~‘m.

SITY LIBRARIES

IIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIII III

3 129300

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the

dissertation entitled

Cross-Functional Sourcing Team Effectiveness:
Critical Factors and Relationships

presented by

Robert J . Trent

has been accepted towards fulfillment
of the requirements for

Doctor of Philosophy degree in Production and Operations
Management

/%47%%««%

Major professor

Robert M. Monczka, Ph.D.
Professor of Strategic Sourcing

Date—W393.— Management

 

 

MSU it an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

 

LIBRARY
Mlchlgan State
Unlvorslty

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date duo.

DATE D E DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU I. An Affirmative Adlai/Equal Opportunity Institution
extrema-no.1

 

CROSS-FUNCTIONAL SOURCING TEAM EFFECTIVENESS:
CRITICAL FACTORS AND RELATIONSHIPS

By

Robert Joseph Trent

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Management

1993

Robert M. Monczka, Advisor

ABSTRACT

CROSS-FUNCTIONAL SOURCING TEAM EFFECTIVENESS:
CRITICAL FACTORS AND RELATIONSHIPS

By

Robert Joseph Trent

Despite an increased interest in the use of cross-functional sourcing teams
within US. industry, 'we actually know very little about what contributes to cross-
functional sourcing team effectiveness or how this type of team might function more
effectively. The primary objective of this dissertation was to address the lack of
research-based knowledge concerning the factors relating directly to cross-functional
sourcing team effectiveness. Accomplishing this required the development and
testing of a series of a prion' research hypotheses that evaluated the variables
potentially related to cross-functional sourcing team effectiveness.

This research collected data from 18 US. based firms over a five-month
period. Each participating company actively used cross-functional sourcing teams for
the purpose of improving firm performance and competitiveness through effective
sourcing and supply base management strategies. A total of 108 cross-functional
sourcing teams provided detailed information about a number or topics. This
research used a cross-sectional survey to evaluate the relationship that team
leadership, organizational resources, performance evaluation and reward systems,
demographic variables, team effort, team effectiveness, communication, cohesiveness,

sUpplier involvement, team authority, and other variables have with cross-functional

ii

iii
sourcing team effectiveness. In addition, the research combined data from three
separate sources. Team members, team leaders, and external team raters provided
various data that formed a team record. The use of separate data sources
introduced greater objectivity and integrity to the data collection process.

Prior to analyzing final survey data, extensive measurement model analysis
occurred. This analysis used a combination of exploratory and confirmatory factor
analysis to develop scales that demonstrated high internal and external consistency
as well as unidimensionality. Final data analysis used hierarchical regression analysis
to identify those variables that consistently related to member-reported and
extemally-reported effectiveness. Post hoc data analysis relied on canonical
regression to identify critical relationships not proposed a priori.

Data analysis revealed that five variables consistently related to cross-
functional sourcing team effectiveness. These variables include the presence of
certain organizational resources, the effort expended by team members towards team
assignments, team leadership, a team's level of empowerment and authority, and
external supplier involvement on team activities. This project also identified future

research opportunities related to cross-functional sourcing team effectiveness.

To Jan, whose patience and support made this possible.

To Jack and Ellen, who always help in their own special way.

iv

ACKNOWLEDGEMENTS

This project would not have been possible without the generous support of
those willing to commit their time and energy towards a successful outcome. First,
I would like to thank and acknowledge the effort of my dissertation committee
members--Robert Monczka, John Hollenbeck, Philip Carter and Robert Handfield.
Each unselfishly shared their insight, knowledge, and positive critiques and
suggestions that helped transform this project into a final outcome. To each I am
very grateful. In particular, I want to recognize the efforts of my committee
chairman, Robert Monczka. His continued support, friendship, and professionalism
throughout will always be remembered and appreciated.

I would also like to acknowledge the support of Shirley Monczka and Greg
Gay. Their support during data collection and reporting was invaluable. I would
also like to express my appreciation to the individuals who took the time to complete
the survey at the participating companies. Their contribution made the successful
completion of this project possible.

Finally, and most importantly, I want to express my appreciation to my wife,
Jan. Her support and commitment to our goal never wavered. Without her, the

completion of this project and the Ph.D. program would have been but a dream.

TABLE OF CONTENTS

List of Tables .............................................. xii
List of Figures ............................................. xviii
CHAPTER I: INTRODUCTION ................................ 1
Research Purpose ........................................... 2
Cross-Functional Sourcing Teams ................................ 4
Increased Interest in the Organizational Team Concept ................ 9
Research Questions ......................................... 12
Research Contribution ....................................... 13
Dissertation Organization ..................................... 15
CHAPTER II: REVIEW AND SUMMARY OF THE LITERATURE ....... 17
Defining and Differentiating Groups/Teams ........................ 17
Defining Team Effectiveness ................................... 22
Key Variables or Factors That Impact Team Effectiveness .............. 25
Team Composition .................................... 26
Organizational Reward Structure .......................... 29
Team Perfomrance Norms ............................... 3O
Organizational Resources ................................ 32
Task Performance Strategies .............................. 33

vi

vii

Knowledge and Skill of Members .......................... 34
Team Leadership ...................................... 36
Team Cohesiveness .................................... 40
Presence of and Commitment to Team Goals ................. 42
Level of Effort ........................................ 43
Team Communication .................................. 44
Other Variables Identified in the Review .......................... 45
Degree of Task Autonomy ............................... 45
Performance Outcome Feedback .......................... 46
Stress Level of Team Members ............................ 46
Team Status ......................................... 47
Task Structure ........................................ 48
Team Longevity ....................................... 51
Information Flow ...................................... 52
Power Within the Team ................................. 52
Technology .......................................... 53
Member Roles ....................................... 54
Decision Making Structure ............................... 55
Potency ............................................. 55
Team Size ........................................... 56
Team Intervention ..................................... 58

Conclusions About the Variables Affecting Team Effectiveness . . . . 59

viii

Literature Related to Cross-Functional Teams ...................... 60
Identifying Current Theoretical Models of Team Effectiveness ........... 66
Nieva, Fleishman, and Reick Model ........................ 66
Gladstein Model ...................................... 67
Cummings Model ..................................... 68
Kolodny and Koggundu Model ............................ 68
Hackman Model of Team Effectiveness ...................... 69
Observations About the Current Team Effectiveness Models ...... 76
Calls for Research .......................................... 77
Literature Review Conclusion .................................. 79

CHAPTER III: MODEL DEFINITION AND

RESEARCH METHODOLOGY ................................. 80
Cross-Functional Sourcing Team Effectiveness Model ................. 80
Variable Definition .......................................... 85
Team Leadership ...................................... 85
Organizational Resources ................................ 86
Organizational Reward System ............................ 87
Team Effectiveness .................................... 87
Team Experience . . . .1 .................................. 87
Team Level of Effort ................................... 88
Team Communication .................................. 88
Team Cohesiveness .................................... 89

Member Attitude ...................................... 89

ix

Research Hypotheses ........................................ 89
Hypothesis 1 ......................................... 9O
Hypothesis 2 ......................................... 93
Hypothesis 3 ......................................... 95
Hypothesis 4 ......................................... 97
Hypothesis 5 ......................................... 100
Hypothesis 6 ......................................... 102

Measurement Model Development .............................. 109

Validity Assessment ......................................... 114

Reliability Measurement ...................................... 116

Sample Size Determination .................................... 117

Research Sample ........................................... 119
Selection Criteria ...................................... 119
Research Process ...................................... 121
Demographic Study Indicators ............................ 124
Team Tasks .......................................... 126

Data Analytic Procedures ..................................... 129

CHAPTER IV: MEASUREMENT MODEL EVALUATION

AND HYPOTHESIS TESTING ................................. 131

Measurement Model Evaluation ................................ 131
Variables Analyzed Using Confirmatory Factor Analysis .......... 132

Variables Analyzed Using Exploratory Factor Analysis ........... 153

X

Hypothesis Testing Results .................................... 173
Hypothesis 1 ......................................... 176
Hypothesis 2 ......................................... 178
Hypothesis 3 ......................................... 180
Hypothesis 4 ......................................... 181
Hypothesis 5 ......................................... 184
Hypothesis 6 ......................................... 186

PostHocDataAnalysis......................... .............. 195

CHAPTER V: RESEARCH DISCUSSION ......................... 205

Discussion of Critical Research Findings .......................... 205
Organizational Resources ................................ 205
Supplier Participation ................................... 224
Team Authority ....................................... 233
Team Effort ......................................... 243
Team Leadership ...................................... 248

Discussion of Demographic Indicators ............................ 259
Team Size ........................................... 260
Functional Diversity .................................... 261
Professional and Cross-Functional Team Experience ............ 262
Geographic Diversity ................................... 263
Team Education ...................................... 264

Within-Team Age Dispersion ............................. 264

xi

Team Longevity ....................................... 265
Discussion of Other Study Variables ............................. 266
Performance Evaluation and Reward ....................... 266
Team Cohesiveness .................................... 275
Team Communication .................................. 277
Research Extensions ......................................... 280
Research Limitations ......................................... 286
Bibliography ............................................... 291
Appendix A
Information Packet Forwarded To Potentially Participating Companies ..... 305
Appendix B
Information Letter Forwarded to Individual Study Participants ........... 312
Appendix C
Follow-Up Letter Sent to Participants With Missing Data .............. 314
Appendix D
Requests for Measuring Instrument .............................. 315
Appendix E

Frequency Reporting of Survey Questions ......................... 316

LIST OF TABLES

Table 1
McGrath's Task Typology ..................................... 50

Table 2
Teams and Cross-Functional Teams Literature Search 1986-1991 ......... 62

Table 3
Cross-Functional Sourcing Team Effectiveness Research Project
Participating Company Data ................................... 120

Table 4
Cross-Functional Sourcing Team Effectiveness Research Project
Demographic Study Statistics ................................... 125

Table 5
Educational Level of Study Respondent ........................... 125

Table 6
Physical Distance Between Team Members ........................ 127

Table 7
Cross-Functional Sourcing Team Effectiveness Research Project
Functional Team Representation ................................ 127

Table 8
Cross-Functional Sourcing Team Tasks ............................ 128

Table 9
Openness/Accuracy of Communication-Confirmatory Factor Analysis ...... 136

Table 10
Team Cohesiveness-Confirmatory Factor Analysis .................... 139

Table 11
Level of Effort-Confirmatory Factor Analysis ....................... 142

xii

xiii

Table 12
Team Member Collectivism-Confirmatory Factor Analysis .............. 144

Table 13
Consideration Behavior-Confirmatory Factor Analysis ................. 148

Table 14
Initiating Task Structure-Confirmatory Factor Analysis ................ 151

Table 15
Team Effectiveness-Exploratory Factor Analysis ..................... 155

Table 16
Performance Evaluation and Rewards-Exploratory Factor Analysis ........ 163

Table 17
Assessment of Leadership Effectiveness-Exploratory Factor Analysis ...... 168

Table 18
Supplier Involvement-Exploratory Factor Analysis .................... 169

Table 19
Team Authority-Exploratory Factor Analysis ........................ 171

Table 20
Hypothesis One Results-Hierarchical Regression Analysis
Team-Reported Level of Effort ................................. 177

Table 21
Hypothesis One Results-Hierarchical Regression Analysis
Externally-Reported Level of Effort .............................. 177

Table 22
Hypothesis Two Results-Hierarchical Regression Analysis
Personal Cohesiveness ........................................ 179

Table 23
Hypothesis Two Results-Hierarchical Regression Analysis
Professional Cohesiveness ..................................... 179

Table 24
Hypothesis Three Results-Hierarchical Regression Analysis
Accuracy of Team Communication ............................... 182

xiv

Table 25
Hypothesis Three Results-Hierarchical Regression Analysis
Openness of Team Communication .............................. 182

Table 26
Hypothesis Four Results-Hierarchical Regression Analysis
Within-Team Frequency of Communication ........................ 183

Table 27
Hypothesis Five Results-Hierarchical Regression Analysis
External-Team Frequency of Communication ....................... 185

Table 28
Hypothesis Six Results-Hierarchical Regression Analysis
Team-Reported Process Performance ............................. 187

Table 29
Hypothesis Six Results-Hierarchical Regression Analysis
Externally-Reported Process Performance .......................... 188

Table 30
Hypothesis Six Results-Hierarchical Regression Analysis
Team-Reported Innovation Performance .......................... 189

Table 31
Hypothesis Six Results-Hierarchical Regression Analysis
Externally-Reported Innovation Performance ....................... 190

Table 32
Hypothesis Six Results-Hierarchical Regression Analysis
Team-Reported Time Reduction Performance ...................... 191

Table 33
Hypothesis Six Results-Hierarchical Regression Analysis
Externally-Reported Time Reduction Performance ................... 192

Table 34
Hypothesis Six Results-Hierarchical Regression Analysis
Team-Reported Supply Base Management Performance ............... 193

Table 35
Hypothesis Six Results-Hierarchical Regression Analysis
Externally-Reported Supply Base Management Performance ............ 194

Table 36
Post Hoc Analysis-Correlations Between Team-Reported
Dependent Variables and Canonical Variables ...................... 200

Table 37
Post Hoc Analysis-Correlations Between Covariates and Canonical
Variables Using Team-Reported Dependent Measures ................ 200

Table 38
Post Hoc Analysis-Correlations Between Externally-Reported
Dependent Variables and Canonical Variables ...................... 201

Table 39
Post Hoc Analysis-Correlations Between Covariates and Canonical
Variables Using Externally-Reported Dependent Measures ............. 201

Table 40
Post Hoc Analysis-Hierarchical Regression Analysis
Team-Reported Process Performance ............................. 202

Table 41
Post Hoc Analysis-Hierarchical Regression Analysis
Team-Reported Supply Base Management Performance ............... 202

Table 42
Post Hoc Analysis-Hierarchical Regression Analysis
Team-Reported Time Reduction Performance ...................... 203

Table 43
Post Hoc Analysis-Hierarchical Regression Analysis
Team-Reported Innovation Performance .......................... 203

Table 44
Post Hoc Analysis-Hierarchical Regression Analysis
Externally-Reported Process Performance .......................... 204

Table 45
Post Hoc Analysis-Hierarchical Regression Analysis
Externally-Reported Supply Base Management Performance ............ 204

Table 46
Externally and Team-Reported Organizational Resource Levels .......... 208

Table 47
Correlations Between Team-Reported Resources and

Team-Reported Effectiveness .................................. 210

Table 48
Correlations Between Externally-Reported Resources and

Externally-Reported Effectiveness ...............................

Table 49
Critical Relationships Between Process Performance Items and

211

Organizational Resource Categories .............................. 214

Table 50
Critical Relationships Between Supply Base Performance Items and

Organizational resource Categories ..............................

Table 51
Correlations Between Team-Reported Effectiveness and

Supplier Involvement Indicators .................................

Table 52
Correlations Between Formal Supplier Team Membership and

Other Key Study Variables ....................................

Table 53
Correlations Between Supplier Support at Team Meetings and

Other Key Study Variables ....................................

Table 54
Correlations Between Team Effectiveness and Team Authority Dimensions

Table 55
Correlations Between Internal Team Process Authority and

Other Study Items ...........................................

Table 56
Correlations Between Effectiveness Dimensions and Cross-Functional

Sourcing Team Leadership ....................................

Table 57
Correlations Between Team Leadership and Cross-Functional

Sourcing Team Effort ........................................

216

225

229

231

. 236

240

251

254

xvii

Table 58
Correlations Between Team Leadership and Coordination of Work Activity . 254

Table 59
Correlations Between Team Leadership and Team
Communication Satisfaction .................................... 255

Table 60
Performance Evaluation and Rewards ............................ 267

Table 61
Responsibility for Evaluating Performance Contribution ............... 269

Table 62
Determination of Performance Evaluation for Team Participation ........ 269

LIST OF FIGURES

Figure 1

Nieva, Fleishman, & Reick's Model of Team Performance .............. 71
Figure 2

Gladstein's Model of Group Effectiveness .......................... 72
Figure 3

Cumming's Model of Team Effectiveness .......................... 73
Figure 4

Goodman's Adaptation of Kolodny and Kiggundu Model ............... 74
Figure 5

Hackman's Model of Group Effectiveness .......................... 75
Figure 6

Cross-Functional Sourcing Team Effectiveness Process Model ........... 81
Figure 7

Hypothesis 1 ............................................... 91
Figure 8

Hypothesis 2 ............................................... 94
Figure 9

Hypothesis 3 ............................................... 96
Figure 10

Hypothesis 4 ............................................... 99
Figure 11

Hypothesis 5 ............................................... 101
Figure 12

Hypothesis 6 ............................................... 104

xviii

I. INTRODUCTION

Management theorist Tom Peters (1988) has written that the organizational
model featuring narrow, functional middle managers operating within a rigid, vertical,
and functional alignment is obsolete. A flern'ble structure that includes suppliers,
customers, and managers from different specialties crossing functional barriers and
moving horizontally throughout the organization is now required. Purchasing and
materials executives have indicated the cross-functional sourcing team is rapidly
becoming an important part of this new flexible organizational structure (Monczka
1989). ' Firms increasingly are relying on cross-functional sourcing teams to reduce
barriers between organizational functions and to improve total performance and
competitiveness through effective sourcing and supply base management strategies
(Monczka 1989, 1992). This includes supporting a firm's performance in the areas
of cost, quality, delivery responsiveness, time to market, and concurrent design and
engineering activities.

Despite a renewed interest in organizational teams and the availability of
previous research cOncerned with team effectiveness, we still lack a well-tested and
accepted body of research and theory to guide managers in the formation and use
of teams (Hackman 1987), including cross-functional sourcing teams. Furthermore,
Loehr (1991) noted that since a number of scholars predict an increase in the use of
cross-functional teams in the coming years (also cross-functional sourcing teams),
theory-building and hypothesis testing must occur to support the development and

appropriate use of this type of team. Too little is known about what contributes to

2

cross-functional team effectiveness or how these teams might come to function more

effectively (Loehr 1991). Even less is known about cross-functional teams in a

procurement and sourcing environment.

RESEARCH PURPOSE

This dissertation addressed the lack of knowledge about the factors that relate

directly to cross-functional sourcing team effectiveness. The purpose of this research

was to develop and test, using actual teams within US. industry, a series of research

hypotheses that evaluated the variables that potentially relate to cross-functional

sourcing team effectiveness. This required the completion of certain activities:

(1)

(2)

(3)

(4)

(5)

Conduct adetailed review of small group/team literature to identify
the factors that potentially impact cross-functional sourcing team
effectiveness

Develop a conceptual model of cross-functional sourcing tea‘m
effectiveness and related research hypotheses based on previous
group/team research and procurement/sourcing research

Collect data from US. firms that actively use the cross-functional
sourcing team approach

Develop and analyze the measurement scales required for testing the
sourcing team effectiveness model and related research hypotheses

Analyze each research hypothesis and discuss relevant research
findings and implications

A comprehensive review of the literature related to organizational work teams

resulted in several conclusions. Each conclusion supports the purpose of this

dissertation.

First, although executive management increasingly looks to

organizational teams, including cross-functional sourcing teams, to improve a firm's

3

competitive position, little is actually known about what determines team
effectiveness (Hackman 1988, Goodman 1986, Tziner and Eden 1985). Furthermore,
the use of formal work teams does not guarantee a successful outcome to an
organizational task. Research must address the factors contributing to team
effectiveness. No previous research was found that evaluated the factors relating
directly to cross-functional sourcing team effectiveness.

Secondly, few studies have explored, empirically, the factors that affect team
effectiveness (Goodman, Ravlin, and Argote 1986, Loehr 1991). A review of the
literature covering 1986-1991 revealed that only a handful of minor studies focused
on some aspect of cross-functional team effectiveness. There is a total absence of
research focusing on cross-functional sourcing teams. While a number of theorists
offer a vision of self-directed specialists who move together, work effectively in
teams, and share action oriented goals, these visions do not provide blueprints for
the successful development and use of a cross-functional team (Loehr 1991).

A third conclusion was that a major portion of previous group and team
research has relied on single laboratory experimental designs. Unfortunately, this
limits our research emphasis to two or three variables while controlling the effects
of extraneous variables through randomized subject assignment. Laboratory
research, while offering insight into the effects of specific team-related variables,
helps explain only selected parts of the team effectiveness phenomena. Previous
laboratory research does supports the development, however, of a conceptual cross-

functional sourcing team effectiveness model. It provides knowledge about the

4

variables that potentially impact cross-functional sourcing team effectiveness and the
relationship between model variables.

McGrath (1986) has argued it is time to take what we have learned about
group/work team dynamics and performance within the laboratory and apply this
knowledge directly to field research. Gladstein (1984) maintained the study of
organizational teams in context is a complex and difficult task. The need for direct
field research, however, is critical as teams become more prevalent in procurement
and sourcing, which can significantly impact the competitiveness of manufacturing
firms. This dissertation sought to address the concerns surrounding the current lack

of scholarly field research within this research area.

CROSS-FUNCTIONAL SOURCING TEAMS

This research focused on a particular type of organizational team--the cross-
functional sourcing team. Before addressing sourcing team specifics, a discussion of
the cross-functional team approach is necessary to further our understanding of this
concept. Because minimal research about cross-functional teams was found to exist,
some of the references made during this section refer to anecdotal literature.

Cross-functional teams, also known as interfunctional teams, consist of
personnel from separate organizational functions brought together to achieve a
specified task(s). Effective use of these teams represents a new organizational
challenge for firms accustomed to maintaining a hierarchical and inflexible

organizational structure.

5

A cross-functional team satisfies Hackman‘s (1982) perception of an
organizational work group or team. This means the team is a real, self-managed
group with an intact social system, has one or more tasks to perform, and operates
within a formal organizational context. Cross-functional teams, however, are unique
compared to most other organizational work teams. First, team membership usually
consists exclusively of mid-level functional specialists. This provides some level of
task knowledge and professional heterogeneity. Secondly, cross-functional teams
often assume responsibility for assignments or projects with a limited duration or
members participate on the team in addition to regular responsibilities. This differs
from an autonomous production team, for example, that operates on a continuous
basis. Thirdly, most cross-functional teams are self-managed work teams. This
means the team has the authority to handle internal processes to generate an
identifiable team output, service, or decision (Hackman and Oldham 1980).

A self-managed cross-functional team differs from a self-regulated work team.
Self-regulated teams, which include self-directed production teams or executive
management teams, are empowered to act as a democratic form of work
organization. This provides total team responsibility for the regulation, organization,
and control of a job or task (Grayson 1991). A self-managed team, while having
significant responsibility for managing its task assignment, usually relies on others
external to the team to make decisions about team goals, structure, and the overall

level of organizational support.

6

Cross-functional teams are becoming increasingly important as an
organizational structure and tool within US. industry (Larson 1988, McKeown 1990,
Simmons 1989, Massimilian and Pedro 1990, Peters 1988). The cross-functional
approach can add a degree of organizational flexibility, multi-functional knowledge,
and control and coordination mechanisms for responses to competitive demands that
traditional hierarchical or matrix structures are unable to achieve (Alavi and Keen
1989). Optimistic predictions of leading industry experts indicates that task—oriented,
cross-functional teams making major business decisions, including major product and
sourcing decisions, will begin to supersede the rigid functional and fixed matrix
organizational structure present in most companies (Morgan 1991).

While cross-functional teams can exist throughout a firm and perform many
separate tasks (Kumar and Gupta 1991), this research focused on cross-functional
sourcing teams directly responsible for purchasing and materials related assignments.
Sourcing teams are increasingly common within manufacturing firms today due to
this team's potential ability to influence the competitive factors of product cost,
quality, product cycle time reduction, and organizational innovation (Monczka 1989).
Today, over half of a typical firm's costs result from externally purchased materials.
The sourcing team represents an effort by management to maximize the performance
potential of the procurement function through effective cross-functional supply base
management.

Previous research revealed that material executives perceive the use of cross-

functional sourcing teams is critical for future competitive success (Monczka 1989).

7

The results of a comprehensive field and Delphi study conducted in 1988 revealed
that executive purchasing and materials managers will emphasize the cross-functional
sourcing team approach for purchasing and materials related tasks, particularly
supplier evaluation and selection and strategic commodity management, throughout
the 19905 (Monczka 1989). Data collected from 54 firms in 1992 supported the
original Delphi prediction. In the 1992 data sample, almost 75% of firms indicated
they used cross-functional sourcing teams (Monczka, Trent, and Gay 1992).
Executive purchasing and materials managers perceived cross-functional sourcing
teams to be critical for future functional and organizational success. Compared to
other purchasing and materials strategies, respondents rated the cross-functional
sourcing team approach as currently one of the highest emphasized strategy
responses at the executive functional procurement level.

Each member of a cross-functional sourcing team, regardless of the team's
specific task, should provide a specialized functional expertise. The sourcing team
can include representatives from purchasing/materials, MIS, research and
development, quality assurance, finance, accounting, manufacturing, or design and
product engineering. Team membership can also include other specialists on an as
needed basis.

Cross-functional sourcing teams are typically responsible for a number of
purchasing and materials related activities. Possible activities or tasks assigned to
sourcing teams include to (1) support material sourcing requirements during new

product development, (2) support material sourcing requirements for items with

8

established production specifications, (3) identify common purchased items between
corporate business units or buying centers for consolidated purchasing, (4) negotiate
corporate-wide material purchase agreements, (5) evaluate and select strategic
suppliers, (6) develop purchased material cost-reduction strategies, (7) develop local,
business unit, or corporate-wide sourcing strategies, and (8) perform value analysis
activities. Sourcing teams can also be responsible for projects with a limited or finite
duration.

A common and critical cross-functional sourcing team task is the evaluation
and selection of strategic suppliers. The team works as a combined unit to evaluate
supplier capability across a number of performance factors. These factor include a
supplier's (1) price/cost competitiveness, (2) quality commitment and capability, (3)
expected delivery performance, (4) financial integrity, (5) management capability, (6)
ability to support early product design, (7) product and process technical competence
and innovativeness, (8) delivery and manufacturing systems, (9) commitment to
supply chain improvement, and (10) business and contingency planning. The team
evaluates not only current supplier capabilities but also the ability to achieve
continuous future improvement.

Another potential cross-functional sourcing team task is the identification of
common company-wide purchased items or commodities. Company-wide purchased
items of interest usually have (1) strategic importance to the competitive success of
a firm, (2) availability limited to relatively few global suppliers, (3) a high degree of

technical content, (4) wide common usage between business units, or (5) a strong

9

impact on total product quality. Once the team identifies likely company-wide
sourcing candidates, it must develop a corporate strategy and the systems to manage
the procurement process.

Specific cross-functional sourcing team task assignments can differ from firm
to firm. The team's primary emphasis or focus is dependent upon a firm's
competitive and operational requirements. All sourcing teams, by definition,
emphasize purchasing and materials related task assignments. This is what
differentiates a cross-functional sourcing team from other cross-functional or

organizational work teams.

INCREASED INTEREST IN THE ORGANIZATIONAL TEAM CONCEPT

The growing interest and use of cross-functional sourcing teams is part of a
larger movement towards the increased use of organizational teams. The work team
is gaining popularity as a basic building block of organizational structure throughout
US. industry (Alavi and Keen 1989). Confronted with intense competition, customer
demands, and the need to develop innovative products and processes in shorter time,
managers must reinvent how they manage the workforce and structure the task
environment (T josvold 1991).

Organizational work teams interest executive management because of the
performance potential the successful use of the team approach offers firms.
Researchers from a variety of disciplines have written extensively about the benefits

resulting from the successful use of the team approach for organizational task

10
assignments (Grayson 1991, McGrath 1984, Likert 1961, Miner 1984). For example,
Tjosvold (1991), citing Hackman (1983), wrote that an effective team provides some
definite benefits to the team and the individual. A sample of the benefits that can
occur when a team meets its performance potential includes:
(1) Team members become committed and identify with their group and
organization. They take a long-term view and have the energy to

pursue future assignments.

(2) Members develop individual skills and effective working relationships
that allow them to address future tasks with confidence.

(3) Members find their work motivating and rewarding so they exert
greater effort, use their abilities, and persist to do their jobs well. .

(4) Groups complete their tasks and accomplish their goals in ways
acceptable to others within the organization.

Maier (1967) noted the potential benefits of the team approach include a
greater sum of total knowledge and information brought together at one time, a
greater number of approaches put forth to solve a problem, joint participation of
problem solving leading to increased group acceptance, and better awareness and
acceptance of team decisions throughout an organization. Hackman and Morris
(1975) have argued the team yields major benefits when "patterns of interaction
develop in which the individual inputs of team members combine to yield an
outcome better than any single person--or even the sum of the individual parts."
This refers to the synergistic effect of team interaction.

As researchers, we cannot believe, however, the mere existence of a team
results in a higher level of organizational effectiveness. The literature cites many

examples of organizational teams failing to reach their performance objectives or

11

creating adverse effects on individual team members (Maier 1967, Likert 1961,
Hackman and Morris 1975, Dyer 1977, Tjosvold 1991). The drawbacks that affect
organizational work teams also can affect cross-functional sourcing teams. While the
cross-functional concept is simple in theory, practical implementation is often
difficult. The cross-functional team approach requires most firms to adopt a new
corporate culture of participative teamwork throughout all levels and across rigidly
defined functional boundaries (Simmons 1989).

Research conducted by Hoffman (1979) concluded that groups by themselves
neither necessarily produce the best decisions their members are capable of nor are
the members necessarily motivated to carry out a group's decision. Altier (1987)
noted that a poorly directed, improperly constituted team can be a disaster. The
team approach can waste valuable time, delay important decisions, reach inferior
decisions, or cause confrontations between participants.

The research literature cites several possible drawbacks to the organizational
(including the cross-functional) team approach. First, a team may not be the most
efficient means to perform a task (Kaplan 1979, Steiner 1972). Second, team
interaction can negatively effect the individual team member (McGrath 1964,
Hackman 1976). For example, team participation may require a member to assume
an uncomfortable role, endorse performance norms that do not match the member's
norms, place an individual on a team the individual does not feel comfortable with,
or a variety of other scenarios that result in individual dissatisfaction (George 1977).

Third, a team can actually arrive at a worse decision or outcome than individual

12
members who work separately (Janis 1981, Levine and Moreland 1990). Finally,

researchers have focused on the possibility that a team's performance norms can
actually be lower than at least some of the performance expectations of the
individual team members. As a result, teams can enforce norms of low rather than
high productivity (Whyte 1955).

This discussion introduced the team approach as a form of organizational
structure gaining increased acceptance within US. industry. We must recognize that
teams can positively or negatively affect an organization and its members. Within
the purchasing and materials arena, it is critical that we gain a greater understanding

of the factors relating directly to cross-functional sourcing team effectiveness.

RESEARCH QUESTIONS

This research addressed questions that are relevant to both academics and
practitioners. Four major questions underlie this project. First, which variables are
thought to relate directly to cross-functional sourcing team effectiveness? The
answer to this question required a literature search to identify the variables thought
to exert the strongest influence on cross-functional sourcing team effectiveness. This
included identification of the domain of variables that may make the difference
between a successful and unsuccessful cross-functional sourcing team.

Second, what were the relationships between the variables thought to relate
to cross-functional sourcing team effectiveness? This question required the

development of a conceptual model of cross-functional sourcing team effectiveness.

13

This model specified direct and indirect variable relationships between independent
and dependent variables. From collected data, we evaluated the hypothesized
relationships between the independent and dependent variables within the model.
Third, which variables identified during the literature search related directly
to cross-functional sourcing team effectiveness or performance? This question
required (1) development and testing of a measurement instrument that assessed the
constructs or variables hypothesized to impact sourcing team effectiveness, (2) data
collection from cross-functional sourcing teams in a field setting, and (3) data
analysis to identify statistically significant relationships between model variables.
Finally, how could the results of this research contribute to the more effective
use of cross-functional sourcing teams? This is a timely and critical topic within US.
industry today. This required the translation of the research results into a set of
findings that increase our knowledge about the deveIOpment and effective use of

cross-functional sourcing teams.

RESEARCH CONTRIBUTION

The primary contribution of this research is to advance the state of knowledge
in an area critical to the practitioner yet largely ignored by the academic. Goodman
et a] (1986) maintained that for such a broad and important topic of research, the
actual number of group or team performance studies conducted is relatively small.

This included research that involves cross-functional and organizational work teams.

14

This research is expected to make an academic and applied contribution.
Academically, it responded to the call for increased team field testing made by
leading group/team researchers (McGrath 1986, Hackman 1988, Goodman, Ravlin,
and Argote 1986, Gladstein 1984). Although this research focused on cross-
functional sourcing teams, many of the variables or constructs assessed within the
cross-functional sourcing team effectiveness model are common to other self-
managed organizational teams. Unfortunately, the constructs defining team
effectiveness and its antecedents have not been operationalized or readily accepted
(Goodman, Ravlin, and Argote 1986). A major contribution of this research
included the development of measures for variables within the cross-functional
sourcing team effectiveness model. In-depth evaluation of the cross-functional
sourcing team effectiveness model provided an opportunity to build upon previous
group/team work and to further our understanding of this topic.

At an applied level, this research provided an opportunity for an in-depth
analysis of cross-functional sourcing teams. The results of this research will help
organizations form and manage sourcing teams with a better understanding of the
variables critical to team effectiveness. This research will also provide a level of rigor
currently not associated with cross-functional team research. Research-based answers

will begin to replace anecdotal accounts of cross-functional team effectiveness.

15
DISSERTATION ORGANIZATION

This dissertation consists of five chapters. The first chapter is the introduction
to the research topic. The remainder of the dissertation focuses on the activities
required to develop, test, and analyze research hypotheses related to a conceptual
model of cross-functional sourcing team effectiveness.

The second chapter reviews and summarizes the literature related directly to
the dissertation topic. The result of the literature review (1) defines and
differentiates groups/teams, (2) defines team effectiveness, (3) identifies the key
variables or factors that impact cross-functional sourcing team effectiveness, (4)
presents and critiques the leading models of team effectiveness, and (5) identifies the
calls for research related to the dissertation topic. The review and summary of the
literature provided the foundation for the development of the cross-functional
sourcing team effectiveness model.

Chapter III discusses the cross-functional sourcing team effectiveness model,
related research hypotheses, and the research methodology used during the research.
Specifically, this chapter includes a discussion of the conceptual model of cross-
functional sourcing team effectiveness along with a discussion of each research
hypothesis. The chapter also addresses (1) the development and testing of construct
measures, including reliability measurement and validity assessment, (2) sample
selection, (3) sample size determination, (4) data collection, and (5) the statistical

approaches used during data analysis.

16

Chapter IV presents key statistics and research results related to measurement
model and research hypothesis testing. This includes the presentation of
correlational matrices, significance test results, variable beta weights, and other
important statistical findings. The chapter also presents final reliability coefficients
for each multi-item construct assessed during the research. Analysis of post hoc data
also occurs.

Chapter V summarizes and interprets the major research findings. These
conclusions have a dual audience--industry managers who look to the cross-functional
sourcing team approach as a means to achieve performance improvement objectives
and others who seek increased knowledge about group or team effectiveness. The
latter group can include academics and industry practitioners interested in
organizational teams other than cross-functional sourcing teams. Finally, the chapter
discusses future research extensions and research limitations identified during the

course of this project.

11. REVIEW AND SUMMARY OF THE LITERATURE

The literature that supports this dissertation topic includes research conducted
within a number of academic disciplines. This is due to two reasons. First, the
theoretical domain of small group or team research lies outside the business areas
of engineering, production, procurement, materials management, or operations. A
review that limited itself to these areas would forgo the benefit of research that
provides the conceptual foundation of team dynamics. Secondly, a review of the
current business-related literature revealed limited research related directly to cross-
functional teams. There currently exists no in-depth study of cross-functional
sourcing team effectiveness.

The research domain of small group or team research is broad. Literally
thousands of studies implicate many variables of team performance (Foushee 1984).
These studies cover a wide range of organizational settings, groups, and tasks. This
dissertation literature review is not a review of the broad domain of small group
research. For reviews of the small group literature, see McGrath and Altman (1966),
Hare (1982), McGrath and Kravitz (1982), or Levine and Moreland (1990). For a
review of intergroup relations, consult Messick and Maclde (1989). Finally, consult
Clark and Reiss (1988) for a review of two-person small group relations.

This review and summary supported directly the research requirements for the
development of a cross-functional sourcing team effectiveness model. The specific
objectives of this literature review included to:

0 Discuss and define key research concepts, including different
definitional perspectives of work teams and team effectiveness.

17

18

0 Identify and define the variables that potentially impact team
effectiveness or performance.

0 Evaluate the literature and previous research related directly to cross-
functional teams (including sourcing teams).

0 Present and critique current theoretical models of work group or team
effectiveness.
0 Identify research opportunities and calls for research related directly

to the team effectiveness research topic.
Throughout the literature review, we will assume that teams in an
organizational business setting are also groups. For this discussion only, work team

and work group are synonymous.

DEFINING AND DIFFERENTIATING GROUPS/TEAMS

Different perspectives of groups and teams exist within the literature. One
problem is that researchers use the label group to refer to a wide variety of social
and organizational forms (Hackman 1990). Because of this, it is necessary to
highlight some of the different perspectives or definitions of groups/teams. The
objective is to arrive at a workable understanding or conceptualization of an
organizational or cross-functional work team.

The basic definition of a group provides a good starting point. A group is two
or more people existing in an arrangement that permits some degree of interaction
among members that share a sense of identity (Rubin 1977, Zander 1980, Shaw
1981). Mink, Mink, and Owen (1987) defined a group as three or more persons

working toward a common goal or purpose. Tjosvold (1991) elaborated when he

19

argues that groups are two or more persons who interact and influence each other
directly, who are mutually dependent and have interlocking roles and common
norms, and who see themselves as a unity in pursuit of common goals that satisfy
their individual aspirations and needs. All of these definitions, while providing a
conceptual perSpective of groups and teams, are too broad to develop a definition
in an organizational or operational sense.

Brett and Rognes (1986) defined organizational groups. They maintained an
organizational group is a set of individuals who constitute an identifiable social
aggregate within a formal organization. They pointed out that a legitimate group or
team requires both members and nonmembers to recognize it as a distinguishable
social aggregate that exists within the boundaries of an organization. This
perspective provides a valuable differentiation between formal and informal
organizational teams.

Organizational attributes are also used to define groups or teams (Hackman
1990). Defining a group or team by its specific attributes allows a researcher to
identify formal organizational groups versus informal groups outside the research
domain. Hackman (1990) defined organizational groups according to three unique
attributes. First, the group must be real. This means it has boundaries,
differentiated member roles, and interdependence among members. Next, the group
must have one or more tasks to perform that results in some collective outcome.
Third, the group must operate in an organizational context and manage relations

with other individuals in the larger system in which it operates. Teams or groups

20

that satisfy these three criteria are actual organizational work groups (Hackman
1987).

Recent attention has focused on self-regulated work groups or teams, self-
managed work groups or teams, and business teams in an organizational setting.
While similar in some respects, the three concepts are technically different. Self-
regulated work teams have almost total control over the internal regulation of their
activities. They are responsible for individual member task assignments, a multiple-
skill based task, and an identifiable or whole piece of work (Pearce and Ravlin 1987).
Grayson (1991) maintained self-regulated work teams are a work related approach
to self-determination. This type of team is a democratic form of work organization
ascribing to a team the responsibility for the regulation, organization, and control of
a job or task.

Self-managed work teams differ slightly from self-regulated work teams.
Hackman and Oldham (1980) referred to a self-managed work team as an intact
social system whose members have the authority to handle internal processes as they
see fit to generate a specific team product, service, or decision. The difference
between self-regulated and self-managed teams lies in the team's degree of control,
autonomy, or authority. A self-regulated work team has total control over its
activities and assignments. Self-regulated work teams are also self-managed work
teams. The opposite, however, is not true. A self-managed work team does not have
the same degree of control or autonomy over its activities as a self-regulated work

team.

21

Another perspective encompasses both self-regulated and self-managed work
groups or teams. Alavi and Keen (1989) noted that a business team is a self-
contained, self-directed task oriented work group usually comprised of ten or fewer
members. Business teams typically focus on assigned tasks distinct from informal
groups within a business. This perspective covers most of the teams or groups
commonly associated with business organizations today--including cross-functional
sourcing teams.

In a study of teams in a business setting, Hackman (1990) concluded there
exists a continuum of organizational teams. One end of the continuum includes
manager-led teams. These teams have virtually no authority--they merely execute the
work that other people design, monitor, and manage. Self-governing (i.e. self-
regulating) teams are at the other extreme of the continuum. These teams have
virtually unlimited authority to choose their own goals, select members and team
leadership, and manage work of their own design. Between the two extremes, but
closer to the self-regulated work teams, are self-managed teams. While these teams
have significant responsibility for managing their work, others usually make decisions
about team goals, structure, and the overall level of organizational support.

In conclusion, the most useful approaches for evaluating a work team for the
purposes of this research (1) define organizational groups and business teams (Brett
and Rogness 1987, Alavi and Keen 1989), (2) differentiate between self-regulated
versus self-managed work teams (Grayson 1991, Pearce and Ravlin 1987, Hackman

and Oldham 1980), (3) describe a continuum of work teams (Hackman 1990), and

22

(4) describe the unique attributes of a real organizational work team (Hackman
1990). By analyzing these different perspectives found in the literature, one can
conceptualize and identify organizational and cross-functional sourcing teams in a
field setting. It also becomes possible to identify the similarities and differences
between cross-functional and other organizational teams. This affects the
researcher's ability to generalize externally certain research findings. Understanding
the different types of teams within an organization also allows easier evaluation of

previous research.

DEFINING TEAM EFFECTIVENESS

The need to define the team effectiveness variable is essential for cross-
functional sourcing team research. The literature review indicated a measure of
team effectiveness is usually the dependent variable in team research studies. Miner
(1984) has argued that comparisons across team research studies reveal that the
variable producing the most inconsistent findings is the dependent variable that
measures team effectiveness or performance. This is due partly to the many ways
researchers have operationalized this construct. No accepted or consistent measure
of effective team performance exists in the literature. Defining team effectiveness
as a performance measure is an issue all team research must address.

Hackman (1985) presented one of the more thorough and potentially
operational approaches to team effectiveness (see also Hackman 1987, Hackman and

Walton 1986, and Hackman 1990). He argued that simple outcome measures of

23

effectiveness, while central to laboratory research, are not comprehensive enough for
field research within complex organizations. Most organizations do not have clear
right or wrong answers about a team's overall performance. In fact, many
organizations formally evaluate individual team member performance rather than
providing an overall measure of team performance.

Hackman (1985) defined team effectiveness in terms of three performance
criteria. The first criteria is that the productive output of the team should meet or
exceed the performance standards of those who receive and/or review the output.
Team performance usually depends more on what external others (i.e. key managers,
clients etc.) perceive about team performance rather than a single performance
measure of output. The second performance criteria is that the team‘s experience
should maintain or enhance the capability of members to work together on future
team assignments. A team that can no longer work together because of poor team
interaction was , not an effective work unit. The third criteria is that the team
experience should, on balance, satisfy rather than frustrate the personal needs of
team members. Hackman clearly provided a multi-dimensional perspective of team
effectiveness.

Rubin (1984) presented a second perspective that focuses on a number of
team effectiveness quality dimensions. He broadly defined effective team
performance in terms of a team's ability to arrive at a quality decision(s). The
components of decision quality include (1) team efficiency, (2) careful development

and analysis of alternatives, (3) fairness to internal team members and affected

24

others, (4) member satisfaction and morale, (5) leadership effectiveness, and (6)
team growth over time. He concluded that team effectiveness, as indicated by the
quality of group decisions or outcomes, must be evaluated according to a complex
set of multiple criteria.

Another approach to team effectiveness emphasizes specific measures of
performance output as indicators of team success (McGrath 1964, Steiner 1972, and
Guzzo 1986). This approach does not define performance dimensions per se but
rather relies on specific output measures (i.e. measures of efficiency). McGrath
(1964) referred to team performance outcomes such as "speed to solution" or
"number of errors." Steiner (1972) focused strictly on measures of productivity to
evaluate team effectiveness. Guzzo (1986) defined effectiveness in terms of a team's
ability to fulfill its charter as reflected in the team's goals and objectives. Guzzo's
definitional perspective emphasizes effectiveness over efficiency.

Another approach defines effective work teams through broadly stated
operational characteristics. McGregor (1960) detailed eleven characteristics of an
effective work team while Likert (1961) described twenty-four characteristics. Mower
and Wilemon ( 1989) have listed ten characteristics that must be present for effective
performance. While these approaches provide a conceptual perspective to effective
team performance, it is difficult to develop a performance measure that captures
each characteristic discussed by these researchers.

Of the different approaches to defining team effectiveness, Goodman (1986)

has concluded that Hackman's dimensional approach is the most useful, particularly

25

in a complex organizational setting. Hackman's approach explains what effectiveness
actually means. A critical issue with the dimensional approach, however, is that a
theoretical or operational model must explain the different dimensions. Overall,
Goodman concluded that team effectiveness per se is not a clear concept. It
becomes necessary, therefore, to apply an approach that defines what effectiveness
means rather than relying on a simple measure of team effectiveness. For this
research, a multi-dimensional perspective of team effectiveness offered the greatest

opportunity for operationalizing the team effectiveness construct.

KEY VARIABLES POTENTIALLY IMPACTING EFFECTIVENESS

A major requirement of this dissertation was to identify and evaluate the
variables that potentially affect cross-functional sourcing team effectiveness. Perhaps
the most critical decision during the construction of a theoretical model is the
selection of model variables (Hildebrand and Ott 1991). While one might suggest
a strategy of developing a model that included many variables, such a model quickly
becomes uninterpretable and increases the opportunity that a significant research
finding occurs by chance.

To be unaware of the potential variables that affect cross-functional sourcing
team effectiveness risks reaching inconclusive or even incorrect research conclusions.
James (1980) has noted that researchers have reported many biased estimates of path
coefficients in the literature, for example, because relevant variables were not

included in the causal model. He wrote "the operative question is not whether one

26

has an omitted variables problem but rather the degree to which the unavoidable
unmeasured variables problem biases estimates of path coefficients or provides a
basis for alternative explanations." James further argued that all models are subject
to some degree to an omitted or unmeasured variable problem.

A primary objective of the literature review was to identify the critical
variables that potentially impacted cross-functional sourcing team effectiveness. The
most important variables were operationalized, measured, and included in the cross-
functional sourcing team effectiveness model. The research also assessed additional
variables that might have had an effect on team performance but were not included
in the original model.

The following section identifies and discusses the factors that potentially affect
team effectiveness as identified during the literature review. This discussion has three
objectives: ( 1) identify the variables or factors that potentially affect cross-functional
sourcing team effectiveness, (2) summarize significant previous research about the
variable, and (3) develop an awareness about potential relationships between the

variables. Discussion of the variables occurs in no particular order.

Team Composition
Extensive research has focused on team composition and its affect on
performance. Even without an abundance of previous research indicating the

importance of team composition to team success, experience with teams across a

27

range of settings indicates the importance of team member composition towards
team effectiveness.

Previous team composition research has usually centered on homogeneity or
heterogeneity of traits, attitudes, or abilities of team members (Goodman, Ravlin,
Argote 1986). Even with the significant amount of research that has occurred,
Tziner and Eden (1985) have argued that little is known about how composition
affects team performance in tasks requiring a high level of interdependence--a
characteristic of the cross-functional sourcing team.

Steiner (1972) argued that a team's task influences the impact of member
composition. Most research supports the finding that heterogeneity of member
characteristics contributes positively to team performance when task requirements
are diverse (Goodman, Ravlin, and Argote 1986, Pearce and Ravlin 1987). Overall,
team performance is likely to be better if a team has heterogeneous skills, or at least
a sufficient range and distribution of skills to address the required tasks (McGrath
1964). The greater the variety of perspectives the more likely a high quality solution
will emerge (Hoffman 1979, Pearce and Ravlin 1987). Teams with different
(although not extreme) personality types, leadership abilities, types of training, and
points of view usually are more creative and innovative than teams with similar
member characteristics, other things equal (Hoffman 1979). Hoffman also argued
that heterogeneous teams are no less cohesive than homogeneous teams, although

heterogeneous teams do sometimes experience communication strain.

28

Previous research has also focused strictly on the effect of individual member
personality as a component of team composition (Shaw 1981). Hoffman (1979) has
cited extensive research that focuses on various personality types and the impact on
team performance. He argued that certain personality characteristics will influence
the rate of participation and the relative influence of members. Team members with
certain personality types can be destructive to the team process. For example, the
extroverted, dominant, socially aggressive individual consistently emerges in team
research as the highest participator with disproportionate influence over the team's
decision. This individual's superior social skills and ability to dominate others
prevents the team from considering opposing or alternative ideas. Unfortunately, no
correlation exists between the dominant member and superior cognitive ability.
Other personality types shown to have a negative influence on the team process are
the self-confident members, individuals with strong affiliative needs, and
counterdependent (i.e. resistant) types.

Since organizations have little control over the personality characteristics of
team members, it is important to be aware of the effects that certain personality
types can have on team performance (Hoffman 1979). Shaw (1981) cited evidence
that teams whose members are different within a reasonable range with respect to
personality profiles perform more effectively than teams whose member's personality
profiles are similar. Potential problems occur when member personalities are no

longer in what is considered the "reasonable range."

29

Other research has focused on the personal and physical traits of team
members. Most personal and physical trait research examines the effect of a team
member's sex (Levine and Moreland 1990). Shaw (1981) reported that personal
characteristics of team members influence the team interaction process by
determining to some extent an individual's behavior and how others respond to that
member. Shaw noted that physical characteristics such as weight, height, race, sex,
and appearance influence the behavior of the individual team member and the
responses of the other members to him or her. He also concluded that "relative to
other variables, these factors are generally weak and can be overcome by the effects
of powerful variables such as personality and ability." The physical characteristics of
members, while possibly impacting team performance, are attributes to measure and

control rather than include as a central focus of team research.

Organizational Reward Structure

Few variables that potentially affect team performance or effectiveness have
resulted in such consistent research findings across different settings as an
organization's reward structure. Guzzo (1986) noted that it is hardly surprising that
rewards impact team effectiveness. He also noted that many of the major theoretical
models of team or group effectiveness reflected the importance of team rewards
directly in the models.

Researchers have noted that the destructive effects of rewarding individual

contributions rather than team performance can be considerable (Lawler 1981,

3O
Hackman and Oldham 1980). McGrath (1964) has argued that teams perform better

when management rewards the entire team rather than differential rewards to each
member. In direct field study, Abramis (1990) noted that firms providing rewards
and recognition only to individual employees found that tension developed as teams
matured. Team members realized that although they surmounted major work
challenges as a team, rewards were provided only to certain individuals.

A reward system that recognizes and reinforces team performance can
complement and amplify the motivational incentives built into the team task
(Hackman and Oldham 1980). Pearce and Ravlin (1987) have posited that for self-
managed and self-regulated work teams, a mix of individual and team level rewards
will be most effective in establishing commitment to individual task responsibilities
as well as the overall goals of the team. Individual evaluation ensures personal
commitment and effort to the team's task while team rewards encourage team effort

and performance.

Team Performance Norms

Team performance norms can positively or negatively impact effectiveness.
Golembiewski (1965) defined team norms as reflections of social pressures that apply
either to all team members or to performers of specific roles. McGrath (1964)
referred to norms as a generic term for team attitudes. He maintained they are the
shared predispositions of an interrelated set of people to respond to classes of

objects in certain positive or negative ways.

31

Team norms influence effectiveness because they exert pressure for individual
members to conform to those norms (McGrath 1964). Teams that maintain a low
performance norm can experience lower team effectiveness. Also, teams with lower
performance norms may pressure members with high individual performance norms
to accept a lower norm. Of course, the opposite situation can also occur. Low
performers may find themselves on a team that enforces a higher performance norm.
Work teams that have higher performance norms or attitudes usually produce at a
higher level (Likert 1961). Dyer (1977) has written that high-performing
organizations must emphasize certain norms of behavior before the formation of any
team because norms are critical to team success.

Swap (1984) has argued that team norms are a potential source of conflict--
particularly when team norms conflict with individual norms. Swap also maintained
that norms often interact with team cohesiveness. Cohesive teams seem best able
to enforce team performance norms. Whether a cohesive team is productive or
nonproductive depends largely on the norms it enforces. The study of team norms
is critical due to the potential relationship between team norms and the individual
and team level of effort. In other words, performance norms directly impact a team's

level of effort.

Organizational Resources
This variable refers to the presence and quality of the organizational resources

provided to accomplish a team's task assignment. Peters and O'Connor (1980)

32

argued that attention to the mundane aspects of a team's performance situation (i.e.
material resources) is increasingly recognized as critical to effective team
performance. Peters and O'Connor have identified the variables that correspond to
team resources as situational resource variables. They maintain eight unique
situational resource variables are vital to team effectiveness: (1) job-related
information and training, (2) tools and equipment, (3) materials and supplies, (4)
budgetary support, (5) required services and help from others, (6) task preparation,
(7) time availability, and (8) the physical work environment. Hackman and Oldham
(1980) have argued that the availability of task-relevant training and technical
consultation are critical resources for creating conditions favorable to team
effectiveness.

Guzzo (1986) noted that apart from the resources brought by the individual
members, teams in an organizational environment rich in resources are more likely
to be effective, other thingsequal, than teams that do not have adequate resources.
One resource that Guzzo discusses is the availability of computerized decision
support systems in organizations. Although computers hold great potential for
enhancing the effectiveness of decision-making teams, few demonstrations of their
potential impact on team decisions and effectiveness ern'sts.

The presence or absence of adequate team resources has the potential to
make a good team ineffective and a less qualified team at least marginally effective.
The availability of organizational resources is potentially a primary determinant of

team effectiveness (Guzzo 1986).

33

Task Performance Strategies

The use of performance strategies is a widely researched and potentially
important determinant of team effectiveness (Hackman and Morris 1975, Hackman
1987, Cummings 1981, March and Simon 1958, Hackman and Oldham 1980).
Performance strategy refers to the collective choices made by team members about
how they will perform an assigned task (Hackman and Morris 1975).

Teams that actively assess the opportunities and demands facing the team and
then consider several alternative methods for proceeding with their task usually
develop appropriate performance strategies compared to teams that do not actively
assess task demands and opportunities (Maier 1963, Hackman 1987). Furthermore,
once members agree to a strategy, whether implicitly or explicitly, they usually
continue using that strategy throughout the team's assignment (March and Simon
1958). Hackman and Morris (1975) indicated that task-oriented teams rarely
consider performance strategies explicitly, particularly for tasks that the members
have some familiarity. Members rely on implicit understanding and past experience
concerning how to address the assigned task.

Researchers have concluded that the strategies members use in working on
their task can affect a team's performance effectiveness (Maier 1963, Hackman,
Brousseau, and Weiss 1976). Shiflett (1972) suggested that because many different
types of tasks and strategies exist, perhaps the only universally effective strategy is
one in which the team shows a willingness to switch from one specific strategy to

another as the need arises.

34

Task performance strategy is a major variable in several of the leading
theoretical models of team effectiveness. Guzzo (1986) noted that both Steiner's
(1972) and Hackman's (1982) model of effectiveness requires that for a team to be
effective, it must adopt performance strategies that are consistent with the
performance requirements of the task. Hackman (1987) has stated that a team's
performance strategy is one of three variables responsible for a large portion of team
effectiveness. Most researcher agree that (1) the presence of a performance strategy
can affect team effectiveness and (2) it is important that the strategy match the
requirements of the performance task. A mismatch between strategy and task can
result in suboptimal team performance.

Research has also focused on methods to improve performance through team
strategy. Hackman and Morris (1975) proposed that three approaches exist to
encourage a team to discuss and possibly change its performance strategies as
needed. The approaches include (1) diagnosis and feedback of current strategies, (2)
process consultation or intervention, and (3) task redesign so the task matches the
performance strategy. The actual strategy formulation process, along with the

modification of task strategy over time, is not a focus of this research.

Knowledge and Skill of Members
Researchers have identified the overall knowledge and skill of team members

as having a disproportionate effect on team effectiveness (Hackman 1987, Hackman

and Morris 1975, Tziner and Eden 1985, Cummings 1981). Hackman (1987) noted

35

that a team is most likely to bring enough talent and expertise to a task when (1) the
team has an appropriate number of members with a good mix of skills, (2) the
education system of the organization offers training or consultation as needed to
support existing knowledge, and (3) team interaction avoids weighing of individual
contribution and instead supports sharing of expertise and collective learning. This
variable actually reflects how much knowledge and skill members apply to the task
and not how much knowledge and skill each member possesses. This is a critical
distinction because a measure of general intellect or education may not accurately
reflect team knowledge and skill.

Tziner and Eden (1985) studied the effects of individual ability on the
performance of tank crews. They found that crews composed of high-ability
members performed significantly better than predicted on an inter-dependent tasks.
Crews with all low-ability members performed worse than predicted. The researchers
noted a clear interactive effect between ability and performance (Levine and
Moreland 1990).

McGrath (1964) has maintained that not all tasks require members with the
highest possible knowledge and skills available. He argued that teams need a
distribution of levels and types of abilities that fit the requirements of the assigned
task. A review and summation of the literature by Shaw (1981) revealed the
importance of the knowledge and skill of team members towards achieving a team's
assigned task. Shaw noted that an individual with specific abilities related to the

team's task will be more active on the team, will make a greater contribution to the

36

team's attempts to complete the task, and will have increased influence on the team's
decisions. The high ability member is more likely to emerge as a leader and improve
team performance. Individual member knowledge and skill is a critical variable when
developing a testable model of cross-functional sourcing team effectiveness.

This research employed a multi-dimensional variable of team member
experience in place of team member knowledge and skill. The dimensions of team
experience included (1) functional team diversity, (2) prOfessionaI and functional
tenure of each team member, (3) total cross-functional team experience, and (4) age
and age dispersion within each team. The collection of the data to support these
dimensions is more feasible than assessing each member's knowledge and skill
applied to the task. While the assessment of knowledge and skill is attractive
theoretically, demographic and experience measures are more practical. This

approach also eliminates reliance on a self-reported measure of knowledge and skill.

Team Leadership

The role and impact of the team leader has been a widely researched team
effectiveness variable (Green 1975, Elmes and Wilemon 1988, Maier 1967).
Golembiewski (1965) has defined leadership as a measure of general status with at
least three major dimensions: individual prominence, aiding team attainment, and
sociability. On a cross-functional project team, a project manager is often the team
leader. An outstanding project manager is one who has a reputation for completing

projects on or near schedule, within budget, within specifications, and whose team

37

members would actively choose to work with again if they had the opportunity (Pitts
1990).

A leader can positively or negatively influence a team's drive, its cohesion, its
goal selection, and its goal attainment (Stogdill 1981). Stogdill also found that teams
with high degrees of consensus about their leadership are more effective and satisfied
than those lacking consensus. Maier's (1967) research revealed that it makes sense
to have a leadership role when there is a significant amount of information for the
team to process. Research also indicates the importance of the leader in linking the
team to the rest of the organization (Likert 1961). McGrath (1964) viewed
leadership from the perspective of leadership behavior. As such, a leader exhibits
three behavioral roles: the leader as an influence on individual members, the leader
as the initiator of the team structure required to solve a task or problem, and the
leader as the completer of essential team functions.

Shaw (1981) focused considerable attention on the subject of team leadership.
From his research review, Shaw developed a series of propositions or hypotheses

about team leadership:

0 In teams without a formally assigned leader, the probability that a'
leader emerges within a team increases with increasing size of the
team

0 Leaders are usually more intelligent than nonleaders in situations

involving emergent versus selected leaders

0 The more dependable the team member, the more probable it is that
he or she will emerge as a leader and will be successful in helping the
team achieve its goal

38

° Persons who actively participate in the team are more likely to attain
a position of leadership

° Emergent leaders tend to behave in a more authoritarian manner than
elected or appointed leaders

0 Effective leaders are characterized by task-related abilities, sociability,
and motivation to be a leader

0 The degree to which team members endorse the leader depends upon
the success of the team in achieving its goals

Goodman, Ravlin, and Argote (1986) have concluded that a review of the
current literature indicates that (1) situational factors are important in determining
the effectiveness of a leader and (2) participative leaders are often more effective
than authoritarian leaders. They also concluded that task and team characteristics
play an important role in determining just what effect leadership has on the team.
Guzzo (1986) concluded that the most critical functions of a leader of a task-
performing team are those activities that contribute to the establishment and
maintenance of favorable performance conditions.

Instead of simply evaluating the team's perception of leadership, a more
interesting approach cited is to evaluate leadership behavior and its relationship to
other model variables. Stogdill has written that the study of leadership is complex.
His analysis of leadership research indicated (1) the selection of leaders because of
personality traits contributed minimally to group success, (2) numerous traits
differentiated leaders from followers, (3) the traits demanded in a leader varied from
one situation to another, and (4) the trait approach ignored the interaction between

the leader and the group.

39

Because focusing on leader personality traits has proved inconclusive, many
research efforts have focused on leadership behavior. This research relied on an
external measure of leader behavior--the Leader Behavior Description Questionnaire
(LBDQ)--to assess the leadership variable. This approach will provide greater insight
concerning leadership behavior and its affect on other variables in the model.

The LBDQ has been tested extensively and consistently supports two
independent leader behaviors--consideration and initiating structure (Stogdill and
Coons 1957). Although the original LBDQ was thought to measure nine separate
behaviors, factor loadings consistently indicate that two predominant leader behaviors
emerged from data collected across multiple settings and time periods. Although the
two dimensions of behavior have been found to correlate to a moderate degree, they
are sufficiently independent to permit the use of the consideration and initiating
structure scales as measures of different kinds of behavior (Halpin and Winer 1957).
The use of the LBDQ provided an external measure of team leadership because
team members evaluated the team leader directly. The team leader does not
evaluate him or herself.

Items related to the consideration dimension are associated with behavior that
indicates friendship, trust, respect, and warmth between the leader and team
members (Halpin and Winer 1957). Consideration is a people-oriented dimension
and indicates the degree to which the team leader is considerate of team members.
Items related to the initiating structure dimension are those indicating whether the

team leader organizes and defines the relationship between him or herself and the

40

team members. An initiating structure leader tends to define the expected role of
each team member and attempts to establish well-defined patterns of organization,
channels of communication, and methods of achieving the team's task (Halpin and
Winer 1957). Initiating structure behavior is task-oriented. A leader can exhibit a

high level of both behavioral dimensions.

Team Cohesiveness

Shaw (1981) noted that most researchers refer to cohesiveness as the degree
to which team participation motivates members to remain on a team. The term also
refers to the level of a member's motivation or attraction to a team. Different
conceptual and operational definitions have resulted in some confusion about what
cohesiveness actually is and how to measure this construct (Levine and Moreland
(1990). As a result, research results can vary considerably across studies due to the
multiple approaches taken to measure and evaluate the cohesiveness construct.

The effects of cohesiveness on a team are complex. They depend on factors
such as (1) the abilities of the team members, (2) the type of task the team is
performing, (3) the leader's style, and (4) which aspect of team cohesion the
researcher assesses (Levine and Moreland 1990).

Most researchers have agreed that a highly cohesive team is a "close-knit
group." Cohesive teams are able to enforce greater participation and loyalty among
members and usually exert greater influence over them, including conformity to team

standards (Swap 1984). Cohesiveness can have a positive or negative effect on team

41

performance given the right circumstances and task. Positive feelings among
members promotes the security to risk suggesting unusual solutions to problems
(Hoffman 1979). Other research has found that the specificity of a team's goals and
the accuracy of performance feedback increases team cohesiveness, which positively
links to performance in self-regulating work teams (Pearce and Ravlin 1987). More
importantly, Pearce and Ravlin (1987) have noted that recent research indicates that
teams performing tasks high in interdependence (such as a cross-functional sourcing
team) show a clear, positive relationship between task-based cohesiveness and team
effort and effectiveness.

Shaw (1981) has discussed a number of hypotheses with tentative support in
the literature: (1) members of high-cohesive teams communicate with each other
more than members of low-cohesive teams, (2) high-cohesive teams exert greater
influence over their members than low-cohesive teams, (3) high-cohesive teams are
more effective than low-cohesive teams in achieving team goals, (4) members of
high-cohesive, compatible teams are usually more satisfied than members of low-
cohesive teams, and (5) the pattern and content of interaction are more positively
oriented in high-cohesive teams. Cohesiveness is potentially a major determinant of

cross-functional sourcing team effectiveness.

42

Presence of and Commitment to Team Goals

Commitment to a team's task or goals is potentially a key variable affecting
team effectiveness. Several major dimensions of this variable include: (1) the
presence of team goals and (2) the commitment to team goals.

Zander (1971, 1980) focused on whether teams with explicit goals produce
more than teams with no goals. He found that teams with goals often use the goal
as a criterion for evaluating how well the team is performing. Additionally, Zander's
research indicated that (1) teams set difficult goals more often than easy goals, (2)
teams with goals perform better than those asked simply to perform their best
without explicit end goals, (3) a team whose members commit to its goals usually
have members who prefer a goal with favorable consequences, and (4) external
pressures on a team to choose a specific goal usually result in the setting of a
difficult rather than an easy goal.

A second dimension involves a team's commitment to its goals. Shaw (1981)
has defined a team goal as an end state desired by a majority of team members.
Shaw also maintained that a team can have more than one goal. He argued that it
is unlikely that a team would attempt a task if it had no goals or objectives. The
reason for attempting a task is that task completion facilitates goal achievement.

The importance of goal setting, commitment, and achievement appears critical
to team success. Furthermore, the team's interest and perceived progression towards
a defined and agreed upon goal can influence member motivation (Pearson and

Gunz 1981).

43

The presence of and commitment to a team's goals has the potential to impact
the effectiveness and performance of a cross-functional sourcing team. Herman and
Herman (1989) have maintained that cross-functional teams should have clear goals
or objectives agreed to by the members and reconfirmed each time the team meets.
These goals should be quantified whenever possible. For further discussion of the
importance of team goals, see Likert (1961), Shaw (1981), and Levine and Moreland

(1990).

Level of Effort

Hackman and Morris (1975) and Hackman (1987) proposed that the level of
effort team members commit or expend to meet their goals and objectives is a
primary determinant of team success. Hackman hypothesized that a team will work
especially hard to achieve its tasks and goals if certain conditions are met:

0 Members must use a variety of high-level skills on the task

0 The task is a whole and meaningful piece of work with a visible
outcome and challenging performance objectives

0 Team outcome has significant consequences for others within and
outside of the organization

0 The team receives positive consequences for excellent performance
and any rewards focus on the team effort

° The task provides members with substantial autonomy

0 Work on the task generates regular, trustworthy feedback about the
team's performance

44

The individual and team level of effort is a primary determinant of team
effectiveness. Hackman's research has indicated that the level of effort expended on

the team's task assignment directly influences team effectiveness.

Team Communication

Team communication reflects both the quantity and quality of information
exchange between team members. Katz (1982), for example, maintained that
research consistently shows that interpersonal communication, rather than formal
reports, publications or other written documents, are the primary means used by
professional teams to collect and transfer important new ideas and information
throughout a project. Foushee (1984) has noted that a significant process variable
is the information flow within a team. A study conducted by Seers (1989) found that
the quality of information exchange between individual team members systematically
related to team performance. -

Swap (1984) noted that an important determinant of a team's decision-making
effectiveness was the formal and informal communication structure of the team. For
example, for more complex, discretionary tasks (like those found with self-managed
work teams in a business setting), a more decentralized communication structure
should produce both better decisions and greater member satisfaction.

A conceptual model of cross-functional sourcing team effectiveness must
include this variable. A primary objective of the cross-functional approach is the

removal of the traditional barriers of communication between functional groups. A

45

significant portion of a team's success may depend on its ability to coordinate and

communicate information across functional boundaries.

OTHER VARIABLES IDENTIFIED DURING THE LITERATURE REVIEW

A number of other variables identified during the literature search are
hypothesized to exert less of an influence on team effectiveness. However, the
research design must measure and/or control several of these variables to isolate

their possible relation to team performance.

Degree of Task Autonomy

This variable refers to the degree of decision authority or task autonomy
provided to a team. Guzzo (1986) has noted that several theoretical perspectives of
group effectiveness emphasize task autonomy as an essential part of performance.
Hackman (1987) regarded autonomy as important because it stimulates effort within
the team, promotes the use of member skills and abilities, and influences the
adoption of an appropriate task performance strategy.

Task autonomy research is not widespread within the literature. As a result,
this variable has received minor attention as a potential determinant of team

effectiveness in a business setting.

46

Performance Outcome Feedback

A variable emphasized in individual decision-making research but receiving
minor attention at the group or team level is decision or performance outcome
feedback (Tindale 1989). Several researchers have investigated the role of outcome
feedback on different aspects of team performance. Zander (1971) found that goal
directed effort is greater in a team that receives feedback about its progress in
attaining its goals or objectives. Furthermore, Zander has argued that the quality of
a team's productivity increases as the feedback becomes more complete. It is
probable that complete performance feedback affects positively team members with
a stronger need or desire for achievement.

Continuing research by Ilgen and Hollenbeck (1989) focuses on a hierarchical
team decision making process in environments where performance feedback is
immediate and known. This is a likely variable to measure for a team performing

a task that receives or requires feedback at various points in time.

Stress Level of Team Members

The research literature search revealed minimal findings about how varying
degrees of stress impact the effectiveness of team decision making or performance.
This variable is currently part of a major research project conducted by Ilgen and
Hollenbeck (1989). Catastrophes such as the downing of an Iranian civilian jetliner

by the US. military indicate that stress can be an important variable affecting the

47

quality of team interaction and decision making. This variable requires further

research to isolate specifically the effects of stress on team members.

Team-Status

Teams that maintain a structure in which the contribution of individual
members are weighted separately face the real possibility of diminished team
effectiveness (Thomas and Fink 1961). Status differences between members of a
work team can have functional and dysfunctional consequences, while the overall
status of the team relates positively to team effectiveness (Pearce and Ravlin 1987).

An individual member's status on a team often forms naturally as a result of
a member‘s knowledge or position. Levine and Moreland (1990) cited extensive
research that people with a higher status among team members (1) have more
opportunities to exert social influence, (2) try to influence other members more
often, (3) are more influential than people with lower status, (4) receive positive
evaluations more often than members with a lower status, and (5) usually have a
higher self esteem than members with a lower status.

The status of team members has the potential to affect positively or negatively
team performance. If a team member has a higher status due to legitimate or
desirable factors, then we would expect a positive influence on the team as a whole
(and vice versa). Research evidence suggests that an organization should enhance
the status of a team. This should not come, however, at the expense of increased

status differences within the team (Pearce and Ravlin 1987).

48

Task Structure

Hackman (1969) has defined a task as consisting of material resources and a
set of instructions specifying an assignment to perform vis a vis the available
resources. The instructions indicate what operations to perform with respect to the
material resources and/or the team's expected goal. Goodman (1987) viewed task
as "a program or set of operating rules, heuristics, and criteria for a transformation
process. Tasks describe the activities in a particular job or the activities requiring
completion between jobs." Finally, Steiner (1972) maintained a task is not a specific
pattern of behavior but rather an assignment to accomplish in one or more ways.

Potentially negative consequences exist from not controlling or measuring the
task variable in team research. As a result, a number of typologies have emerged
that differentiate team tasks. Steiner's (1972) typology, for example, subdivides
unitary tasks according to the manner in which member contribution results in a
team outcome. Unitary tasks can be disjunctive, conjunctive, additive, or
discretionary (Steiner 1972). While it is beyond the scope of this review to detail
Steiner's typology, this typology is valuable because researchers can use it to analyze
directly team productivity in a mathematical form (McGrath 1984).

A more basic typology is put forth by Hackman (1968). He focuses on three
types of tasks--production, discussion, and problem-solving tasks. Production tasks
require the production of ideas, images, or a physical product. Discussion tasks

require a discussion of values or issues, usually with a requirement of team

49

consensus. Problem—solving tasks require a solution arrived at under constraining
conditions.

McGrath (1984) developed a widely recognized and cited task typology. He
took the best features and concepts from previous typologies and included the
features in a model of group or team tasks. The foundation of McGrath's task
model is four task performance processes (McGrath 1984). The four performance
processes include generate, choose, negotiate, and execute. Two types of tasks are
present within each process. Table 1 details McGrath's task typology. The cross-
functional sourcing teams of research interest are those assigned planning or creative
type tasks.

The type of a team's task impacts its interaction and performance outcome.
The influence of task, however, is probably not direct. It appears that task impacts
performance by interacting with other variables such as team size, team interaction,
performance strategy, or individual skills and knowledge. The nature of a team's task
is a particularly strong influence on the team's interaction process (Hackman and
Morris 1975). Guzzo (1986) has noted that depending on the task, certain
performance strategies are likely to be more effective than others. He also says that
research findings suggest that factors such as the importance of the task and the
extent to which a team engages in a whole task affects team member motivation.

Motivation, in turn, influences individual member effort.

:1er of Task

Planning Tasks
Creativity Tasks

Intellective Tasks
Decision-Making Tasks

Cognitive Conflict
Tasks
Mixed-Motive Tasks

Contests/Battles
Performance Tasks

Source: McGrath (1984)

50

Table l
McGrath's Task Typology

ms;

Generate

Choose

Negotiate

Execute

Key Features

Action-oriented plans
Creative ideas

Single correct answer
Single preferred answer

Resolving policy conflicts
Resolving pay-off conflicts

Winning
Excelling

51

Hackman (1968) has concluded that team task makes a difference in the study
of organizational work teams. He argued that tasks affect the nature of team output
and the characteristics of team interaction. In Hackman's words, "Unless the type
of task is held constant, counterbalanced, or sampled throughout the research, a real
possibility exists that the results of a study may be seriously confounded with
unintended task effects." Gladstein (1984) has argued that task directly moderates
team performance or effectiveness. In a conceptual model of cross-functional
sourcing team effectiveness, the team's task warrants the attention of the

researcher.

Team Longevity

Few studies have evaluated team performance over an extended period of
time. One of the major problems in the study of small groups and teams has been
the general neglect of a temporal perspective of team performance over time (Katz
1982). Recognizing the lack of a temporal approach to team effectiveness research,
Ilgen and Hollenbeck (1989) focus on team performance over an extended period
of time as part of their current research.

Katz (1982), in a comprehensive study of technical project team performance
over time, found that the early positive effects of team formation taper off when
members continue to work together over an extended period of time. Katz also
reviewed several studies that reached the same general conclusion. Performance

gains reach a point of diminishing return, indicating the possible need for rotation

52

of members over the life cycle of a team. Ironically, a later study by Allen, Katz,
Grady, and Slavin (1988) led to the conclusion that no evidence existed for the
previously reported curlinear relationship between performance and tenure. Team
longevity is potentially an important variable for study using a longitudinal research

study approach.

Information Flow

This variable has also received minimal research attention. As such, it is not
entirely known how the quality or quantity of available information effects team
process and performance. Foushee (1984) noted that the way a team uses its
informational resources and the procedures it employs for communicating essential
information are important characteristics of the team process. The exchange of
quality information helps in the development or coordination of strategies through
assurance that members have access to all relevant information. Research

opportunities exist to analyze the effect of information flow on team effectiveness.

Power Within the Team

Related to the status of members and leadership is the distribution of power
within a team. Power distribution within a team can affect the interaction process
and, depending on the basis for that power, can positively or negatively affect team

performance.

53

Fiorelli (1988) studied individual power and its effect on team performance
in a professional work setting. He concluded that expert power was the only power
base having a positive impact on team productivity. In his study, the exercise of
reward and coercive power had a negative effect on productivity and individual
participation. He also concluded that the opportunity for meaningful participation
was far more important than the exercise of the traditional "carrot and stick" (i.e.
reward and coercive power) approach to team motivation. In fact, even the use of
reward power in a professional setting provided an unprofessional basis for making

task decisions.

Technology

Technology represents a system of components directly involved in acting on
and/or changing an object from one state to another (Goodman 1986). Goodman
further elaborated by saying that four components exist in a technological system--
equipment, materials, physical environment, and programs. While the technological
system a team operates under potentially plays a powerful role in team performance,
most of the current models of team effectiveness ignore the role of technology.

Cummings (1981) has presented one of the few models that includes
technology as a key variable of individual and team effectiveness. He defines
technology in terms of its relevant characteristics and argues that technology exerts
its influence on performance outcome through a variety of different summary

variables. Technology, while largely ignored throughout the literature in favor of

54

studying the social processes of teams, nonetheless has the potential to impact overall

team performance.

Member Roles

Role refers to the set of expectations that team members share concerning the
behavior of a person who occupies a given team position and how certain positions
relate to the team's task (Hare 1982). Levine and Moreland (1990) have defined
roles as shared expectations about how a particular person on a team ought to
behave.

The potential for dysfunctional team performance exist if confusion
concerning the role of individual members is present. Levine and Moreland (1990)
concluded that member role potentially influences team effectiveness when (1) a
person lacks the knowledge, ability, or motivation to play a role effectively (such as
a leadership role) or (2) team members disagree about how to carry out a role or
who should play it. Research further indicates that role conflict contributes to
increased tension and decreased individual and team productivity. Furthermore,
transition from one role to another is often a source of team conflict. This variable
can exert a dysfunctional influence on team performance when members do not
understand or accept their roles on a team. As a variable, it is potentially more

relevant when analyzing directly team process and interaction.

Decision Making Structure

This variable refers to the interactive processes that teams use to arrive at a
decision. Swap (1981) has argued that a major variable affecting team performance
is the decision rules and processes within a team (i.e. the decision-making structure).
Examples of different decision making structures include a centralized versus
decentralized decision structure or a consensus versus a majority decision criteria.
Research has indicated that a team's task also affects the appropriate form of
decision making structure (Hackman 1968). Within the literature, the decision
making structure represents a process while actual team effectiveness is an outcome

of that process.

Potency

Potency refers to the collective belief of members that a team is effective. A
team's level of potency is a function of three factors (Guzzo 1986). Guzzo (1986)
noted that these factors include (1) the perception the members have of the team's
level of skill and knowledge, (2) the support provided the team by leaders and others
within an organization, and (3) the availability of resources needed to complete the
task. Relatively few references within the literature cite this variable. Guzzo,
however, included potency as a major variable in his model of team effectiveness.
Potency actually appears to be a variable composed of multiple dimensions, each of

which is a unique determinant of team effectiveness.

Team Size

Researchers often include team size in studies as an afterthought simply
because it is an obvious and available team characteristic (O'Dell 1968). As a result,
a comprehensive review of this single variable would require dozens of pages. Team
size research usually focuses on two dimensions: (1) the relative benefit of team
performance versus individual performance (i.e. do teams perform better than
individuals?) and (2) the processes and outcomes of teams of various sizes (Ziller
1957).

Some agreement exists within the literature about the effects of size on team
performance. Goodman (1986) has noted that an inverted U-shaped relationship
exists between size and performance, depending to some degree on the diversity of
the task and its coordination requirements. As coordination becomes increasingly
difficult through the addition of new members, performance declines with size.
Hackman and Oldham (1980) noted the danger associated with overstaffing a work
team. They have argued that adding additional members can decrease team
productivity. The literature provides some boundary between small and large
organizational work teams. Smaller teams consist of two to seven members while
larger teams consist of twelve to twenty members.

Steiner's (1972) research is perhaps the most widely cited research concerning
the relationship between size and team productivity. Steiner has argued that teams
never perform at their level of potential productivity due to "process losses." Process

loss results from difficulties associated with coordinating member activities,

57

motivational problems, and inefficiencies that result when members work together
on a team. Process loss increases at an increasing rate as a team adds members.
Steiner created a useful set of guidelines that relates size to performance and process
loss.

One relatively strong research conclusion within the literature is that members
of larger teams, compared to members of smaller teams, often report less personal
satisfaction from team participation, feel they have less opportunity to influence team
decisions, and complain of poor coordination of team activities and assignments
(Wicker, Kirmeyer, Hanson, and Alexander 1976). Additionally, Stogdill (1981)
wrote that the size of the team affects the emergence of a leader. He notes that as
size of the team increases, individual members have less opportunity to lead and
fewer members initiate leadership acts.

Shaw (1981) developed plausible research propositions about team size from

an extensive review of the literature:

- Members report greater dissatisfaction with larger teams than smaller
teams.
0 Team performance increases with increasing size when the task is

additive or disjunctive. Team performance decreases with increasing
size when the task is conjunctive (See Steiner's typology of tasks for
task definition).

0 Team members have fewer opportunities to contribute and participate
on larger teams. This leads to greater dissatisfaction on the part of the
individual member.

0 Pressure on the individual participant to conform with a team's
majority position increases with team size, up to some maximum point.

58
Refer to Hare (1982), Thomas and Fink (1963), Hackman (1990), Belbin

(1981), Shaw (1981), and Swap (1984) for additional discussions of team size.
Steiner (1972) is the best single source that addresses the effects of size on team

effectiveness and performance.

Team Intervention

Team intervention, also known as process consultation, is a method of
diagnosing and acting upon the human processes of work teams (Kaplan 1979). It
is a mechanism in which team members, usually with the assistance of an expert
third-party, work together to solve problems and improve team performance (Schein
1969). The objectives of team intervention are to remove the barriers to effective
team functioning and to develop an ability to manage team processes more
effectively (Woodman and Sherwood 1980). Blake and Mouten (1975) provide a
solid discussion of the specific role of the interventionist.

Previous research concerning process or team interventions has indicated
mixed results. Green (1975) found that interventions that structure team interaction
to minimize the opportunities for process losses (such as developing rules and
boundaries) improve team effectiveness for certain kinds of teams and tasks. On the
other hand, Friedlander and Brown (1974) found that interventions usually succeed
in changing member attitudes and sometimes member behavior but have no

consistent effect on performance from one team to another.

59

Hackman (1987) concluded that the research findings regarding process
interventions suggest that structured techniques that minimize process loss can be
successful. Conversely, interventions that seek to improve the quality of
interpersonal relations among members to promote "team synergy" do not appear to
yield reliable improvements in team effectiveness. Team intervention is a critical
variable when the research focus includes a temporal perspective of the team

interaction process, including assessing the effect of an intervention.

Conclusions About the Variables Affecting Team Effectiveness
The review and summary of the literature resulted in several conclusions
about the variables that impact team effectiveness:

(1) Certain variables potentially have a greater influence on a work team's
performance as compared to other variables.

Given the proper circumstances, each of the variables discussed can influence
team effectiveness. Difficulty arises in that a researcher can not evaluate or include
every conceivable determinant of team effectiveness. Hackman and Morris (1975)
have noted that the number of factors that can affect team performance is so great
that managing more than a few factors at a time in an experiment, for example,
becomes prohibitive. Fortunately, the research literature does reveal a number of
variables that likely exert a stronger influence on team effectiveness.

Hackman and Morris (1975) and Hackman (1987), for example, argued that
three variables account for a major portion of the variation in measured team

performance: (1) the level of effort brought to bear on the task by team members,

:‘j

60

(2) the task performance strategies used by the team, and (3) the knowledge and
skills that team members have relative to the task. Overall, selected variables
identified during the literature review Should explain a significant amount of
variation of team performance or effectiveness.

(2) Certain variables, while perhaps not major antecedents of team
effectiveness within this research, nevertheless require measurement
and/or control. For example, team size represents a variable outside
the sourcing team effectiveness model that requires measurement.

(3) A measure of team effectiveness is the dependent variable in almost
all team research studies. Furthermore, it will be necessary to test
multiple models if the team effectiveness variable is determined to be
multi-dimensional.

(4) Certain variables mediate the effect of other variables on team
effectiveness. Not all variables exert a direct influence on team
effectiveness. Some will exert an indirect influence through mediating
variables.

These conclusions are the result of a comprehensive, multi-disciplinary review

of group and work team literature. The conclusions from this review, combined with
the potential research opportunities identified in the literature, provided the

foundation for the development of a testable model of cross-functional sourcing team

effectiveness.

LITERATURE RELATED TO CROSS-FUNCTIONAL TEAMS

A data base search covering the period 1986-1991 revealed hundreds of
journal articles including "teams" or "cross-functional" teams as part of their focus.
These articles included both rigorous academic journals along with trade and industry

publications. A summarized content review of each article reveals a wide variety of

61

areas related to the topic of teams. Table 2 presents a distribution of topics under
the "team" and "cross-functional" team domain. As indicated in this table, relatively
few articles relate directly to cross-functional teams.

A limited amount of academic research currently exists that focuses on cross-
functional teams. A review of the last five years indicated that only a handful of
studies have examined any aspect of cross-functional interaction and performance.
Pinto and Pinto (1991), for example, focused on the antecedents of cross-functional
cooperation. Their research examined the factors under a project manager or
leader's control that can foster cross-functional cooperation among team members.
They examined the effects of four factors-~superordinate goals, physical proximity,
team member accessibility, and rules and procedures. They found that all four
factors were significant predictors of intra-team cooperation and eventually cross-
functional success.

Pinto and Pinto (1990) also conducted research that focused on project team
communication and cross-functional cooperation during the development of new
programs at hospitals. This research revealed that a high cooperation team differed
from a low cooperation team in terms of (1) its increased use of informal methods
for communication and (2) its reasons for communicating. High cooperation teams
exhibited a greater likelihood to engage in brainstorming, obtaining project-related
information, reviewing the team's progress, and seeking performance feedback. Pinto
and Pinto also maintain that a strong relationship existed between communication

patterns and ultimate project success.

62
Table 2

Teams and Cross-Functional Teams

Literature Search
1986-1991

Category:

Not related to the research
topic
General business teams
Self-directed production
teams
Quality improvement teams
Cross-functional teams, including
new product development teams

Research related (6)
cross-functional
articles

Executive level management
teams

Training teams

Accounting auditing teams

Interfirm teams

General employee participation
teams

Quality of work life teams

Information systems teams

Cost reduction teams

MRP implementation teams

Process improvement teams

Total Sources Reviewed:

Articles:

240

80
41

38
36

MMQI—i

NWUJ-B-fi-

522

63

Murray (1989) examined the connection between team composition and firm
performance for executives of Fortune 500 firms. He found partial support for the
hypothesis that heterogeneous groups facilitated adaptation and therefore were
preferable under conditions of environmental change. He also concluded that his
results highlighted the many difficulties associated with untangling and identifying the
causes of a firm's performance.

Loehr (1991) observed directly communication on a cross-functional team in
a manufacturing firm. She found three areas of major concern surrounded successful
cross-functional team design and interaction. First, the interaction process requires
equity among team members. Each team member must have an equal opportunity
to participate and contribute. Secondly, team members must trust one another. A
lack of trust supports increasingly inequitable levels of participation and reduces the
effectiveness of the team process. Finally, equal levels of authority must exist within
the team. When different organizational levels of authority exist within the cross-
functional team, lower authority members display a reluctance to express varying
perceptions of the task and to raise questions about their roles in bringing the project
to completion. Loehr concluded from her case study that crossing functional
boundaries enhanced team effort while crossing levels of authority and position
hampered team effort.

In a survey of high—technology team leaders and members, Meyers and
Wilemon (1989) studied team experiences with learning in a technologically advanced

new product development team environment. Over a third of the respondents

64

surveyed said that "failure to establish clear project goals and objectives" was the
leading error committed by new product development teams. This included poor
goal definition at project kick-off, inadequate definition of goals and functional
departmental roles, and underestimating project complexity. Also, the researchers
reported that informal networks, cited by almost half the respondents, were the
primary method learning was transferred from project to project.

Barczak and Wilemon (1989) analyzed leadership differences and roles in new
product development teams. They identified four major roles played by leaders of
new product development teams. Their research also indicated that leaders of
operational teams rely on a narrow range of familiar techniques to accomplish team
goals. Leaders of teams assigned tasks requiring creativity and innovation rely on a
variety of familiar and unfamiliar steps to meet team objectives. Barczak and
Wilemon's research provided insight into the importance of innovative leadership for
cross-functional teams pursuing certain types of tasks.

Another study, conducted by Thamhain and Wilemon (1987), focused on the
success drivers and performance barriers for engineering related project teams.
These authors identified certain variables that had the strongest positive association
with project team performance: (1) interesting and stimulating work, (2)
organizational recognition, (3) experienced and qualified project team personnel, (4)
proper technical direction and leadership, and (5) professional growth potential. The

authors also noted that (1) unclear project objectives, (2) insufficient resources, (3)

65

conflict, (4) uninvolved management, and (4) shifting goals and priorities were the
predominant barriers to project team performance.

There exists a limited amount of research related directly to cross-functional
teams. Most of the current literature reports strictly anecdotal evidence (Welter
1990, McKeown 1990, Musselwhite 1990, Raia 1991, Kumar and Gupta 1991,
Massimilian and Pedro 1990). Most reporting surrounds the development of new
products, with the coordination of R&D and manufacturing receiving a lesser
amount of attention (Crittenden 1991). While many firms have realized successful
new product introductions resulting from the cross-functional approach, few firms
ever report team failures (Morris 1991).

Team failure provides an excellent learning opportunity, both in a practical
and academic sense. Furthermore, the self-reported successes in the trade journals
may not be reliable and are subject to exaggeration. For example, the trade
literature often cites a major US. automobile producer for its use of the team
approach for the successful introduction of a car line during the mid-19803 (Welter
1990). While the new cars met initial market timing goals, early models failed to
achieve the superior product quality envisioned during team product development.
Over 80 percent of the early buyers reported some kind of problem while different
consumer groups pressured the US. government for product recalls (Simmons 1989).
This indicates that while the team approach can lead to market timing success, the

self-reported success stories in the literature are often incomplete. The industry and

66

trade literature ignores team failures with likely exaggeration of many of the
successes.

The review of the cross-functional literature resulted in three conclusions.
First, current cross-functional research is often limited to a narrow research question
and lacks methodological rigor. Furthermore, no research focuses on cross-
functional sourcing teams. Secondly, the cross-functional research that is available
reveals the importance of team interaction and cooperation, leadership, and
communication to cross-functional team effectiveness. Finally, the majority of
literature related to cross-functional teams is anecdotal or stated in generally
descriptive terms. This highlights the need for additional research in the area of

cross-functional teams.

CURRENT THEORETICAL MODELS OF TEAM EFFECTIVENESS

This section presents an overview of leading theoretical models of team
effectiveness or performance. This supports a later discussion concerning the
research opportunities related to full model testing. Figures 1-5 present the basic
framework or a summarized version of various models of group or team

effectiveness.

Nieva, F leishman, and Reick Model
This model, presented in Figure 1, maintains that team performance is a

function of four variables. It defines team performance as a combination of

67

individual task performance and team performance functions. The four variables in
this model include (1) the external conditions under which the team operates, (2)
individual member resources and ability, (3) team characteristics such as size,
communication, and authority, and (4) task characteristics and demands. While
providing a basic taxonomy of group performance, the model fails to define the
independent and dependent variables at a level that allows empirical verification.
The constructs in this model are actually a composite of several factors of team

effectiveness. Model variables are multi-dimensional.

Gladstein Model

This model, an extension of McGrath's (1964) input-process-output model, is
the only current model of team effectiveness subjected to empirical testing. This
model lends itself to empirical testing because of its relatively straightforward
relationships. However, as with the Nieva, Fleishman, and Reick model, some of the
variables in this model (i.e. group structure) may actually be multi-dimensional
variables in a causal sense.

The research results from the testing of Gladstein's model were inconclusive.
In particular, there was no significant relationship between the external measure of
team performance and the model constructs. Also, investigation of the moderating
effect of the team's task proved inconclusive. There was insufficient task variability
in Gladstein's (1984) study to identify the moderating effect of task on team

performance. Furthermore, the actual study presented no evidence of construct

68

measurement reliability or validity. This further weakens the research conclusions.

Figure 2 presents the Gladstein Model of Group Effectiveness.

Cummings Model

Socio-technical theory provides the foundation for the Cummings model. This
theory maintains that team effectiveness is a function of the joint optimization of
task and social ends in the workplace. Effective groups are those that fulfill both the
task requirements of an organization and the social needs and goals of team
members (Cummings 1978). The major drawback of this model is its complexity.
The model presented in Figure 3 is a somewhat condensed version of the actual
number of dimensions or sub-variables in the original Cumming's model. With the
number of variables included in this model, it would be difficult to identify which
variables, if any, have a significant effect on team performance. Furthermore,
Cumming's relies extensively on two-way relationships between variables. From a
modeling perspective, this makes it difficult to evaluate empirical relationships.

Almost all the variables in this model are a cause and an effect.

Kolodny and Kiggundu Model

The Kolodny and Kiggundu model is another socio-technical model. Six
variables define this model: (1) organizational arrangements-the basic organizational
decisions that constrain the work group, (2) task conditions--the physical environment

of the work, (3) group characteristics--various attributes of each member, (4) group

69

interaction-~pattern of work activity among members, (5) technical skills, and (6)
leadership and supervision—-the formal and informal influences within the group.

Figure 4 is the simplified Kolodny and Kiggundu model as presented in
Goodman, Ravlin, and Argote (1986). This figure reveals that leadership affects
organizational arrangement. Task conditions, organizational arrangements, and team
characteristics affect task skills, team interaction, and leadership. The latter three
variables combine to determine performance outcome. This model also contains a
performance feedback loop.

The positive features of this model include its selection of variables as well as
a refined presentation of the relationship between the variables. A negative feature
is this model is patterned after a logging team. The model applies to a very specific

type of task that may not be generalizable to different task settings.

Hackman Model of Team Effectiveness

Hackman's (1987) model of team effectiveness, also referred to as the
Normative Model, is currently one of the more widely referenced models of team
effectiveness and performance within the literature (see Figure 5). This is due to
Hackman's refinement of the model over of a period of years. The criteria that
define team effectiveness is a major strength of the Hackman model. While other
models usually define effectiveness in relation to some measure of output, Hackman

presents a multi-dimensional definition (refer to the section in this chapter that

7O

discusses team effectiveness). The model is similar to Gladstein's in that both models
acknowledge the importance of organizational and unique group level variables.
The Hackman model has certain attributes that makes it an attractive process
model of team performance. Besides providing a multi-dimensional perspective of
team effectiveness, the model contains a realistic mix of variables that potentially
influence team effectiveness. The model also specifies the general relationships
among the variables in a straightforward manner. A testable model of team
effectiveness could incorporate many of the variables and concepts presented in the

Hackman model.

Observations About the Current Team Effectiveness Models

Goodman, Ravlin, and Argote (1986), in their critique of the leading team
effectiveness models, concluded there exists a viable set of theoretical models of
group performance and effectiveness. On the positive side, the models provide a
useful way to organize our thinking about how groups or teams operate in
organizations. They also concluded that the models describe a general concept of
group effectiveness generalizable across different types of groups or teams.
Furthermore, the models specify the relationships between the variables in terms of
direction rather than in terms of weight (except for the Cumming's model). This
supports empirical modelling and testing of the relationships between the variables

(although minimal testing has actually occurred).

71

 

 

mesons»; H88
085508

 

............................................................................................................
................................................................

 

ASSESS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Oegmaea wadsegeznm
30:59. . .
I4 wags... 5952a saw
. Handgun
H88
goonommmom
Homw I H33
Ogononmmom Volcanoes».
r 85 UoBEam
Eng—w u

 

 

 

7:2». Emma—Boo. an Nana—ma 32E o». H85 Halon—Boone

72

use 5..» wmoanm OCH». Qua.
D85 8

 

 

  

986 008888

 

$888 WEE

$0838an 3»

588302: H380 GHQ—aw Haw w
IHOG 9850 .............................................................................................. , ......................... . ....... . ................ ....H

............
............................................................................
............................................ n. ............................

 

 

 

 

 

 

Hawk 008885.

 

 

-880888— 9.8385.
-Eaavganooo

 

03% @8888

 

 

 

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

-wos. Ba can. 93a. mmooméoomm
-335. 2% 28... . . - .. ._ 2 : _ V . ._
-Hmmw 0033— ....... . ........................................................ . ........................ . ..................................... . ...................... m aI.
mono»— 8§Ev $3.». on
-mommmoomoo
-9888. 0588.8
.8888 Fain—E: Eve.»
H88 ”80:88 -858 2253.8.
Hgom 8m Han—88
088—888
-3888 madam
.635 Saw a w Boon-£2 $5820

 

 

 

 

988N808— mndogc

 

~ .............................................................................................................

batman m2. 035 H888

is... St Ema N

Quamnmqm 7.8%— o». 0.86 mfiaoaénmmm

 

magma

.v

o o ..

.0.

.u.

..

.c.

c.

in:

u.”

I ................................................. . ................ . ........................ . ...........
,.IL .VLP . .LPP .‘4hbu ?.th.rrb.h vr-.~| -- .II~I.~ r... . n..n >:.P.h.->n.-

- Nxvonsngm
- 3352. 330538

 

 

 

 

 

 

 

 

 

 

 

 

FHQHEBODB

583308 _

 

 

 

 

 

 

 

 

 

- moan—=3. . .
- 3.3ch # Hun—S95—

J

 

 

 

Hon—5205
9388138

c OH Gag 00
- 32:9»an 05888

 

 

 

 

 

 

 

 

 

 

 

 

 

mafia u

nus—banana :2?— OH H35 Hmonmguomm

74

 

 

annmaw BE
mcvaammos

 

 

. 1.. .4 <.4.< . .a . .144 4.4.. 1:...

. .. ... .2.3... . .... ... ... ....................................... ....,................. .... ...
..... . ...3.... .. .. 2.... .. .3............................................. ...................3...... .
........... .........7.......n............... ..

 

 

 

GREECE:
ggmaBgflm

 

 

 

...............................................................
.....................................................................................................

 

 

awn—n
Oouamonm

 

.........u..p........‘..--.44’........--...... ...J-1. 1..
............................................................
............................................................
............................................................
............................................................

..........................................................
............................................................

 

 

 

Hmmw mem 03.6
EnaBomoB
Smog—V Ea

 

 

 

 

   
 

 

OEooBam

 

 

 

 

Eng A
moan—Emaw >9683c= cm Mme—23v. Ea Emma—E: 32E

 

 

 

035

 

 

 

 

 

 

 

   

0600550..»— 0008...

 

 

 

 

   
    

 

 

 

 

 

 

 

 

 

 

> 805.. EB «.6688 8683. Sr
tank in.
I ”at: ”E o 0
. mace—n01 Q08... 589nm 9.52.5.
.5333 A_o.m..§§§ ................
.b
03.6 $59.3.

—. macaw Una—ma 9.0.6 583030.. 8.

........................................................................................................................................ 659.8 68098 .088

-988 3.5830 6388 up...»

 

>w$§936~08§u£9w3905rg
53.8.8 39....er

-noBvoncoa 03.5 «3.6
.gaggu

Snag m
flan—3.00.0 30%— 0m 9.0.6 “5.8350000

76

While subject to varying degrees interpretation, the variables in each of these
models are basically similar. Each model includes antecedent and mediating
variables. Each model also relies on a measure of team effectiveness or performance
as the primary outcome variable. In conclusion, the commonalities between the
models exceed the differences. Goodman et al have concluded, however, that these
models require improvement in a number of areas to promote empirical testing. The
following section, Calls For Research, identifies specific research opportunities that

result from the absence of clearly specified team effectiveness models.

CALLS FOR RESEARCH

There currently exists an abundance of "how to" literature concerning team
formation directed toward practitioners and consultants (Bertcher and Maple 1977,
Dyer 1977, Mink, Mink, and Owen 1987). Leiva and Obermayer (1989), Barner
(1989), and Bradford (1989) all offer their insight about forming and managing a
cross-functional team. Many general team building and team development guides
also exist (Blake and Mouten 1975, Beer 1976, Stumpf, Zand, and Freedman 1979,
Shaw 1981, Tolle 1988, Mergerison and McCann 1989, Parker 1990). While
providing some helpful guides and potentially worthwhile assessment and evaluation
tools, they provide little theoretical understanding or insight into the factors affecting
cross-functional sourcing team effectiveness. In fact, some of these sources assume

that a team's composition, particularly in a business setting, is an uncontrollable

77

factor due to personnel limitations or political factors. As a result, these sources
focus on maximizing team performance given team composition and leadership.

From an academic perspective, the area of cross-functional sourcing team
research offers excellent research opportunities. Hackman ( 1987) has argued that
no robust set of generalizations exist that allows us to predict, from prior assessments
of team effectiveness variables, how well a team should perform. He also asserted
that despite the thousands of research studies of group and team performance, we
still know very little about why some teams are more effective than others, including
cross-functional sourcing teams. Designing a team to make an important judgmental
decision or carry out an organizational assignment is a difficult task. Despite the
importance of team design, minimal systematic guidance is available to decision-
makers in an organizational work setting (Stumpf, Zand, and Freedman 1979).

Goodman (1986) has stated that the absence of clearly specified models
affects the quality of data analytic procedures. While most theoretical models
acknowledge the presence of multiple variables that affect group performance, few
models bring to bear the appropriate analytic techniques required to bring out the
relationship among those variables (Tjosvold 1991).

While the current theoretical models present a conceptual mapping of
variables, they are not in a form that allows us to identify the critical variables or to
understand the relationships among the variables. Goodman, Ravlin, and Argote
(1986) stated that research opportunities exist to improve these models by (1)

creating a better understanding of the critical variables that affect performance, (2)

78

creating a better conceptualization of these variables, (3) creating a better
understanding about the functional relationships among variables, and (4) stating
models in a way that supports testing. This will allow researchers to identify
nonobvious yet interesting relationships between model variables.

Researchers (McGrath 1986, Hackman 1987) have also argued that it is time
for greater field research and movement away from focused laboratory study. While
they recognize the contribution of laboratory research, they also acknowledge that
the time has come to develop a fuller understanding of the factors influencing team
effectiveness. Goodman, Ravlin, and Argote (1986) have maintained "when groups
in existing organizations are studied, a broader range of interacting variables must
be accounted for to understand the results. Complex relationships between variables
exist in the real world, and the only way to get at these complexities is to study
systems of variables, rather than two or three variables at a time."

Academic leaders in group and team research clearly make the call for field
testing of team effectiveness models. Even greater opportunities exist for research
that focuses on cross-functional sourcing teams. These teams will play an
increasingly important role in supporting or enhancing a firm's competitive position
(Monczka, Trent,and Gay 11992). The absence of research to support our
understanding of cross-functional sourcing teams limits our ability to maximize the
organizational contribution of this type of team. Clearly, developing and testing a
model of cross-functional sourcing team effectiveness provides a major opportunity

to make a valuable scholarly and applied contribution.

79
LITERATURE REVIEW CONCLUSION

This review and summary of the literature identified the key concepts,
definitions, and potential factors influencing cross-functional sourcing team
effectiveness. This process helped establish a foundation from which to develop a
conceptual model of cross-functional sourcing team effectiveness. While the study
of group and team effectiveness has been popular for many years, only now are we
recognizing the performance potential that teams, including cross-functional sourcing
teams, provide firms within a competitive business environment. We must expand

our understanding of cross-functional sourcing team effectiveness.

III. MODEL DEFINITION AND RESEARCH METHODOLOGY

This chapter presents a conceptual model of cross-functional sourcing team
effectiveness and defines specific model linkages. The first section discusses the
cross-functional sourcing team effectiveness model. The second section defines key
model variables. The third section details each research hypothesis subjected to
empirical testing. The fourth section addresses model construct development,
including a discussion of the literature sources contributing to measurement scale
development. Subsequent sections include a discussion of validity assessment,
reliability measurement, sample selection, sample size determination, and data

analytic procedures.

CROSS-FUNCTIONAL SOURCING TEAM EFFECTIVENESS MODEL

The cross-functional sourcing team effectiveness model created and evaluated
during this research identifies the relationships thought to affect team effectiveness.
The model, presented in Figure 6, resulted from an in-depth review of the team
effectiveness literature along with an evaluation of other leading process models of
team effectiveness. As such, this model follows the same pattern of input-process-
output defined by McGrath (1964). Direct experience with cross-functional teams
in an organizational setting also influenced model development.

The model proposes that team effectiveness is a complex phenomena resulting
from the effects of multiple variables. As such, the model does not predict that

cross-functional sourcing team effectiveness results from the effect of a single

80

81

MXOQNZOCm APP—graham

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

68... En
68298.88.— fur-$3
Hagan... cor-$0-
..83 38.20888 t
95.5.8: 5. .2. 89.3
038.?ng 06918
. 35.8... 0.555 2 088.9588... «930m -
- > E60383 6383 . 8 Illl
. a.“ III. 68... 55.. l- 8-... 028%98
:8...“ >388 r23— 0». mad: Mafia-5. 30 .68....ng Beacon—80
00935.88: o 9088 01.3.- 3.—
. 068095895 .. 9.1.0.358
. £93.85 ENS. a. 8355.88 8.3.3. .8:
.. gag won—E 0». 8335.838
on... 00.08.38
383.53— ... —II
3.8 NZUOQMZOCM <§E
Danna-maul ”89.38
En 89:8 8 «38
Can-BEE ”333 was
OCH—VGA“
23.3“ .530 n

 

End—cw a
0339.533000— mecnnmam an»... maooaégmm 3000—

82

variable. Each team provides certain micro and macro level inputs which can affect
the team interaction process. Micro-level inputs include team leadership, the
composition and experience of team members, and individual team member attitude
towards the collective team process. Macro-level inputs include the level of
organizational resources present to support team activities and the organizational
reward system which may encourage or inhibit team participation.

The model theorizes that a number of critical process variables result from a
cross-functional sourcing team's interaction. Conceptually, process variables occur
between time one and time two. Time one represents the initiation of team
interaction while time two represents a point in time in which team effectiveness is
evaluated for a given period. Model process variables include (1) the level of effort
put forth towards team assignments, (2) the level of team cohesiveness, and (3) the
quantity and quality of internal and external team communication. Within this
model, micro and macro-level variables can directly impact process variables. Micro
and macro-level variables should also demonstrate a direct relationship with cross-
functional sourcing team effectiveness.

The cross-functional sourcing team effectiveness model contains exogenous
or independent variables whose causal antecedents are outside the scope of this
research. Exogenous variables help explain other variables but are not themselves
explained within the model (Hunter and Gerbing 1982). The exogenous variables
in this model include (1) the organizational reward system, (2) team composition and

experience, (3) team leadership behavior, (4) member attitude, and (5) the

83

organizational resources provided to the team. While these variables are critical to
the study of team effectiveness, the model does not explain their origin. It is
possible, perhaps even likely, some dependence exists between the exogenous
variables. However, the model does not detail any specific relationships between
exogenous variables.

A primary objective of the literature search was to identify preliminarily the
variables that might relate to cross-functional sourcing team effectiveness. The
specification of all meaningful antecedents of a phenomena greatly increases the
quality of estimation parameters (Hunter and Gerbing 1982). Hunter and Gerbing
argue that the presence of all relevant antecedents of a variable improves the model
conceptually and provides improved statistical estimation. Perfect prediction from
one variable to another can never occur unless the model includes all significant
antecedents for each dependent variable. To the extent the model includes all
meaningful predictors, good parameter estimation should occur even with a modest
sample size.

The model presented in Figure 6, however, does not include all factors
potentially influencing cross-functional sourcing team effectiveness. It would be
impractical to construct a model that relates each potential determinant of sourcing
team performance. The cross-functional sourcing team effectiveness model reflects
a tradeoff between depth and sc0pe. It is better to develop an in-depth
understanding about a reasonable number of variables than to generate minimal

knowledge about many variables. The variables specified in the cross-functional

84

sourcing team effectiveness model are expected to exert, however, a disproportionate
effect on team performance.

Several assumptions underlie model development. First, the model assumes
the relationships between the variables are linear. While most relationships between
variables, if taken to extremes, might exhibit nonlinear tendencies, there is no reason
to believe the linearity assumption is not valid. Secondly, the model identifies only
recursive, or unidirectional, relationships. If the unidirectional assumption is not
valid, this research cannot apply certain analytic techniques. Use of the product rule
is no longer valid, for example, when a model specifies nonrecursive relationships
between variables (Hunter 1991).

Development of the cross-functional sourcing team effectiveness model
preceded development of the measurement model. Hunter (1986) argues that the
practical concerns often associated with the measurement model should not influence
the development of the theoretical model. For example, models often require the
invention of new variables or constructs. Preoccupation with existing constructs and
measurements take away from a researcher's ability to create and innovate. Once
the model specifies the relationships among the variables, then, and only then, should
‘ a researcher devise methods to measure the variables.

Previous research and experience supports directly a portion of the conceptual
links between model variables. Specific linkages between independent and

dependent variables are discussed within the research hypothesis section. The

85

hypothesis section operationalizes the cross-functional sourcing team effectiveness

process model subject to empirical testing.

VARIABLE DEFINITION

This section discusses each variable included within the cross-functional
sourcing team effectiveness model. This provides an understanding of each model
variable before discussion of the of research hypotheses. Each variable consisted of
a self-reported or externally-reported evaluation from the team leader, team
members, or external team rater. A self-reported measure is one requiring some
indication of the respondent's feeling, perception, or opinion about an item. An
externally-reported measure can occur in three ways. First, an externally-reported
measure can be an answer or response to a non-judgmental item (i.e respondent's
age or years of experience). Second, an external measure can be a response reported
by an individual outside the team. For example, an external rater's evaluation of
team performance is an externally reported measure. Finally, an external measure
can be a respondent's perception or evaluation to an item that does not pertain
directly to the respondent. For example, member rating of team leadership behavior
qualifies as an externally reported measure.

Team Leadership Team leadership consisted of three externally-reported
dimensions of team leadership. Team members rated the team leader against two
behavioral dimensions--consideration and initiating structure. Leaders rated high on

the consideration dimension have a people-oriented focus and are supportive and

In-’

86

interested in others. A leader high on consideration enjoys working with others, has
a positive view of the contributions made by others, and wants to be perceived
positively by others (Gratzinger, Warren, and Cooke 1990). Leaders rated high on
the initiating structure dimension are task and goal-oriented. This type of leader is
usually adept at securing team resources and organizing and coordinating team
activities (Gratzinger, Warren, and Cooke 1990). Each team member evaluated team
leadership behavior using a modified version of the Leadership Behavior Description
Questionnaire (LBDQ) (Stogdill 1957). The team leader does not evaluate his or
herself. Please refer to Chapter II for a discussion of the LBDQ. Team members
also evaluated the team leader against nine key tasks of leadership. These nine items
combined to form a single scale titled resource/commitment attainment.
Organizational Resources This variable included the different resource
categories required to accomplish a team's task assignment. A single organizational
resource value resulted from combining linearly the quantitative rating of each
resource category. The categories presented in this research represented a slightly
modified version of Peters and O'Connor's (1980) situational resource constraints.
The ten resource categories included (1) job-related information, (2) tools and
equipment, (3) materials and supplies, (4) budgetary support, (5) required services
and help from others, (6) task preparation, (7) time availability, (8) the physical work
environment, (9) executive level support, and (10) supplier participation. Team
members, team leaders, and external team raters evaluated the availability of each

organizational resource category.

87

Organizational Reward System This variable assessed the degree to which
individual participation on team assignments affected individual performance
evaluation. It supported evaluation of the impact that individual contribution
towards team activities and/or overall team performance might have on
organizational rewards and individual performance evaluation. Organizational
rewards include salary increases, promotions, bonuses etc.

A second dimension measured the respondent's perception about the
organizational performance evaluation and reward system. This determined whether
respondents perceived that the organizational evaluation and reward system
supported team interaction and encouraged personal commitment to team activities.
The team leader and team members responded to the items within this construct.

Team Effectiveness This measure focused on the team's ability to generate
a productive output that met or exceeded organizational performance standards or
expectations. The external rater, team leader, and individual team members
evaluated empirically team effectiveness against ten performance indicators. Internal
and external raters also assessed the team's ability to meet or exceed eighteen
sourcing performance indicators. The results of exploratory factor analysis presented
in Chapter IV identify four unique dimensions of team effectiveness identified during
measurement model scale analysis.

Team Experience The team experience variable included four planned
dimensions, each of which represented a separate study variable. The first dimension

was functional diversity. This is simply the number of unique functions represented

88

on a team. The second dimension was the average professional and/or functional
tenure of team members. This supported comparisons of professional and functional
longevity and team effectiveness. The third dimension measured age dispersion
within a team. The standard deviation of team member ages represented the
dispersion measure. Finally, calculation of the average number of cross-functional
teams each member has been formally a part of occurs. The minimum scale value
for the cross-functional experience dimension was 1.0 for each team member and
team leader.

Team Level of Effort This multi-item construct evaluated the level of effort
each team expended towards the accomplishment of its task assignment. The team
leader, team members, and the external rater evaluated total team effort.
Respondents did not evaluate their personal level of effort per se but rather the
team as a whole or the effort of the other team members.

Team Communication Team communication included three planned
dimensions which are unique study variables. The first dimension included a self-
reported multi-item assessment of the openness and accuracy of communication
between team members. A second dimension focused on the frequency of within-
team communication. A third dimension evaluated the frequency of communication
between the cross-functional sourcing team and individuals or groups external to the
team. The team leader and team members provided responses for each

communication variable.

89

Team Cohesiveness This measure contained multiple items relating to self-
reported perceptions of the team's ability to work together to accomplish its goals
and objective. A cohesive team is one characterized by ( 1) trust between members,
(2) a willingness to provide mutual support between team members, (3) a lack of
destructive conflict, (4) a feeling of team togetherness, and (5) a willingness of team
members to pursue future assignments together. The team leader and team
members responded to the team cohesiveness measure.

Member Attitude Member attitude is a multi—item measure of each team
member's attitude toward collective team interaction. Individuals scoring high on the
collectivism measure indicated a desire to work and cooPerate with other team
members as well as a willingness to sacrifice for the sake of the team. Team
members with a low score on this scale prefer individual versus group or team

assignments. The team leader and team member completed this measure.

RESEARCH HYPOTHESES

Specification of the relationships within the cross-functional sourcing team
effectiveness model supported the development and testing of separate research
hypotheses. Assessing the relationship between dependent and independent variables
allowed an examination of the statistical relationship between model variables. Six
dependent variables were subject to in-depth hypothesis testing and discussion within
this study: (1) team level of effort, (2) team cohesiveness, (3) openness/accuracy of

team communication, (4) within-team frequency of communication, (5) external-team

9O

frequency of communication, and (6) team effectiveness. Figures 7-12 graphically
present each a priori hypothesis evaluated during this research. Please note that all

references to teams in each hypothesis refer to cross-functional sourcing teams.

Hypothesis 1

Hypothesis I examined the linkages between a cross-functional sourcing team's
level of effort and (1) an organization's reward system, (2) leadership behavioral
style, (3) team cohesiveness, (4) average team professional tenure, and (5) member
attitude towards team interaction as measured by an indicator of individual member
collectivism.

Hypothesis l-A: Team level of effort relates positively to a system that

recognizes and rewards individual contribution to team activities and/or

overall team performance.

As organizational rewards (i.e. salary increases, promotions, bonuses etc.)
directly connected to member contribution and/or team performance increased, the
level of team effort should also increase. A well-researched link between
organizational reward incentives and productivity exists within the literature.
Pritchard, Jones, Roth, Stuebing, and Ekeberg (1988), in their review of the
literature related to performance incentives, concluded that (1) incentive or reward
systems are effective in increasing group productivity (and presumably group level

of effort) and (2) a large research literature exists supporting the effectiveness of

different types of incentive systems on individual and team level of effort.

91

Cosmos/mam

$33330:
Woémam

+
/ $~mo Homo-

filo/i 0». mmmon Emacs”: Hanan

.\ /

Emmmbm mgogo H08:
FREQ. wag/AS Corommébomm

I—I

Emmi—w q
MESH—5am H

92

Hypothesis l-B: A cross-functional sourcing team whose leader exhibits a

higher level of initiating structure behavior demonstrates a higher level of

effort towards team activities than a team whose leader exhibits a lower level
of initiating structure behavior.

In his review of leadership and group performance literature, Stogdill (1981)
concluded that task-oriented leadership behavior more often than relationship-
oriented leadership behavior relates at a higher level to team effort and output. This
results in the proposition that a cross-functional sourcing team whose leader
demonstrates a higher level of initiating structure leadership behavioral style should
also demonstrate greater team effort. Stogdill (1981), however, also cites a
longitudinal study that concluded relationship-oriented behavior and task-oriented
behavior were both found causally antecedent to group/team effort. Some ambiguity

exists in this area which can make the results of this research even more interesting.

Hypothesis 1—C: Highly cohesive teams exert a higher level of effort towards
team assignments than less-cohesive teams.

Although this hypothesis expects a relationship between cohesiveness and
effort, we must recognize that under the right conditions a highly cohesive team can
actually enforce a lower team performance norm and therefore a lower level of team
effort. Because teams included in this research consisted of professional managers,
a lower team effort resulting from the interaction of low performance norms and
high cohesiveness was not thought a priori to be a major concern.

Hypothesis 1-D A team whose members have a positive attitude towards the
team process exhibit a higher level of effort towards team assignments.

A perceptual measure of team member collectivism (versus individualism)

evaluated each team member's attitude towards the team process. A team whose

93

members demonstrated a higher level of collectivism should relate positively to a
team's effort.

Hypothesis l-E: Average team member professional tenure relates negatively
to team effort.

This examined the proposition that younger employees exert a higher overall
level of team effort compared to employees with longer tenure. Presumably, younger
team members have been exposed to the cross-functional team concept at an earlier
stage in their professional career compared to team members with longer tenures.
As a result, younger teams will generally exert a higher level of effort towards team

assignments. This line of reasoning is consistent with results obtained by Ancona

and Caldwell (1989).

Hypothesis 2
Hypothesis 2 focused on variables hypothesized to relate directly to team
cohesiveness. It proposed that a positive relationship exists between the presence of
a leader exhibiting a considerate behavioral style, member attitude as represented by
a measure of team collectivism, and team cohesiveness.
Hypothesis Z-A: A positive relationship exists between a leader who
demonstrates consideration leadership style and a team's level of
cohesiveness.

Stogdill (1981), in his summary of leadership and group performance, cites

extensive literature that concluded that relationship-oriented leadership behavior

94

Panama 0955330:
wag/mow

H88 Coromwébomm

7

Oozoomima

and—cw m
mafia—Emma N

95

contributes directly to group cohesiveness. However, Stodgill was also able to cite
literature supporting the position that task-oriented, directive leadership behavior was
seen to increase cohesiveness. Finally, he also presents evidence that both leadership
styles (relationship and task-oriented) were important to cohesiveness. Although the
evidence is ambiguous regarding leadership behavior and group/team cohesiveness,
this research tests the proposition that a direct relationship exists between
relationship-oriented leadership and team cohesiveness. Cross—functional sourcing
teams with leaders rated high on the consideration dimension of leadership, as
measured by the Leadership Behavior Description Questionnaire (LBDQ), should
exhibit a higher level of team cohesiveness compared to leaders rated lower on the
consideration dimension of the LBDQ.

Hypothesis 2-B: A positive relationship exists between a team whose
members support team interaction and a team's level of cohesiveness.

Team members who share a positive attitude about team interaction and the
team process should demonstrate a higher level of cohesiveness compared to
members who prefer an individual work environment or are unwilling to make

sacrifices for attaining team goals.

Hypothesis 3

Hypothesis 3 focused on the openness/accuracy of cross-functional team
Communication. Communication should be a particularly critical in a cross-functional
Sourcing environment. Ancona and Caldwell (1989) cite numerous studies that

conclude (1) the amount and pattern of communication within the team and (2) the

96

Emmnbm mgnga + Ovoitwnofimov.

bananas? IIIIIII cm 438
worming 0085:3830:
‘ .
Hog
OOWomwébomm
Efidfim w

£5535me m

97

amount and pattern of communication between the team and outside groups relates
directly to cross-functional team performance.

Hypothesis 3-A: Teams with higher levels of cohesiveness exhibit greater
openness and accuracy of within-team communication.

Shaw (1981) has cited extensive research that indicated the pattern and
content of communication interaction (i.e. the openness-accuracy) was more
positively oriented in highly cohesive groups. We expect teams that show a
willingness and desire to interact as a group will demonstrate a higher level of open
and accurate communication between team members. Openness/accuracy is a
positive communication dimension which does not include negative or destructive
within-team communication.

Hypothesis 3-B: A positive relationship exists between leaders rated highly on

initiating structure leadership behavior and the openness and accuracy of

within-team communication.

Organizing positive channels of communication is a critical aspect of the team
leadership role. Halpin and Winer (1957) have maintained that a key activity
associated with an initiating structure behavioral style is to organize channels of
communication within a group or organization. Team leaders rated higher on the

initiating structure behavioral dimension should lead teams with a higher level of

open and accurate communication exchange between team members.

Hypothesis 4
Hypothesis 4 examined the linkages between within-team frequency of

communication and the following independent variables: (1) leadership behavioral

98

style, (2) team cohesiveness, (3) within-team age dispersion, and (4) within-team
functional diversity.

Hypothesis 4-A: A team whose leader exhibits higher initiating structure
behavior will have a higher level of within-team frequency of communication.

As mentioned, Halpin and Winer (1957) have maintained, a key function of
the initiating structure leader is to establish channels of within-team communication.
Once the leader establishes within-team communication channels, we would expect
increased frequency of within-team communication.

Hypothesis 4-B: A team's level of cohesiveness relates positively to within-
team frequency of communication.

Shaw (1981) cited extensive research indicating that members of highly
cohesive groups communicate with each other more often than members of less
cohesive groups. Numerous studies have shown that the amount of communication
and interaction is greater in groups/teams composed of members who are highly
attracted to the group (Shaw 1981). Therefore, we expect members of highly
cohesive cross-functional sourcing teams to demonstrate higher levels of within-team
communication compared to members of less cohesive teams.

Hypothesis 4-C: Within-team age dispersion relates negatively to within-team
frequency of communication.

Zenger and Lawrence (1989), in their study of the differential effects of age
on technical communication, found that the frequency of communication among
research team members related positively to age similarity. This link investigates this
relationship between age dispersion and within-team communication frequency in a

cross-functional sourcing environment.

99

How:—
Coromwébomm

T

Emmmbm + $567633 .
922:8 homaonrllv $2338. 0». All. ”wow/MWMME
mam wormfiou. 0083:3330: - v.

>mo
memaamwg

and: 3
MESH—Emma A

100

Hypothesis 4-D: Higher levels of within-team functional diversity relate
negatively to within-team frequency of communication.

Although most firms perceive that the use of cross-functional teams supports
increased inter-departmental communication, this link proposes that teams with
higher levels of functional diversity will actually exhibit lower levels of within-team
communication frequency. Cross-functional sourcing team assignments are usually
part-time for team members. Furthermore, most team members are not co-located
together. Higher levels of functional diversity means that a higher proportion of
team members will be physically separated. This separation requires greater effort
to initiate within-team communication, particularly informal or impromptu
interaction. Other things equal (such as the frequency of formally scheduled
meetings or physical distance between team members), we expect teams with higher
functional diversity to demonstrate lower levels of within-team communication

frequency.

Hypothesis 5
This hypothesis evaluated the relationship between extemal-team frequency
of communication and (1) leadership behavioral style and (2) a team's level of
functional diversity.
Hypothesis 5-A: A team whose leader demonstrates a higher level of
initiating structure leadership behavior will have a higher level of extemal-
team frequency of communication.

The positive relationship between initiating structure and external-team

communication frequency should occur for the same reasons presented in

101

Emmmbm madam—:0 +
$39.me Illnllvv
wormioH

Eadwm E
mica—Emma m

9:93: H08:
33:38. am
0033:3030:

_ .

3532me
Umfifiqu

102

Hypotheses 3-B and 4-A for the within-team frequency and the openness-accuracy
of communication dimensions.

Hypothesis S-B: Frequency of external-team communication relates positively
to a team's level of functional diversity.

Ancona and Caldwell (1989) noted that representatives of a cross—functional
team have a higher level of contact and greater ease of external communication with
members of their own functional areas. This results from a shared language,
socialization, and worldview. Because of this, we expect that cross-functional
sourcing teams with a higher level of functional diversity to communicate externally
with nonteam members or groups about team-related issues at a higher level than

teams with a lower level of functional diversity.

Hypothesis 6

Hypothesis 6 evaluated directly cross-functional sourcing team effectiveness.
The independent variables within this hypothesis include (1) level of team effort, (2)
Organizational resource availability, (3) leadership behavioral style, (4) the quantity
and quality of within-team and external communication, (5) team cohesiveness, (6)
functional team diversity, (7) average team member professional tenure, and (8)
previous cross-functional team experience.

Ancona and Caldwell (1989), in their study of cross-functional product teams,
identified three dimensions of a multi-item team effectiveness measure through
principal components analysis. These dimensions included (1) team efficiency in

developing technical innovations, (2) adherence to budgets and schedules, and (3)

103

a composite measure of overall performance. As detailed in Chapter IV, this
research identified and evaluated four separate dimensions of cross-functional
sourcing team effectiveness.

An external rater, team leaders, and team members responded to the team
effectiveness measures. External reporting comprises a major dimension of team
effectiveness as defined by Hackman (1987). The rater determined whether the
collective output of the team met or exceeded organizational expectations or
objectives across multiple performance items.

Previous work by Hackman (1987) provided conceptual support for the direct
linkages between certain variables and team effectiveness. The basic proposition that
underlies Hackman's (1987) normative model of team effectiveness is that team
effectiveness is a joint function of (1) the level of effort group members expend to
carry out the team's task and (2) the amount of knowledge and skill members bring
to bear on the team's task (i.e. professional team member experience in the sourcing
team effectiveness model). A third variable in the Hackman model, the
appropriateness of the performance strategies used by the team throughout its
assignment, is not part of the cross-functional sourcing team effectiveness model.
Hackman has argued that these variables are responsible for the majority of a team's
effectiveness.

Kolodny and Kiggundu (cited in Goodman 1986) have maintained team
effectiveness is primarily a function of three variables. Although they use slightly

different terminology, their theoretical model of team performance specifies that

104

H92; 0». mad: OHmBmNmfimoaa ”80:88

><0Sm0 50m00m~0ba+ 000mao~mm00
H0050 \ramaaBEH warming.

><o~mmo 0H0mm/’. + 0n0mm-m:boaobm_ Hammmbm $2088
9530:»— mxuongoa .Ilv mofiombm H003 All] Ramafiwau warm—son
mmooaénamm +

95308: 9’585 \ Owoubommtwoofimov.

+ 0». 0083:3030:
+
+

H008 00:3?0003 + 2505 H008 330003
0m 008803030:

mxSBE H005 mawncabow
0m 0088:3830

EGG—Wm 5
magi—Emma 0

105

member knowledge and skill (i.e. professional tenure and cross-functional team
experience in the sourcing team effectiveness model), within-team communication,
and team leadership will exert a dominant and direct effect on team performance.

Hypothesis 6-A: The level of effort a team demonstrates towards its
assignments relates positively to cross-functional sourcing team effectiveness.

Hypothesis 6-B: The level of organizational resources provided to a cross-
functional sourcing team relates positively to team effectiveness.

Hackman (1987) also maintained that the availability of adequate material
resources required to accomplish a team's task is a primary determinant of team
effectiveness. Research by Peters and O'Connor (1980) has also supported the direct
link between resources and team effectiveness. They argued that attention to the
various categories of material resources required by a team is critical to effective
work performance. The direct link between resources and team performance also
appears in the Nieva, Fleishman, and Rieck (1978) theoretical model of team
performance. Finally, Guzzo (1986) has maintained that teams in an environment
rich in resources are more likely to be effective than those that lack adequate
resource support.

Hypothesis 6-C: A team whose leader exhibits a higher level of consideration

behavior is more effective than a team whose leader exhibits a lower level of

consideration behavior.

Hypothesis 6-D: A team whose leader exhibits a higher level of initiating

structure behavior is more effective that a team whose leader exhibits a lower

level of initiating structure behavior.

Previous research has supported the direct link between leadership and team

effectiveness (Stogdill 1981, Foushee 1984, McGrath 1964). Stogdill (1981) found

106

that a leader can positively or negatively affect a group's total effectiveness. Within
this research, a direct relationship should exist between teams whose leaders rate
highly on consideration and initiating structure dimension of the LBDQ.

Hypothesis 6-E: Open and accurate information exchange between team
members relates positively to cross-functional sourcing team effectiveness.

The link between communication and effectiveness is especially critical for
cross-functional teams. One of the primary benefits sought from cross-functional
interaction is the improvement of organizational communication. Presumably, a
reduction in communication barriers through cross-functional interaction should
relate directly to a higher level of team effectiveness.

Pinto and Pinto (1991) evaluated the direct link between communication and
team effectiveness. A key conclusion of their research was that the vast majority of
ideas originating in cross-functional teams come from the direct, face-to-face contact
and communication of team members. Their research supported the position that
positive communication, cooperation, and coordination between team members
directly leads to a successful performance outcome. They argued that successful
communication between team members results directly in a higher level of team
performance. Research conducted by Seers (1989) found that team performance
related systematically to the quality of information exchange (i.e. openness and
accuracy) between individual team members.

Hypothesis 6-F: Teams with higher levels of within-team frequency of

communication will demonstrate a higher level of cross-functional sourcing
team effectiveness.

107

Hypothesis 6-G: Teams with higher levels of extemal-team frequency of
communication will demonstrate a higher level of cross-functional sourcing
team effectiveness.

Ancona and Caldwell (1989) have argued that cross—functional teams are
particularly dependent on communication patterns and the communication process.
This includes communication within the group and with external groups. Cross-
functional teams must obtain information and resources from groups external to the
team, communicate internally to transfer this information, and transfer back
processed information to external groups. Thus, within-team and external-team
frequency of communication should demonstrate a direct relationship with cross-
functional sourcing team effectiveness.

Ancona and Caldwell (1989) also noted that high performing cross-functional
teams rely on an interrelated two step communication process. First, team members
obtain information from outside the group. Second, those individuals obtaining
information within the group then transmit this information to the rest of the group.
When this process flows smoothly, more effective team interaction and performance
can result. Other studies have concluded that the amount of communication
(particularly technical communication) within the team and between the team and

outside groups related directly to team performance (Allen, 1984, Katz 1982).

Hypothesis 6-H: A positive relationship exists between team cohesiveness and
sourcing team effectiveness.

In their literature review, Pearce and Ravlin (1987) noted that teams
responsible for interdependent tasks (such as cross-functional sourcing teams) show

a clear, positive relationship between task-based cohesiveness and team performance.

108

Additional research cited by Pierce and Ravlin indicated a positive relationship
between cohesiveness and performance but raised the issue of causal direction.
Performance may, at least in part, influence cohesiveness as opposed to the more
widely accepted unidirectional reverse relationship.

Hypothesis 6-1: Higher levels of functional diversity relate negatively to team
effectiveness.

While this appears to counter a popular belief and perception about cross-
functional interaction, there is reason to speculate that an inverse relationship exists
between effectiveness and diversity. For the teams included in this research, team
task assignments were predominantly purchasing or materials related. A cross-
functional sourcing team with a higher level of functional diversity (and hence
nonpurchasing functions) may not accept purchasing related assignments as readily
as a team weighted more heavily with purchasing personnel. Ancona and Caldwell
(1989), in their study of cross-functional product development teams, found that a
higher level of functional diversity related negatively to each team performance
variable. .

Hypothesis 6-J: A team whose members have a higher level of cross-

functional team experience is more effective than a team whose members

have a lower level of cross-functional team experience.

This hypothesis proposes that a learning and experience curve exists for cross-
functional team participation. Because direct cross-functional interaction is a
relatively new experience at most US. firms, it was expected that teams with higher

levels of previous cross-functional team experience would perform more effectively

than teams with lower levels of experience.

109

Hypothesis 6-K: Cross-functional sourcing teams with a higher level of
average professional team member tenure are more effective than teams with
a lower level of average professional team member tenure.

This hypothesis reflects a belief that a team composed of higher professionally

tenured members should demonstrate greater task knowledge, information, and

experience. This, in turn, should impact directly team performance.

MEASUREMENT MODEL DEVELOPMENT

This research required the development of a measurement scale for each
multi-item variable included in the cross-functional sourcing team effectiveness
model. This required the statistical analysis of each multi-item scale to develop final
measures that were accurate, efficient, and evaluated the. construct of interest. A
major contribution of this research is the development of the measures included in
the cross-functional sourcing team effectiveness model.

Literature sources from a variety of disciplines supported the development of
each construct measure. A review of prior research related directly to each construct
within the model comprised an additional detailed literature search. The review of
previous research and measurement scales cannot be overemphasized. Advances in
measurement occur when researchers take the time to review and evaluate previous
work related to a current research interest (Price and Mueller 1986). This allows a
researcher to benefit from previous research while offering an opportunity to

improve scale items or construct measures.

110

The evaluation of existing measurement scales and items followed a certain
sequence. First, the constructs of research interest were identified and defined
during the literature review discussed in Chapter 11. Next, a second literature review
identified and evaluated existing scales and items related directly to the variables
hypothesized to impact cross-functional sourcing team effectiveness. This increased
the pool of items while contributing to the efficiency of measurement instrument
development. Third, the items from all sources, including items developed
specifically for this research project, were evaluated for content or face validity. A
lack of content validity resulted in the elimination of items from the pool of total
items.

The following section provides a brief description of the sources consulted
during development of the measurement scales for the model constructs.

Mountain Bell Marketing Development Project Team Survey This survey,
developed at the Graduate School of Business at Columbia University, is one of the
few recent attempts to evaluate a model of team effectiveness (Gladstein 1984). The
Mountain Bell survey included items developed by the staff at Columbia University
as well as survey items used in the Yale work team effectiveness questionnaire
developed by Hackman, the Group Effectiveness Survey by Nadler, the Michigan
Assessment of Organizations Survey, and Van de Ven's Measuring and Assessing
Organizations Survey. Items adapted from this resource appeared in the level of

effort and team cohesiveness construct.

111

Michigan Assessment of Organizations Survey This is a major survey
developed by the Survey Research Center at the University of Michigan (1975). One
section of this document pertained to individuals working in group or team task
settings. Items included (and adapted where required) from the MAO survey
appeared in the (1) team level of effort and (2) team cohesiveness constructs.

Mowday and Steers The original scale developed by these authors, as
presented in Price and Mueller (1986), assessed the possible feelings an individual
might have about the company or organization for which they worked. Several of
the questions from the Mowday and Steers survey were modified and included in the
level of effort and team cohesiveness constructs.

Yale Effectiveness Work Team Survey The team effectiveness project,
sponsored by Yale University and directed by J. Richard Hackman, is one of the
more comprehensive attempts at analyzing team interaction and performance. A key
part of this project was the development of a self-reported work team survey.
Several of the constructs included in the Yale research were directly or indirectly
included in Hackman's project. Unfortunately, the items or scales presented in
Hackman's project have not been subject to validity assessment or reliability
measurement. Nonetheless, the work team survey provided several items included
during the development of the cross-functional sourcing team measuring instrument.
The measures that included scale items from the Yale project are: (1) team level of

effort and (2) team cohesiveness.

112

Ohio State University Leadership and Group Dimensions Scales- This
research, which dates back to the 19505, provided several items related to the team
cohesiveness scale. The cross-functional sourcing team effectiveness survey also
included a factor analyzed version of the Leadership Behavior Description
Questionnaire consisting of approximately 48 items measuring the two behavioral
dimensions--consideration and initiating structure.

Pinto and Pinto Pinto and Pinto provided one of the few research studies
that has focused directly on some aspect of cross-functional teams. These
researchers concentrated on the role of cross-functional communication and
cooperation throughout the development of new programs. To evaluate their
hypotheses, the researchers developed a fifteen item scale related directly to cross-
functional communication and cooperation. This scale exhibited a high degree of
reliability as measured by coefficient alpha. The openness/accuracy dimension of the
team communication measure included several of the Pinto and Pinto scale items.

O'Reilly and Roberts These researchers developed a ten item scale to
measure the accuracy and openness of communication within a group or team.
These items, as presented in Price and Mueller (1986), related directly to the
openness/accuracy dimension of team communication. Several items from the
O'Reilly and Roberts scale were included in the final version of the measuring
instrument for this research.

Van de Ven's Organizational Assessment Survey Van de Ven (1980)

developed the Organizational Assessment Survey to discover how various jobs, work

113

groups, sections, and levels within an organization work together. Because of its
comprehensiveness, many of the survey's items related directly to different
dimensions of the cross-functional sourcing team effectiveness variable.

A key feature of the Organizational Assessment Survey included its variety of
questions and answer formats. Van de Ven developed a series of valuable items that
assess team effectiveness. These items form part of the team effectiveness measure.
Items modified or included from Van de Ven's survey appeared in the following
measurement scales: (1) team level of effort, (2) team cohesiveness, and (3) within-
team and external frequency of communication, and (4) team effectiveness.

The literature review of related measurement scales served two purposes.
First, the review supported the evaluation of hundreds of potential items. Secondly,
this research benefited from previous work conducted by excellent researchers.
While not all of the questions from these secondary sources related directly to cross-
functional sourcing teams, many did address some of the factors hypothesized to
impact cross-functional sourcing team effectiveness.

The final measurement model modified or included a portion of the items
identified during the second literature review. Some items clearly lacked content
validity and were eliminated from further consideration. Others failed the statistical
tests of internal consistency and parallelism (i.e. external consistency) during the final
item analysis.

Three versions of the measurement instrument exist: (1) team member

version, (2) team leader version, and (3) external rater version. The team member

114

and team leader survey instrument are identical except for the last section. The first
section of each version includes demographic items such as education, sex, age,
professional experience, and previous cross-functional team experience. Subsequent
sections evaluated each multi-item construct--team communication, team
cohesiveness, performance measurement and evaluation, team level of effort,
member attitude, organizational resources, and organizational culture and ,
environmental uncertainty. A following section requested responses to questions
specific to the cross-functional sourcing team process.

The last sections of the leader and member survey instrument differ. For the
team member version, the last section required the rating of team leadership
behavior. For the team leader, the last section asked for responses to questions that
the leader answered for the entire team. Of particular importance is the leader's
evaluation of supplier participation and cross-functional sourcing team authority.

The external rater evaluated team performance, level of effort, and the

presence or absence of organizational resource support.

VALIDITY ASSESSMENT

The content and construct validity of the measurement scales was a critical
research issue. Validity refers to the degree to which an instrument or scale
measures the construct or factor under investigation (Bohmstedt 1969). While

validity is an important topic, researchers often ignore the assessment of each

115

construct's validity. Construct validation often gives way to the more exciting
dimensions of the actual research (Schwab 1980).

Several methods are available to assess construct validity. This research relied
on confirmatory factor analysis to address the validity of each construct measure
(Nunnally 1967 and 1978, Hunter and Gerbing 1982, Bohrnstedt 1969). Factor
analysis is useful (1) to determine the dimensionality of a domain and (2) to select
the items that fit best into the various strata of the domain (Bohrnstedt 1969).
Hunter and Gerbing (1982) have argued that from the substantive viewpoint of
theory building, the most important criterion is the content validity of selected items.
Evaluating the content or face validity of each item in a construct is the initial step
when assessing validity through confirmatory analysis.

The face validity of each item is a necessary but not sufficient condition of
construct validity. Construct validity must move beyond the assessment of the face
or content validity of each item. This requires the statistical evaluation of the
measurement model (Hunter and Gerbing 1982).

This research employed a statistical approach defined by Hunter and Gerbing
(1982) to assess each measure's validity. The first statistical step in the validation
process required the assessment of items within each multi-item scale. If all the
items in a cluster of items measure the same factor, then we expect the correlation
between the items to satisfy a product rule for internal consistency. Internal
consistency implies that if all the items in the measurement scale have equal quality,

any two items will have the same correlation to within sampling error (Hunter and

116

Gerbing 1982). This means each item has the same correlation with the cluster true
score. A "flat" correlation matrix satisfies the condition of internal consistency.

A second statistical analysis evaluated the external consistency or parallelism
of each construct. This analysis evaluated how items in a construct correlated with
items outside the construct. Items of equal quality within a construct should have
equal correlations within sampling error with other constructs. Furthermore, the
construct itself should not exhibit a high correlation with other constructs.

The objective of this process was to develop unidimensional measurement
scales. This required each scale item to exhibit face or content validity, internal
consistency, and a reasonably low correlation with other factors or item clusters.
Chapter IV reports on the final measurement scale items used during the testing of

research hypotheses.

RELIABILITY MEASUREMENT

Reliability refers to the extent to which measurements are repeatable
(Nunnally 1978) and the degree to which a measurement instrument is free of
random error and exhibits a high degree of accuracy or precision (Kerlinger 1986).
The primary objective of measuring the reliability of each scale was to calculate a
reliability coefficient associated with each construct's measure. The reliability
coefficient is the estimated square of the correlations of scores on a collection of

items with true scores (Nunnally 1978). Nunnally also stated that the reliability

117

coefficient is the estimated average correlation of that test with all possible tests with
the same quantity of items obtainable from sampling a domain.

This research relied on coefficient alpha to evaluate the measurement error
associated with each multi-item construct. Nunnally has argued that coefficient alpha
is a key deduction from the domain-sampling theory of measurement error. The
domain sampling model theory states that any particular measure contains a random
sample (or a sample considered to be random) of items from a hypothetical domain.
Coefficient alpha is the expected correlation of one test with another test of the same
length from a domain when the two tests purport to measure the same construct.
The closer the value of coefficient alpha is to 1.0, the more reliable or accurate is the
measure of the underlying variable (Hunter and Gerbing 1982).

Hunter and Gerbing (1982) have argued that coefficient alpha provides an
unbiased estimate of the reliability of the cluster score only if the scale is
unidimensional. A researcher should interpret the value of coefficient alpha only
after a scale demonstrates face validity as well as internal and external consistency.
This reveals the importance of assessing the validity of the measurement model

before determining each construct's final level of reliability.

SAMPLE SIZE DETERMINATION
Sample size determination was a critical methodological issue in this research
for several reasons. First, sample size affects statistical power. Power is the ability

of avoiding a Type II error--accepting the null hypothesis when it is actually false.

118

Cohen and Cohen (1983) have argued that power analysis is an integral part of the
research planning process. They maintain that attention to statistical power supports
the related decisions concerning sample size, alpha, the power to which one can
aspire during the planning of research, and to some degree, the interpretation of the
research results afterwards.

Each research hypothesis required the testing of a specific link between a
dependent and independent variable. A power analysis for partial correlation and
regression coefficients supports determining the sample size necessary for testing the
null hypothesis that any partial correlation or regression coefficient is zero (Cohen
and Cohen 1983). A minimum acceptable level of power for this research was set
at .7, which is at the lower end of the conventional .7-.9 range of acceptable power
(Cohen and Cohen 1983). This value was selected because of the high cost of adding
additional cases (i.e. teams) to the total sample.

Power analysis indicated that a sample of 70-100 teams should support the
statistical requirements of each research hypothesis included within this research.
Note, however, that this sample size estimate was highly subjective. Power analysis
required an a priori estimate of the unique contribution of each independent variable
to the total value of R Square. There was no way of knowing a priori the total R
Square, the unique contribution of each independent variable, or the redundancy

between independent variables.

1 19
RESEARCH SAMPLE

This research collected data from 107 cross-functional sourcing teams at 18
major US. corporations. Represented industries included aerospace, automotive,
computing devices, consumer products, electronic, photographic, pulp and paper,
tractor and farm implement, and transportation equipment. Table 3 provides data
about each participating company. Of particular importance is the overall response
rate of almost 90 percent. This reflects the commitment that executive management
at each participating company extended to the project as well as the quality of the

firms and individuals responding to the survey.

Selection Criteria

Teams included in this study satisfied certain selection criteria. Each cross-
functional sourcing team had to be a recognized work team at the time of the study
or recently have completed its sourcing task. This means each team satisfied, at the
least, the three organizational attributes set forth by Hackman (1988). First, the
team was a real group with organizational boundaries, differentiated member roles,
and interdependence among members. Second, the team must have had one or
more tasks to perform that results or resulted in a collective outcome. Finally, the
team must have operated within an organizational context and manage relationships
with personnel or groups external to the team. These three attributes were necessary

but not sufficient conditions for final research selection.

120

Table 3
Cross-Functional Sourcing Team Effectiveness Research Project
Participating Company Data

Company Primary Number of Response

Number Industry Teams Rate
01 Communication 8 87%
02 Automotive 12 91%
03 Computing Devices 5 88%
04 Electrical 7 85%
05 Pulp and Paper 8 87%
06 Electronics 3 84%
07 Electronics 7 71%
08 Electronics 5 94%
09 Communication 3 91%
10 Electrical Devices 1 83%
11 Electronics 5 88%
12 Photographic 2 85%
13 Auto Components 3 85%
14 Communication 7 82%
15 Industrial Engines 4 91%
16 Farm Equipment 11 100%
17 Transportation Equipment 8 93%
18 Consumer Products 8 89%
Total 107 88%

121

As mentioned, teams selected for this research must have had a recent task
output against which to evaluate team effectiveness or an identifiable record of team
interaction and performance over time. Participating firms were asked to include
teams that demonstrated a range of success in meeting team or organizational
performance objectives. This was to avoid the possibility that organizations would
offer only their most qualified or successful teams for study. If this occurred, the
study would have had limited variability across key study variables

Because this research focused on cross-functional sourcing teams, additional
selection criteria existed. First, the team's predominant task must be purchasing and
materials related. Second, the team must have had at least one member from
purchasing or a materials related area. Third, team membership must have included,
at a minimum, three separate cross-functional specialties. Teams meeting these

additional qualifications were candidates for participating in this research study.

Research Process

Once the literature search and study proposal were complete, this research
followed a specific series of steps. First, approximately 35 firms received a packet
of information outlining the goals and objectives of the study. The packet included
(1) an introductory letter describing the purpose and requirements of the study, (2)
a description of the responsibility of the external team rater, (3) a flow chart
outlining the research process and timing objectives, (4) an executive

acknowledgement sheet used to indicate willingness to participate, and (5) a form to

122

identify the names and addresses of participating teams, members, leaders, external
raters. The researchers contacted firms directly from a nonrandom list. Requests
for participation were based on familiarity with the executive management of each
firm along with the knowledge that each firm used cross-functional teams for
sourcing decisions.

The executive acknowledgement provided initially to each contacted firm was
used to indicate a willingness to support or not support the research. If a firm
indicated its willingness to participate, management also indicated the number of
teams to be included in the study.

The second step of the research process involved identifying teams, team
members, leaders, and external team raters. Each firm also designated an individual
to act as an organizational coordinator for the project. This position acted as the
liaison between the researcher and each participating company. Executive
management was also requested to forward a message during this step to each team
member, leader, and rater notifying him or her of the firm's support of the project.

The third step involved sending questionnaires directly to team members,
leaders, and external raters. The questionnaire packet included (1) an introductory
letter, (2) questionnaire, (3) a sheet identifying which team the questionnaire
referred, and (4) a stamped self-addressed return envelope. Participants were
requested to complete the survey within ten to fourteen days of receipt and forward

the questionnaire directly to the researcher.

123

The fourth phase of the research involved monitoring and coordinating data
collection. This involved sending reminder letters, sending new questionnaires to
those participants requiring an additional copy, and contacting organizational
coordinators to facilitate data collection. In several instances executive management
sent letters to team members who had not returned the survey reminding them of
their firm's commitment to participate. Data collection covered a period from late
May, 1992 to early November, 1992.

The fifth phase involved data keypunching and analysis. Data analyses
occurred at the individual and team level. Initial measurement model testing, which
occurred midway through the data collection process, involved the use of team
member responses. Hypotheses testing, which involved team level data, did not
occur until data collection was completed. Measurement model analysis also
occurred again later in the project using team level data to verify the integrity of the
initial scales.

The final phase of the research project involved the formal presentation of
research results and conclusions. During this phase each participating company
received a white paper detailing the overall conclusions of the study. Furthermore,
each company that contributed at least three teams to the study also received an
analysis presenting survey data at the individual company level. This analysis also
included benchmark comparisons of certain measures to allow firms to compare their
performance or responses against other participating companies. This occurred

during the first quarter of 1993.

124

Appendices A-C present the packet of documents forwarded to companies
when requesting formal support for this project and to individual team members.
Appendix D provides an address to forward requests for a copy of the measuring

instrument(s) used during the research.

Demographic Study Indicators

Tables 4-7 provide profile data about the participants in this research project.
The demographic profile of participants is unique to this sample only. The profile
of teams and participants may or may not be externally generalizable to a random
sample of all cross-functional sourcing teams.

Table 4 indicates that the average team participating in this study consisted
of six to seven members, included four separate organizational functions, and had
worked together for 15 months. Team size information includes data only about
members recognized as formal team members and not other participants included
on a temporary or as-needed basis. Table 4 also indicates that a high probability
exists that the typical team member is male and has had previous cross-functional
team experience.

Table 5 reveals that the majority of study participants (over 70 percent) had
a college degree. Only 3.6 percent of participants had a high school education or
less. The cross-functional sourcing teams included in this study were comprised
almost exclusively of members of management. This helps explain the higher level

of education attained by the average team member.

125
Table 4

Cross-Functional Sourcing Team Effectiveness Research Project
Demographic Study Statistics

Number of teams in the study:
Number of firms:
Number of study respondents:

Average team size:
Average functional representation per team:
Average member age:

Member sex:

Average member professional experience:

Average member experience in current function:
Average member experience in current position:
Average member cross-functional team experience:

Average team longevity at time of study:

Table 5

Educational Level of Study Respondent

Educational Level

Some High School

High School

Some College

Junior College or Technical
Certificate or Degree

Non-Technical Bachelors Degree

Technical Bachelors Degree

Non-Technical Masters Degree

Technical Master Degree

Doctorate

N = 620 team members

Percent of
Responses

.2%
3.4%
15.5%
9.4%

21.3%
31.1%
10.3%
8.6%
.2%

107

18

622 members/leaders
107 external raters
6.7 members

4 functions

42 years

81% male

19% female

19.7 years

10.7 years

5.5 years

6 teams

17 months

126

While this study included participants from over twenty-five organizational
functions, Table 6 indicates that almost 70 percent of study participants are from
purchasing or engineering-related functions. The high degree of purchasing
membership is expected given the nature of sourcing team tasks. Furthermore, given
that the emphasis of many teams in the study focused on technical issues, greater
engineering involvement is also expected. Purchasing and engineering involvement
also reflects the technological emphasis of many of the participating companies.

Table 7 reveals that almost 60 percent of participating teams were located
flying or driving distance apart. This provided an opportunity to assess the
relationship between distance and other study variables. Again, geographic distance

between team members is also a function of the teams included in this study.

Team Tasks

Table 8 details the primary responsibilities of participating cross-functional
sourcing teams. The most common sourcing tasks included (1) developing cost-
reduction strategies, (2) evaluating and selecting suppliers, and (3) supporting
Sourcing requirements for items with established production specifications. Fewer
teams were responsible for developing sourcing strategies at the local, business unit,
01' corporate level. Just over half of the teams provided sourcing support during new

PTOduct development with less than half directly supporting new product design.

127

Table 6
Physical Distance Between Team Members
Percent of
Description Teams
Located flying distance apart 38.3%
Located driving distance apart 18.3%
Located walking distance apart 33.0%
Co-located in the same general area 9.6%
N = 107 teams
Table 7

Cross-Functional Sourcing Team Effectiveness Research Project
Functional Team Representation

Function Percent Function Percent
Purchasing 29.2% Planning/Control 1.5%
Quality Engineering 11.0% Maintenance 1.5%
Product Engineering 9.1% M18 1.3%
Process Engineering 8.8% R & D 1.1%
Other 6.8% Packaging 1.0%
Materials Management 6.6% Product Development .8%
Design Engineering 4.1% Advanced Technology .8%
Administrative 3.9% Legal .6%
Manufacturing 3.2% Personnel .5%
Marketing 2.9% Logistics .3%
Finance 2.4% Accounting 3%
Project Management 2.3%

N=617

128
Table 8

Cross-Functional Sourcing Team Tasks

Sourcing Team Task

Develop cost-reduction strategies

Evaluate and select suppliers

Support sourcing requirements for items with
established production specifications

Support sourcing requirements during
new product development

Perform supplier development activities

Support product design

Perform value analysis activities

Develop local sourcing strategies

Negotiate corporate-wide purchase agreements

Identify common purchased items between
corporate business units

Develop sourcing strategies for the entire
business unit

New technology development

Develop corporate-wide sourcing strategies

N = 107 teams

Percent of Teams
Pursuing Task

74%
73%
67%

56%

47%
44%
40%
38%
38%
36%

36%

34%
26%

Note: Team leaders were allowed to check off more than one

task assignment

129

There is no way of knowing whether this mix of sourcing team tasks is
generalizable across all US. sourcing teams. As with demographic indicators,
sourcing team task responsibility is a function of the teams and firms selected from

a nonrandom list.

DATA ANALYTIC PROCEDURES

This research followed a data analytic procedure outlined by Hunter (1986).
First, development of the conceptual model occurred before development of the
measurement model. During data analysis, testing of the measurement model
occurred before testing the linkages specified by the research hypothesis.
Verification of the measurement model should occur before conceptual hypothesis
testing (Hunter 1986).

This research employed a specific procedure to evaluate each a priori scale
included in the testing of the cross-functional sourcing team effectiveness model.1
This occurred after each scale item satisfied the criterion of content validity for
inclusion in the initial measurement instrument:

0 Input raw data for each item into a conventional statistical package

and create a correlation matrix. This matrix represents the

correlations between each item in the survey.

0 Use the correlation matrix for confirmatory data analysis. Specify a
priori which items belong with each factor or construct. The software

an

\

 

Exploratory factor analysis was used to evaluate scales with no a priori knowledge of
multiple dimensions. This includes team effectiveness, performance evaluation and
rewards, team authority, and supplier involvement.

130

output provides a new correlation matrix based on the a prior
clustering of items onto factors.

0 Verify the internal consistency of each item in each cluster or factor.
This requires the calculation of the average r and standard error for
each cluster.

0 Using the standard error and average r for each cluster, calculate
confidence intervals for each factor. Items with an r value outside the
range of the confidence interval are not internally consistent within the
cluster. Eliminate items that fail the test of internal consistency from
the cluster.

0 Compare the relation of the items in each cluster against other factors
to identify significant deviations from parallelism (i.e. external
consistency).

0 Once each factor contains items that satisfy the statistical procedures

of internal consistency (i.e. unidimensionality) and parallelism,
calculate final measurement scale reliability.

The second phase of data analysis involved evaluating the statistical
relationships between variables specified within each research hypothesis. This
research used hierarchical multiple regression analysis to identify the unique variance
attributable to each independent variable included within a hypothesis. Hierarchical
analysis is one of the most useful and powerful tools for extracting information from
a data set. (Cohen and Cohen 1983). It is also a valuable data analytic approach
because the results obtained from a simultaneous multiple regression analysis of all
independent variables may often be incomplete. Furthermore, hierarchical analysis
is the only basis on which variance partitioning can proceed with correlated

independent variables (Cohen and Cohen 1983).

IV. MEASUREMENT MODEL EVALUATION AND HY POTHESIS TESTING

This chapter consists of two major sections: measurement model evaluation
and the results of research hypothesis testing. The measurement model evaluation
section is further divided into multiple sections. The first section presents the multi-
item scales analyzed using a confirmatory analysis approach. This includes the team
communication, cohesiveness, level of effort, collectivism and consideration and
initiating task leadership behavioral factors.

The second measurement model section presents multi-item scales analyzed
initially with exploratory factor analysis and later with confirmatory factor analysis
to determine external consistency or parallelism. Variables analyzed strictly with an
exploratory approach are also discussed. These variables include supplier
involvement, team authority, and leadership effectiveness.

The second section of this chapter reports the results of research hypothesis
testing. The final section of Chapter IV reports the results of additional critical
variables suggested by the data during post hoc data analysis. Chapter V discusses

the research findings in detail.

MEASUREMENT MODEL EVALUATION

The statistical evaluation of multi-item constructs was a critical prerequisite
to research hypothesis testing. As Hunter and Gerbing (1982) have argued,
measurement model analysis should occur before analysis of individual research

hypotheses. A researcher should not use observed data values as measurements

131

132

unless there has first been an accounting of the nature and degree of error present
within the observations (Hunter and Gerbing 1982). Because of the importance of
reliable and valid measures, measurement model analysis received significant
attention throughout this project.

Measurement analysis focused on individual team member and leader
responses at the item level. The overall objective of measurement evaluation was to
arrive at a set of scales which demonstrated internal and external consistency.
Combining member and leader surveys into team groupings for analyzing individual
survey items was not required initially for measurement model evaluation.

The statistical evaluation of each individual multi-item measure used either
confirmatory or exploratory factor analysis. Analysis of certain multi-item constructs
developed a priori used a confirmatory approach. For other measures, no a priori
assumption of item loadings on construct or factor dimensions occurred. Thus,
exploratory factor analysis with principle components-varimax rotation helped identify

unique dimensions that were present within certain factors.

Variables Analyzed Using Confirmatory Factor Analysis

Tables 9-14 presents data for the multi-item constructs analyzed using
confirmatory factor analysis. The primary objectives of the confirmatory analysis
included:

0 Identifying the degree to which an item correlates with other items
within its hypothesized factor (internal consistency)

 

133

Identifying the degree to which an item correlates with external factors
and items (external consistency or parallelism)

Determining whether a hypothesized construct or factor is
unidimensional

Determining the reliability of each measurement scale (coefficient
alpha)

Statistical analysis with a confirmatory approach supports identifying which

items warrant final inclusion into a factor. Final factors did not include items that

failed the statistical criteria of internal and external consistency. Also, the analysis

supported the identification of multiple dimensions that may be present within a

construct.

The presence of unidimensional factors is a strict requirement of

measurement scale development.

Tables 9-14 each consist of five sections:

(1)

(2)

(3)

Section A presents the original scale items as they appear in the
measurement instrument. The symbol "(r)" following an item indicates
a reverse coded item.

Section 8 presents the final scale items that satisfied the statistical
criteria of internal scale consistency and parallelism vis-a-vis other
factors. If the statistical analysis indicated the presence of multiple
factor dimensions, Section B details each construct dimension and its
final items.

Section C presents the inter-item correlation matrix for a factor or
dimension of a factor (i.e. the internal consistency matrix). The values
along the top and the side of each matrix correspond to the item
number referenced and detailed in Section A. The values along the
diagonal of the correlation matrix represent each item's communality
(i.e. the correlation between an item and its underlying trait). The
nondiagonal values are correlations between the items included within
the factor. These inter-item correlations should be similar within
sampling error to satisfy the criteria of internal consistency between
items. Section C also presents the reliability of each scale (coefficient
alpha).

134

(4) Section D presents each individual item's correlation with its own factor
as well as its correlation with other selected factors. An item-factor
correlation is simply an individual item correlated against the
combined items that make up a factor. When we compare an item to
another factor we refer to this as the test of external consistency or
parallelism. This evaluation reveals the degree to which items within
a factor correlate with outside factors. An item should have a higher
correlation with its own factor and lower and nonsystematic
correlations with other factors to which it does not belong. Ideally,
items within a factor should demonstrate a somewhat consistent
correlation pattern when compared to the factor which the items
belong (within the range specified by sampling error). Items within a
factor should also demonstrate a somewhat consistent pattern when
correlated to factors which the items do not belong. Evaluating
external consistency verifies that an item does not actually "belong"
within another factor.

(5) Section E lists the items deleted from final consideration within a
factor or dimension of a factor. In almost all cases, deleted items were
not internally consistent with other factor items.

Openness/Accuracy of Communication Although originally hypothesized as

a unidimensional construct, analysis indicated that openness and accuracy are two
distinct dimensions within the team communication construct (Table 9). This means
that a team can have accurate communication while lacking open communication
(and vice-versa). The openness and accuracy dimensions of within-team
communication demonstrated acceptable levels of scale reliability (.90 and .83
respectively) and external consistency. Each dimension was considered as a separate
factor during hypothesis testing.

Team Cohesiveness Team cohesiveness also demonstrated multiple

dimensions (Table 10). Analysis indicated that team cohesiveness has a personal and

professional dimension, each with a high level of scale reliability (.89 and .87

respectively). The difference between the two dimensions, as indicated in Section D

135

of Table 10, is not dramatic. It would be acceptable to combine the two dimensions
into a single dimension of team cohesiveness for hypothesis testing. The separate
dimensions are presented for researchers that may require specifically a personal or
professional focus when evaluating team cohesiveness.

Level of Effort Factor analysis indicated a single dimension of the level of
effort factor existed (Table 11) with a reliability of .90. The scale items also
demonstrated acceptable levels of external consistency with other key factors. Both
external rater and team member reported effort demonstrated similar levels of
consistency.

Team Member Collectivism This scale, originally hypothesized to be
unidimensional, also contained several dimensions (Table 12). The first dimension
assessed a team member's willingness to work with others versus working alone on
organizational tasks. This factor, titled team willingness, consisted of four items and
demonstrated a marginal reliability of .63. The scale was externally consistent with
other key factors.

The second dimension assessed a participant's belief about team member
willingness to sacrifice for the sake of the team. This four item factor, titled team
sacrifice, had a scale reliability of .72 and demonstrated high external consistency with
other factors.

Consideration Leader Behavior The final consideration leader behavior scale,
consisting of 13 items, was a partial version of the original consideration dimension

of the Leadership Behavior Description Questionnaire (LBDQ). The 28 item LBDQ

136

Table 9
Openness/Accuracy of Communication
Confirmatory Factor Analysis

Section A: Original Scale Items

Nv—I

99°99???”

10.

11.
12.

13.
14.
15.

In general, it is difficult to contact other team members (r)

Some individuals on this team provide misleading information to other team
members (r)

Team members openly communicate their ideas with other team members
Team members often fail to communicate information to each other (r)
Members can communicate their feelings freely on this team

It is easy to talk openly with the members of this team

The information I receive from other team members is often incomplete (r)
Communication between the members of this team is very open

It is often necessary for me to go back and check the accuracy of the
information I receive from other team members (r)

I sometimes feel that others do not understand the information they receive
from me (r)

I find it enjoyable to talk to other members of this team

The accuracy of information passed among members of the group could be
improved (r)

There are some significant barriers to communication on this team (r)
Complete and open exchange of team-related information occurs on this team
I can think of a number of times when I received inaccurate information from
other members of this team (r)

Section B: Final Scale Items

Openness of Communication Dimension:

Team members openly communicate their ideas with other team members
Members can communicate their feelings freely on this team

It is easy to talk openly with the members of this team

Communication between the members of this team is very open

I find it enjoyable to talk to other members of this team

There are some significant barriers to communication on this team (r)
Complete and open exchange of team-related information occurs on this team

137

Table 9 (continued)

Accuracy of Communication Dimension:

2. Some individuals on this team provide misleading information to other team
members (r)

4. Team members often fail to communicate information to each other (I)

7. The information I receive from other team members is often incomplete (r)

9. It is often necessary for me to go back and check the accuracy of the

information I receive from other team members (r)

12. The accuracy of information passed among members of the group could be
improved (r)

15. I can think of a number of times when I received inaccurate information from
other members of this team (r)

Section C: Inter-Item Correlation Matrix
Openness of Communication:

3 5 6 8 11 13 14
51

46 44

59 69 75

50 45 67 56

11 49 47 60 60 50

13 55 44 60 57 49 53

14 66 55 68 59 59 66 71

OOO\UIUJ

Coefficient Alpha: .90
Accuracy of Communication:

2 4 7 9 12 15
2 39
4 41 30
7 37 42 53
9 46 37 61 58
12 36 38 47 45 39
15 54 33 56 64 48 60

Coefficient Alpha: .83

138
Table 9 (continued)

Section D: Item-Factor Correlations
Openness of Communication:

3 5 6 8 11 13 14
Openness of Communication 71 66 87 75 71 73 84
Accuracy of Communication 53 36 56 47 49 62 62

Accuracy of Communication:

2 4 7 9 12 15
Accuracy of Communication 62 55 73 76 63 77
Openness of Communication 54 50 46 43 41 46

Section E: Deleted items from final scale(s): 1,10

139

Table 10
Team Cohesiveness
Confirmatory Factor Analysis

Section A: Original Scale Items

1.

9‘93”!"

99°89

11.
12.
13.

14.
15.

I would accept almost any type of assignment in order to keep working with
this team

This team is a close-knit group

The members of this team have a good working relationship together

I feel I am really part of this team

Team members know that each one looks out for the other as well as for
himself

A friendly attitude exists among the team

I do not have much confidence and trust in the other team members (r)
Team members recognize each other's professional talents and expertise
Team members from one department often criticize team members from
other departments (r)

When problems arise, team members perceive them as "mutual" problems that
need to be solved

The members of this team are more like competitors than teammates (r)
Team members help each other to more effectively perform their tasks
When problems arise, team members search for solutions that are agreeable
to each member

Our team is a cohesive group

Destructive conflict among members of our team often occurs (r)

Section B: Final Scale Items

Personal Cohesiveness Dimension:

999’!"

F"

This team is a close-knit group

The members of this team have a good working relationship together

I feel I am really part of this team

Team members know that each one looks out for the other as well as for
himself

A friendly attitude exists among the team

Our team is a cohesive group

140
Table 10 (continued)

Professional Cohesiveness Dimension:

7. I do not have much confidence and trust in the other team members (r)

8. Team members recognize each other's professional talents and expertise

10. When problems arise, team members perceive them as "mutual" problems that
need to be solved

11. The members of this team are more like competitors than teammates (r)

12. Team members help each other to more effectively perform their tasks

13. When problems arise, team members search for solutions that are agreeable

to each member

Section C: Inter-Item Correlation Matrix

Personal Cohesiveness Dimension:

2 3 4 5 6 14
2 48

3 59 72

4 44 66 54
5 59 61 51 53

6 42 68 58 52 54

14 63 63 63 57 62 68

Coefficient Alpha: .89

Professional Cohesiveness Dimension:

7 8 10 11 12 13
7 45
8 56 50
10 46 47 58
11 56 53 67 61
12 46 46 56 48 48
13 44 57 59 56 59 58

Coefficient Alpha: .87

141
Table 10 (continued)

Section D: Item-Factor Correlations
Personal Cohesiveness Dimension:
2 3 4 5 6 14

Personal Cohesiveness 69 85 73 73 74 82
Professional Cohesiveness 53 75 73 62 71 76

Professional Cohesiveness Dimension:

7 810111213

Professional Cohesiveness 67 71 76 78 69 76
Personal Cohesiveness 6O 59 72 66 69 69

Section E: Deleted item(s) from scales: 1,9,15

142

Table 11
Level of Effort
Confinnatory Factor Analysis

Section A: Original Scale Items

1.

PS”

95"

."°.\1

11.
12.

13.
14.

15.

It is clear this team, as a group, is willing to put forth a great deal of effort
to accomplish team assignments

There are many times when it is obvious some of the members of the team
do not feel like pursuing the team's assigned tasks (r)

This team works hard to meet its goals and objectives

Some individuals on this team do not contribute their share towards a
successful outcome (r)

There is virtually no wasted effort by the members of this team

This team works hard to perform its assignments in ways management finds
acceptable

This team can put substantially more effort into its assignments (r)

When you consider the tasks required of each team member, everyone does
his/her fair share of the work

Team members are not concerned enough about this team's assignments (r)
Given a choice between a team assignment and a non-team assignment, some
team members clearly would rather pursue non-team assignments (r)

Team members work as hard as can be expected

The level of effort put forth by this team is lower than what it really should
be to meet management's performance expectations (r)

The lax effort put forth by this team is fiustrating (r)

This team definitely works at a level that meets or exceeds our organization's
performance expectations

I have little doubt this team has the drive to be the best

Section B: Final Scale Items

Level of Team Effort:

1.

PS"

It is clear this team, as a group, is willing to put forth a great deal of effort
to accomplish team assignments

There are many times when it is obvious some of the members of the team
do not feel like pursuing the team's assigned tasks (r)

This team works hard to meet its goals and objectives

Some individuals on this team do not contribute their share towards a
successful outcome (r)

This team works hard to perform its assignments in ways management finds
acceptable

143
Table 11 (continued)

7. This team can put substantially more effort into its assignments (r)

8. When you consider the tasks required of each team member, everyone does
his/her fair share of the work

11. Team members work as hard as can be expected

12. The level of effort put forth by this team is lower than what it really should
be to meet management's performance expectations (r)

Section C: Inter-Item Correlation Matrix
Level of Team Effort:

1 2 3 4 6 7 8 11 12
53 '

44 41

71 45 54

41 48 39 38

43 34 47 31 32

42 39 40 40 40 41

48 49 42 55 36 46 50

11 50 38 51 49 36 52 59 53

12 56 54 60 38 49 52 51 58 60

macs-nuts.)—

Coefficient Alpha: .90

Section D: Item-Factor Correlations
Level of Team Effort:
1 2 3 4 6 7 8 11 12

Effort 73 64 73 62 57 64 71 73 78
Open Communication 64 37 58 41 47 45 48 52 51
Accurate Communication 50 36 41 49 36 51 55 58 52
Personal Cohesiveness 71 38 64 39 53 40 50 48 55
Professional Cohesiveness 71 46 63 46 56 47 54 53 53
Team Willingness 34 18 26 19 24 9 24 29 19
Team Sacrifice 23 -6 21 10 14 0 3 4 12

Section E: Deleted items from scale: 5,9,10,13,14,15

144

Table 12
Team Member Collectivisrn
Confirmatory Factor Analysis

Section A: Original Scale Items

. I prefer to work with others on this team rather than to work alone

. Given the choice, I would rather do a job where I can work alone rather than do
a job where I have to work with others (r)

. I like it when members of my team do things on their own, rather than working
with others all the time (r)

. The members of this team should be willing to make sacrifices for the sake of the
team

. The members of this team should realize that they sometimes are going to have
to make sacrifices for the sake of the team as a whole

. Team members should recognize that they are not always going to get what they
want

. Members should be made aware that if they are going to be part of this team
they are sometimes going to have to do things they don't want to do

. Members of this team should do their best to cooperate with each other instead
of trying to work things out on their own

Section B: Final Scale Items

Team Willingness:

1. I prefer to work with others on this team rather than to work alone
2. Given the choice, I would rather do a job where I can work alone rather than do

a job where I have to work with others (r)

3. I like it when members of my team do things on their own, rather than working

with others all the time (r)

8. Members of this team should do their best to cooperate with each other instead

of trying to work things out on their own

145

Table 12 (continued)

Team Sacrifice:

4.

The members of this team should be willing to make sacrifices for the sake of the
team

. The members of this team should realize that they sometimes are going to have

to make sacrifices for the sake of the team as a whole

Team members should recognize that they are not always going to get what they
want

Members should be made aware that if they are going to be part of this team
they are sometimes going to have to do things they don't want to do

Section C: Inter-Item Correlation Matrix

Team Willingness:

1 2 3 8
1 51
2 48 36
3 31 29 24
8 27 18 22 14

Coefficient Alpha: .63

Team Sacrifice:

4 5 6 7
36

64 59

20 34 28
30 39 50 40

\IONUI-P

Coefficient Alpha: .72

146

Section D: Item-Factor Correlations

Team Willingness:

1 2 3 8
Team Willingness 72 60 48 37
Team Sacrifice 22 14 10 46
Open Communication 36 22 13 11
Accurate Communication 21 8 10 3
Personal Cohesiveness 39 16 18 9
Professional Cohesiveness 42 23 14 14
Level of Effort 35 16 13 8
Team Sacrifice:

4 5 6 7
Team Sacrifice 60 78 52 63
Team Willingness 22 30 32 22
Open Communication 13 12 0 5
Accurate Communication 5 1 2 0
Personal Cohesiveness 23 26 7 13
Professional Cohesiveness 20 20 6 6
Level of Effort 13 10 -1 3

Section E: Deleted items from final scale: None

147

scale included in the measuring instrument (before statistical item analysis) had a .91
reliability, the same as the final 13 item scale (Table 13). This reflects a trade-off
between item quantity (increasing factor items increases reliability) and item quality
(a smaller number of internally consistent, higher quality factor items).

Although this scale has existed for many years, a portion of the consideration
behavior items did not demonstrate internally consistency for this data sample.
There are several possible explanations for this. First, the original LBDQ analysis
relied exclusively on exploratory factor analysis to identify item-factor loadings. A
review of the factor loadings for consideration behavior items indicated that certain
loadings were originally marginal or weak. Secondly, a portion of the original LBDQ
consideration behavior items may not be applicable for the teams sampled during this
research. For example, the item "The leader of our team does personal favors for
the team members" is not as applicable in a setting involving part-time team
assignments compared to a setting involving a full-time supervisor with daily contact
with a structured work crew.

The final consideration behavior scale, which demonstrated strong inter-scale
reliability (coefficient alpha = .91), also demonstrated a high level of external
consistency with the initiating task structure dimension of leadership behavior (the
second dimension of leadership behavior evaluated within this research).
Consideration behavior and initiating task structure are clearly separate dimensions
of leadership behavior. This is consistent with the data findings of the original

LBDQ. The final 13 item consideration behavior scale used in this research is highly

148

Table 13
Consideration Behavior
Confirmatory Factor Analysis

Section A: Original Scale Items

The leader of our team:

PWSPV‘PP’N!‘

15.
16.
17.
18.
19.
20.
21.
22.

24.
26.

27.
28.

Refuses to give in when team members disagree with him or her (r)
Does personal favors for the team members

Expresses appreciation when one of the team members does a good job
Is easy to understand

Demands more than we can do (r)

Helps team members with personal problems

Criticizes team members in front of others (r)

Stands up for team members even if it makes the leader unpopular
Insists that everything he done his or her own way (r)

Sees that a team member is rewarded for a job well done

Rejects suggestions for change (r)

Changes the duties of team members without first talking it over with them
(r)

Treats the team members without considering their feelings (r)

Tries to keep the team members in good standing with those in higher
authority

Resists changing ways of doing things (r)

Rides the team member who makes a mistake (r)

Refuses to explain his or her actions (r)

Acts without consulting the other team members first (r)

Stresses the importance of high morale among the team members
Backs up the team members in their actions

Is slow to accept new ideas (I)

Treats the team members as his or her equal

Criticizes a specific act rather than a particular individual

Is willing to make changes

Makes the team members feel at ease when talking with him or her

Is friendly and can be easily approached

Puts suggestions that are made by team members int operation

Gets the approval of the other team members on important matters before
going ahead

149

Table 13 (continued)

Section B: Final Scale Items

Consideration Behavior:

The leader of our team:

3.
4.

11.
15.
17.
18.
20.
21.
22.
24.
26.
27.
28.

Expresses appreciation when one of the team members does a good job
Is easy to understand

Rejects suggestions for change (r)

Resists changing ways of doing things (r)

Refuses to explain his or her actions (r)

Acts without consulting the other team members first (r)

Backs up the team members in their actions

Is slow to accept new ideas (r)

Treats the team members as his or her equal

Is willing to make changes

Is friendly and can be easily approached

Puts suggestions that are made by team members into operation

Gets the approval of the other team members on important matters before
going ahead

Section C: Inter-Item Correlation Matrix

Consideration Behavior:

3 41115171820212224262728
3 41

4 48 36

11 28 36 38

15 37 32 47 39

17 43 39 57 48 57

18 42 36 46 47 66 50

20 55 39 29 30 38 31 37

21 31 42 53 57 47 49 35 40

22 34 43 39 27 50 39 37 37 34

24 46 44 49 53 50 49 54 42 39 57

26 53 53 37 42 55 515142 55 54 60

27 50 31 34 37 45 42 48 35 30 58 56 42
28 42 37 4142 55 63 43 38 36 57 59 49 50

Coefficient Alpha: .91

150
Table 13 (continued)

Section D: Item-Factor Correlations
Consideration Behavior:
3 41115171820212224262728

Cons. Behavior 64 60 62 62 75 71 61 63 58 75 77 65 71
Initiating Task 53 22 6 15 26 18 49 6 18 35 35 44 31

Section E: Deleted items from final scale: 1,2,5,6,7,8,9,10,12,13,14,l6,l9,23,25

151

Table 14
Initiating Task Structure
Confirmatory Factor Analysis

Section A: Original Scale Items

The leader of our team:

PwflgMPPNZ‘

Encourages overtime work

Tries out his or her new ideas

Controls tightly the team's activities

Criticizes poor work

Talks about how much should be done

Encourages slow-working team members to greater effort

Waits for team members to develop new ideas before he or she does (r)
Assigns team members to particular tasks

Asks for sacrifices from the team members for the good of the entire team
Insists that team members follow standard ways of doing our task

Sees to it that team members work up to their performance potential
Offers new approaches to problems

Insists on being informed about decisions made by the team members
Lets the team members do their work the way they think is best (r)
Stresses being ahead of other organizational teams or groups

Pushes team members for greater effort

Decides in detail what shall be done and how it shall be done
Emphasizes the meeting of deadlines

Emphasizes the quantity of work

Is concerned that we understand our specific team assignments

Section B: Final Scale Items

Initiating Structure:

The leader of our team:

Talks about how much should be done

Encourages slow-working team members to greater effort

Assigns team members to particular tasks

Asks for sacrifices from the team members for the good of the entire team
Sees to it that team members work up to their performance potential
Offers new approaches to problems

Insists on being informed about decisions made by the team members
Pushes team members for greater effort

152
Table 14 (continued)

18. Emphasizes the meeting of deadlines
19. Emphasizes the quantity of work
20. Is concerned that we understand our specific team assignments

Section C: Inter-Item Correlation Matrix
Initiating Structure:

5 6 8 911121316181920
5 28

6 33 48
8 38 42 34

9 30 36 50 34

11 30 54 36 35 41

12 25 44 32 37 59 41

13 29 43 21 33 27 31 26

16 43 52 40 38 39 45 35 48

18 29 36 37 28 29 37 37 4137

19 29 37 28 41 32 33 26 45 46 32
20 38 41 36 29 50 44 36 40 48 32 43

Coefficient Alpha: .87

Section D: Item-Factor Correlations
Initiating Task:

5 6 8 911121316181920
Initiating Task 53 70 59 58 64 64 51 69 60 57 65
Consideration Behavior 6 21 17 15 31 42 15 18 25 11 53

Section E: Deleted items from final scale: 1,2,3,4,7,10,14,15,17

153

reliable and externally consistent with the initiating task structure factor. It is also
more efficient compared to the original 28 item consideration behavior scale.
Initiating Task Structure Leader Behavior The final version of the initiating
task structure scale (Table 14) contained eleven of the original twenty items within
the LBDQ version. The eleven item version, which demonstrated a high level of
internal reliability (.87), also exhibited a high degree of external consistency with the
consideration behavior factor. The reason the final scale for this research did not
include certain original LBDQ items are the same as those discussed for the

consideration behavior dimension.

Variables Analyzed Using Exploratory Factor Analysis

This research also used exploratory factor as the primary tool to identify
multiple dimensions of certain constructs. A lack of a priori dimensional expectations
within a construct supported the use of this approach. For example, the team
effectiveness construct originally contained 28 cross-functional sourcing team
effectiveness related items. Exploratory factor analysis helped identify four team
effectiveness dimensions. This approach strengthened the measurement model by
identifying construct dimensions originally not thought to exist and providing
additional depth to the measurement model.

Besides providing rotated factor loading data, Tables 15 and 16 present the
same information as the confirmatory analysis tables. This includes inter-item

correlation matrices, scale reliability (coefficient alpha) and item-factor correlations.

154

Analysis of the internal and external consistency of scale items further strengthened
measurement model evaluation by providing data not available solely through an
exploratory approach.

Team Effectiveness The team effectiveness construct contained items covering
a variety of team process performance outcomes and specific cross-functional
sourcing team performance indicators. Exploratory factor analysis identified four
dimensions of cross-functional sourcing team effectiveness-~process performance,
innovation introduction, time reduction effectiveness, and supply base performance
and management effectiveness.

The identification of nine items loading highly on a single factor of process
performance was not unexpected (Table 15). It is not unusual for a team to receive
a consistent or near consistent rating across each process performance item. Few (if
any) teams included in this study received detailed quantified assessments or
feedback about team performance at a level which allowed team differentiation
across each process performance scale item. It was therefore difficult for members
and raters to segregate team performance across individual items. For hypothesis
testing, nine process performance items combined to form a single dimension of
overall team process performance.

The process performance dimension of the team effectiveness factor
demonstrated a high level of internal consistency (reliability of .93) and an acceptable
level of external consistency when compared to the other three dimensions of team

effectiveness.

155

Table 15
Team Effectiveness
Exploratory Factor Analysis

Section A: Original Scale Items

S9999’N2‘

10.
11.
12.
13.
14.
15.
16.
17.

18.
19.
20.
21.
22.

23.
24.
25.
26.
27.
28.

The quantity or amount of work produced

The quality or accuracy of work produced

The number of innovative ideas introduced by the team

Reputation for work excellence

Efficiency of team operations

Morale of team personnel

Ability of the team to communicate and coordinate activities across functional
boundaries

Ability of the team to work with individuals and groups outside the team
Ability to meet timing and task schedule targets

Ability to meet executive management's performance expectations
Reduce purchased item cost

Improve purchased item quality

Support reduced product development cycle time for new products
Improve delivery performance from external or internal suppliers
Support the increased use of supplier technical capabilities

Reduce material ordering cycle time between the buyer and supplier
Develop procurement sourcing strategies that directly support business unit
strategies

Support the achievement of new product performance targets

Achieve best-in-class supplier selection

Improve supply base responsiveness

Support inventory reductions

Provide access to new product and process technology before competitors
have access to the technology

Develop supplier performance capabilities

Achieve supply base optimization targets

Support early sourcing and supplier participation during product design
Support standardization of purchased items and/or finished products
Foster development of new technology by suppliers for company use
Establish strategic relationships with suppliers

156
Table 15 (continued)

Section B: Rotated Factor Loadings:

Factor 1 Factor 2 Factor 3 Factor 4

1. .72191

2. .73953

3. .59403

4. .73881

5. .79019

6. .77397

7. .82864

8. .81559

9. .76588

10. .81606

11. .50219
12. .58277
13. .75060

14. .61353

15. .73692
16. .61774

17. .58541
18. .73983
19. .74602
20. .71989
21. .76260

22. .74626
23. .72342
24. .64595
25. .67893
26. .55640

27. .75446
28. .73708

Factor 1 = Process Performance

Factor 2 = Innovation Introduction

Factor 3 = Time Reduction Effectiveness

Factor 4 = Supply Base Management Performance

157

Table 15 (continued)

Section C: Final Scale Items

Process Performance:

$999!"?

t—sxopo

The quantity or amount of work produced

The quality or accuracy of work produced

Reputation for work excellence

Efficiency of team operations

Morale of team personnel

Ability of the team to communicate and coordinate activities across functional
boundaries

Ability of the team to work with individuals and groups outside the team
Ability to meet timing and task schedule targets

Ability to meet executive management's performance expectations

Innovation Introduction:

3.

The number of innovative ideas introduced by the team

Time Reduction Effectiveness:

13.
14.
16.
21.
26.

Support reduced product development cycle time for new products
Improve delivery performance from external or internal suppliers
Reduce material ordering cycle time between the buyer and supplier
Support inventory reductions

Support standardization of purchased items and/or finished products

Supply Base Management Performance:

11.
12.
15.
17.

18.
19.
20.
22.

Reduce purchased item cost

Improve purchased item quality

Support the increased use of supplier technical capabilities

Develop procurement sourcing strategies that directly support business unit
strategies

Support the achievement of new product performance targets

Achieve best-in-class supplier selection

Improve supply base responsiveness

Provide access to new product and process technology before competitors
have access to the technology

158

Table 15 (continued)

23.
24.
25.
27.
28.

Develop supplier performance capabilities

Achieve supply base optimization targets

Support early sourcing and supplier participation during product design
Foster development of new technology by suppliers for company use
Establish strategic relationships with suppliers

Section D: Inter-Item Correlation Matrices

Process Performance:

HQOOQQM-th—b

0

1 2 4 5 6 7 8 9 10
51

75 66

62 72 68

61 61 72 68

46 54 61 67 51

55 63 68 64 68 69

47 59 55 56 52 72 54

47 54 49 60 44 56 59 49

56 63 70 70 56 64 59 72 69

Coefficient Alpha: .93

Innovation Introduction:

One Item Scale

Coefficient Alpha: 1.0

Time Reduction Effectiveness:

13 14 16 21 26

13 54

14 58 58

16 56 67 64

21 58 54 59 58

26 50 50 55 56 48

Coefficient Alpha: .87

Table 15 (continued)

159

Supply Base Management Perfomrance:

11 12 15 17 18 19 20 22 23 24 25 27 28

11 42

12 48 46

15 38 46 42

17 56 40 43 50

18 58 52 40 53 52

19 50 54 49 61 54 57

9 55 49 70 55 39 66 54

20 49 58 59 58 49 74 63

22 45 44 42 44 0 44 48 52

23 41 56 55 56 4 53 65 63 59

24 34 51 52 49 4 49 59 53 61 49
25 46 46 35 43

27 49 45 46 48

28 47 53 56 59

Coefficient Alpha: .93

Section E:

Process Performance:

Process Performance
Innovation Introduction
Time Reduction

Supply Base Management

Innovation Introduction:

Innovation Introduction
Process Performance
Time Reduction

Supply Base Management

6
5
4
53 46 42 61 45 52 47
5
4

4 6170 48 56 6158 54 61

Item-F actor Correlations

1245678910

72 81 83
53 56 61
56 53 5

5

3
5

5 2
646561

8
5
5
6

100
66
53
54

Table 15 (continued)

Time Reduction:

Time Reduction

Process Performance
Innovation Introduction
Supply Base Management

Supply Base Management:

Supply Base Management
Process Performance
Innovation Introduction
Time Reduction

160

13 14 16 21 26

74 76 80 76 69
48 56 48 45 50
41 39 38 39 43
59 76 71 68 69

11 12 15 17 18 19 20 22 23 24 25 27 28

64 68 65 71 72 76 80 72 77 70 68 74 78
57 57 44 54 52 53 58 48 53 55 57 52 63
48 39 32 47 38 35 33 44 37 33 36 43 40
62 65 61 70 67 63 71 66 66 63 65 67 65

161

The identification of time reduction as a separate dimension of team
effectiveness is critical to the type of team (i.e. product and commodity sourcing
teams) included in this research. In separate research, Monczka, Trent, and Gay
(1992) noted that time reduction abilities have become the highest rated capabilities
required for competitive success on a worldwide basis by 1995. Firms achieve time
reduction in multiple ways. For example, the effectiveness scale item "support
inventory reductions" related directly to time reduction effectiveness because
inventory reductions often involve a time-focused just-in-time purchasing system with
suppliers. Additionally, a team's ability to "support standardization of purchased
items and/or finished products" and "improve delivery performance from external or
internal suppliers" are also recognized ways to support improved time-related
performance. The time reduction items logically belong within that dimension. Time
reduction effectiveness is a critical performance outcome sought from many of the
teams included in this study. Time reduction is a primary motivator for establishing
cross-functional teams within complex organizations today.

The time reduction dimension of team effectiveness demonstrated a high level
of scale reliability (.87). It also demonstrated a high level of external consistency
with the process performance and innovation introduction dimensions. It
demonstrated acceptable levels of differentiation with the supply base performance
and management dimension.

Innovation Introduction is a third dimension of team effectiveness consisting

of a single item. The need for organizational innovation, like time reduction

162

effectiveness, is another primary motivator for establishing cross-functional sourcing
teams. This dimension is valuable because it offered the opportunity to compare
those factors which drive higher levels of innovation introduction versus other
dimensions of team effectiveness. For example, are the variables that related highly
to time reduction effectiveness the same as those that related to innovation
introduction?

A fourth dimension of team effectiveness is a composite of 13 items relating
directly to supply base management effectiveness. This dimension included items
unique to the teams in this study and have minimal application outside a population
of cross-functional sourcing teams. Each item is a unique cross-functional sourcing
team performance indicator. This scale demonstrated a high degree of reliability
(coefficient alpha .93) and acceptable to high levels of external consistency.

Performance Evaluation and Reward System Analysis indicated that the
original 15 items included in the performance evaluation and rewards construct relate
to three factor dimensions (Table 16). One three-item factor demonstrated poor
reliability and inconsistent item content and was not included as part of the final data
analysis.

The first dimension related to the presence or absence of a performance
appraisal and evaluation system that motivated or encouraged team member
participation on team assignments. This dimension, titled system motivation to

participate, consisted of seven items with high reliability (.87). The scale also

163

Table 16
Performance Evaluation and Rewards
Exploratory Factor Analysis

Section A: Original Scale Items

The performance appraisal system used to evaluate and/or reward my professional
work:

1 Does not acknowledge or evaluate my performance on this team (r)

2. Encourages a total commitment to the team's assignment

3 Relies too much in subjective indicators of individual or team performance
(r)

4. Relies on performance assessments by manager(s) outside the team who are
not really familiar with our individual or team effort

5. Provides rewards, incentives or bonuses for superior individual team member
effort and contribution

6. Provides rewards, incentives or bonuses for superior total team effort and
performance

7. Emphasizes the right combination of individual and team performance
measures

8. Provides no salary increases, bonuses or promotions for devoting a lot of my
time and resources to this team's assignments (r)

9. Strictly stresses individual team member effort or performance over the team's
combined performance (r)

10. Provides the right weight or emphasis on the team's assignments as compared
to non-team assignments

11. Stresses my personal participation on this team very much

12. Is out of date with the realities of the cross-functional sourcing team
organizational structure (r)

13. Penalizes a lax individual effort on team assignments

14. Is actually a barrier to participating actively on this cross-functional sourcing
team (r)

15. Encourages individual and team effort at a level that supports meeting our

team goals and objectives

164
Table 16 (continued)

Section B: Rotated Factor Loadings:

Factor 1 Factor 2 Factor 3
1. .70140
2. .65127
3. .69836
4. .52841
5. .81569
6. .82178
7. .68807
8. .71548
9. .73987
10. .66277
11. .78700
12. .60228
13. .47016*
14. .56289*
15. .7342]
Factor 1 = System Motivation to Participate
Factor 2 = Participation Rewards
Factor 3 = Note: This factor dropped due to inconsistent item content and

poor internal consistency
* Items 13 and 14 dropped from final scale due to poor internal consistency
identified within the inter-item correlation matrix
Section C: Final Scale Iterrrs
System Motivation to Participate:

The performance appraisal system:

1 Does not acknowledge or evaluate my performance on this team (r)

2. Encourages a total commitment to the team's assignment

7 Emphasizes the right combination of individual and team performance
measures

10. Provides the right weight or emphasis on the team's assignments as compared

to non-team assignments
11. Stresses my personal participation on this team very much

165
Table 16 (continued)

12. Is out of date with the realities of the cross-functional sourcing team
organizational structure (r)
15. Encourages individual and team effort at a level that supports meeting our

team goals and objectives
Participation Rewards:

The performance appraisal system:

5. Provides rewards, incentives or bonuses for superior individual team member
effort and contribution

6. Provides rewards, incentives or bonuses for superior total team effort and
performance

8. Provides no salary increases, bonuses or promotions for devoting a lot of my

time and resources to this team's assignments (r)

Section D: Inter-Item Correlation Matrices

System Motivation to Participate:

l 2 7 10 11 12 15
1 53
2 54 47
7 51 47 58
10 49 44 59 50
11 61 47 48 44 43
12 42 44 48 48 38 39
15 51 55 66 54 44 51 59

Coefficient Alpha: .87

Participation Rewards:

5 6 8
5 56
6 65 73
8 45 54 40

Coefficient Alpha: .78

166
Table 16 (continued)

Section E: Item-Factor Correlations

System Motivation to Participate

1.2 710111215
System Motivation to Participate 73 68 76 70 66 63 77
Participation Rewards 51 53 56 49 41 41 51

Participation Rewards

5 6 8
Participation Rewards 54 86 62
System Motivation to Participate 43 63 49

Deleted items: 3,4,9,13,14

167

differentiated itself well from the second dimension of performance evaluation and
rewards.

The second dimension of performance evaluation and rewards assessed the
presence or absence of rewards, incentives, or bonuses for superior individual and
team performance. This scale, consisting of three items, has acceptable reliability
(.78) and external consistency when compared to the system motivation to participate
dimension.

Other Research Variables Tables 17-19 present the results of exploratory
factor analysis for three additional research variables. These variables include (1)
leadership resource/commitment attainment, (2) supplier involvement, and (3) cross-
functional sourcing team authority. Exploratory analysis indicated the nine items
included in the resource/commitment attainment scale related to a single factor with
a reliability of .95. This factor, which represents the ability of a leader to secure
resources and team member commitment through the performance leadership-related
tasks, is a third leadership variable. This is in addition to the task-oriented and
relationship-oriented dimensions of the Leadership Behavior Description
Questionnaire.

Exploratory analysis indicated that multiple dimensions of supplier
involvement existed (Table 18). The first dimension included items relating directly
to supplier performance support. Suppliers who provide performance support (1)
provide cost reduction ideas, (2) provide quality improvement ideas, (3) support

actions to improve material delivery, (4) offer process technology ideas, and (5)

168

Table 17
Leadership Resource/Commitment Attainment
Exploratory Factor Analysis

Section A: Original Scale Items

Securing individual member involvement

Managing internal conflict

Maintaining team focus and direction

Securing resources as required

Preventing team domination by a member(s) or function(s)

Dealing with obstacles confronting the team

Coordinating multiple tasks and managing the status of team assignments
Establishing specific team goals

Clarifying and/or defining each member's role

99°99‘99pr

Section B: Factor Matrix

Factor 1

.84861
.80461
.87553
.80041
.80950
.86837
.82498
.83405
.83011

999999959!"

Section C: Final Scale Items

Leadership Resource/Commitment Attainment is a single dimensional variable which
includes all of the above nine items combined into a liner combination factor score.

Coefficient alpha: .95

169

Table 18
Supplier Involvement
Exploratory Factor Analysis

Section A: Original Scale Items
Supplier Support Category:

Provide cost reduction ideas

Provide quality improvement ideas

Supports actions to improve material delivery

Offer new product ideas

Offer process technology ideas

Formally meet with the team during team activities
Perform product design activities

Support new product development cycle time reductions
Support material ordering cycle time reductions

999999999?

Section B: Factor Matrix

Factor 1 Factor 2

1. .84465

2. .82745

3. .94198

4. .74355
5. .76546

6. .48331* 51671“
7. .89413
8. .62774
9. .80639

Factor 1 = Performance Support

Factor 2 = New Product Support

* Item 6 not included in either factor due to inconsistent item content

170
Table 18 (continued)

Section C: Final Scale Items
Supplier Performance Support:

Provide cost reduction ideas

Provide quality improvement ideas

Supports actions to improve material delivery
Offer process technology ideas

Support material ordering cycle time reductions

99993!”

Coefficient alpha: .84

Supplier New Product Support:

4. Offer new product ideas

7. Perform product design activities

8. Support new product development cycle time reductions

Coefficient alpha: .70

171

Table 19
Team Authority
Exploratory Factor Analysis

Section A: Original Scale Items
Team authority to:

Make supplier selection decisions

Select the team leader

Allocate budget and material resources to support team activities
Determine the team's performance goals and objectives

Make decisions that affect product design such as technology selection, design
parameter establishment, quality requirements, etc.

6 Make decisions that affect more than one function of department

7. Schedule team meetings

8

9

MPPN!‘

Make timing decisions for the completion of team activities and assignments
Solicit non-team members to support team assignments

10. Select new team members as required

11. Establish purchase prices

12. Develop sourcing and negotiating strategies

Section B: Factor Matrix

Factor 1 Factor 2 Factor 3 Factor 4

1. .84607

2. .74893
3. .66674 .
4. .73077

5. .56698

6. .48614

7. .87187
8. .63474

9. .72266

10. .63137
11. .80962

12. .91502

Factor 1 = External Decision Authority

Factor 2 = Internal Team Process Authority
Factor 3 = Member/Leader Selection Authority
Factor 4 = Scheduling Authority

172
Table 19 (continued)

Section C: Final Scale Items
External Decision Authority:

1. Make supplier selection decisions

6. Make decisions that affect more than one function of department
11. Establish purchase prices

12. Develop sourcing and negotiating strategies

Coefficient alpha: .87
Internal Team Process Authority:

Allocate budget and material resources to support team activities
Determine the team's performance goals and objectives

Make timing decisions for the completion of team activities and assignments
Solicit non-team members to support team assignments

wwsw

Note: Item 5 not included in Factor 2 due to inconsistent item content
Coefficient alpha: .71

Member/Leader Selection Authority:

2. Select the team leader
10. Select new team members as required

Coefficient alpha: .54
Scheduling Authority

7. Schedule team meetings

173

support material ordering cycle time reduction. The performance support dimension
had a reliability of .84.

The second dimension of supplier involvement related to new product
support. This three item scale examined a supplier's contribution towards (1)
offering new product ideas, (2) performing product design activities, and (3)
supporting new product development cycle time reduction. This scale had an
acceptable reliability of .70. A third dimension included a single item examining the
level of support suppliers provided during formal team meetings.

An exploratory factor analysis of the team authority variable indicated the
existence of four unique dimensions. These included (1) scheduling authority, (2)
member/leader selection authority, (3) internal team process authority, and (4)
external decision authority. Table 19 details the specific items and dimensions
comprising team authority. Internal and external authority dimensions demonstrated
acceptable reliability while the member/leader selection dimension demonstrated

marginal reliability. The scheduling authority dimension consisted of a single item.

HYPOTHESIS TESTING RESULTS

This section presents the results of testing each a priori research hypothesis.
Data analysis used hierarchical regression to identify the unique contribution of each
independent variable in relation to the total variance accounted for by the combined
variables within each regression equation. Hierarchical analysis is an appropriate

approach because the story told by a single simultaneous analysis for multiple

174

independent variables may, for many purposes, be incomplete. Also, it is the only
basis on which variance partitioning can proceed when confronted with the possibility
of correlated independent variables (Cohen and Cohen 1983).

The ordering of independent variables was critical to the use of hierarchical
regression analysis. Order of data entry was critical because the incremental variance
attributable to an independent variable may change considerably if one changes its
position within the hierarchy. Changing the order of independent variables modifies
what has and has not been partialled from the total pool of variance (Cohen and
Cohen 1983).

The sequencing of variables within each research hypotheses followed an
order of presumed causal priority and/or temporal precedence.l The first group of
variables entering the analysis consisted of demographically related variables. This
included, for example, professional tenure, the functional diversity of the team, age
distribution, and previous cross-functional team experience. These always preceded
the variables that resulted from or evolved throughout the team interaction process.

The second group included variables that were present as a result of team
interaction. This included team leadership, organizational resources, and team effort.
This was also the general sequence of introduction for these variables within the

hierarchical analysis. Leadership preceded organizational resources and effort

 

The ordering of the variables in hierarchical analysis is decision made by the
researcher. It is important to recognize that the ordering of independent variables
affects the results of the study. A different order will result in a different partialling
of variance and hence produce different results. It is the responsibility of the
researcher to defend or justify his or her ordering of independent variables.

175

because of evidence that an effective leader can influence effort and resources.
Organizational resources preceded effort because of a presumption that resources
can affect a team's level of effort.

The final group included self-reported perceptual variables which logically or
temporally followed the second group. This included, for example, team
communication and cohesiveness. There was a significant relationship within this
study between team cohesiveness and the length of time a cross-functional sourcing
team has worked together. This naturally placed cohesiveness towards the latter part
of the hierarchical analysis. Also, openness and accuracy of team communication
naturally followed leadership and team effort.

This section presents two sets of results whenever a research hypothesis
includes a team-reported and externally-reported dependent measure. This situation
involves the team effort and cross-functional sourcing team effectiveness hypotheses.
Also, the chapter reports data about each dimension of an independent variable
where applicable. For example, team effectiveness included four separate
dimensions--process performance, innovation performance, time reduction
performance, and supply base management performance. The objective was not to
present a vast amount of data but rather to develop a complete understanding of
cross-functional sourcing team effectiveness.

This analysis presents the major findings associated with the testing of each
a priori research hypothesis. However, minimal discussion of the research findings

and conclusions occurs. Chapter V presents a detailed discussion of critical research

176

findings. Chapter V also discusses findings about variables originally hypothesized
as critical but found not to relate to cross-functional sourcing team effectiveness.

Tables 20-35 present the results of testing each hypothesized linkage discussed
in Chapter 111. Each data table identifies the dependent variable and the order of
entry for independent variables analyzed through hierarchial regression analysis.
Within the data tables, significant research findings are highlighted and italicized.
Furthermore, the specific hypothesis that each independent variable referred to in
Chapter III is identified in parentheses. This allows the reader to readily identify
which hypotheses are supported by the data. Data provided in each table include a
cumulative total of the amount of variance the independent variables explain, the
unique variance contribution of a particular independent variable, the beta weight
associated with an independent variable, and the significance level associated with
that beta weight.

Hypothesis 1 This hypothesis originally proposed that team effort related
directly to (1) a system that rewarded team participation, (2) the collectivism of team
members towards the cross-functional sourcing team process, (3) task-oriented
leadership, and (4) team cohesiveness. The hypothesis also proposed that
professional tenure related negatively to team effort.

The team-reported and externally-reported models agreed that only task-
oriented leadership and team cohesiveness related to team effort (Tables 20 and 21).

Furthermore, the two models agreed that professional tenure, team collectivism, and

177

Table 20
Hypothesis 1 Results
Hierarchical Regression Analysis
Team-Reported Level of Effort
Dependent Variable: Team-Reported Level of Effort

Cumulative SR

Variable: R Square: Square: Beta: Sig.:

Professional Tenure (1-E) .00539 .00539 .07 .45

Team Willingness (l-D) .01512 .00973 .10 .31

Participation Rewards (l-A) .01531 .00019 -.01 .88

Initiating Task Structure (I-B) .12996 .11465 .35 .00

Team Cohesiveness (I-C) .64500 .51504 .80 .00
Table 21

Hypothesis 1 Results
Hierarchical Regression Analysis
Externally-Reported Level of Effort
Dependent Variable: Externally-Reported Level of Effort

Cumulative SR

Variable: R Square: Square: Beta: Sig.:
Professional Tenure (1-E) .00211 .00211 .05 .64
Team Willingness (1-D) .00351 .00140 -.04 .71
Participation Rewards (LA) .01205 .00854 .09 .35
Initiating Task Structure (I-B) .15345 .14140 .39 .00
Team Cohesiveness (I-C) .21075 .05730 .27 .01

178

participation rewards accounted for a negligible amount of the variance explained by
the independent variables.

Both models agreed that cohesiveness related directly to team effort. The
externally-reported model, however, indicated that cohesiveness was much less of a
predictor of team effort than the team-reported model. This difference likely results
from comparing a self-reported measure of cohesiveness to a self-reported measure
of team effort. Teams which indicated effort was high also indicated that
cohesiveness was comparably high. A strong bivariate correlation existed between
team cohesiveness and team-reported effort.

The results of this hypothesis indicated that several variables related directly
to team effort. The final research results minimized, however, the importance of
team cohesiveness as a predictor of team effectiveness and effort. The presence of
variables critical and unique to the cross-functional sourcing process provided greater
insight into team success than self-reported cohesiveness.

Hypothesis 2 This hypothesis proposed that personal and professional team
cohesiveness related directly to (1) a team whose members indicated a favorable
attitude towards team interaction (i.e. team collectivism) and (2) a leader that
demonstrated considerate or relationship-oriented behavior.

Tables 22 and 23 reveal that a key predictor of personal and professional team
cohesiveness was relationship-oriented (i.e. consideration) leadership behavior and
a team's cohesiveness. A review of bivariate correlations, however, revealed that

task-oriented leadership and effective leadership also related directly to team

 

Depel

Van;

lean
Cons

Dq

Va]

C01

179

Table 22
Hypothesis 2 Results
Hierarchical Regression Analysis
Personal Cohesiveness

Dependent Variable: Personal Team Cohesiveness

Cumulative SR

Variable: R Square: Square: Beta:

Team Collectivism (2-B) .02210 .02210 .15

Consideration Leadership (2-A) .29748 .27538 .53
Table 23

Hypothesis 2 Results
Hierarchical Regression Analysis
Professional Cohesiveness

Dependent Variable: Professional Team Cohesiveness

Cumulative SR
Variable: R Square: Square: Beta:

Team Collectivism (2-B) .01876 .01876 .14
Consideration Leadership (2-A) .40999 .39123 .63

Sig.:

.12
.00

Sig.:

.15

 

180

cohesiveness. Post hoc analysis revealed that cohesiveness also related strongly to
other self-reported measures such as effort and the openness and accuracy of team
communication.

The level of multicollinearity between team cohesiveness and other self-
reported variables was large (.6 and higher). This indicated that a response pattern
probably occurred between the self-reported variables of cohesiveness, effort, and
communication. The results of Hypothesis 6 indicated that the importance of self-
reported cohesiveness, as well as other uncorroborated self-reported variables, was
relatively minor when these variables assumed their logical position within the
hierarchical regression analysis.

Hypothesis 3 This hypothesis examined a series of linkages predicting the
openness and accuracy of team communication. Specifically, the hypothesis proposed
that open and accurate team communication was a direct function of team
cohesiveness and task-oriented leadership. Hypothesis results indicated that task-
oriented leadership and team cohesiveness related directly to the openness and
accuracy of team communication (T ables 24 and 25). Again, please recall the
warning about the unusually high correlations between self-reported research
variables, including the correlation between cohesiveness and he openness/accuracy
of team communication.

It was not unexpected that task-oriented leadership related directly to

openness and accuracy of team communication. Additional analysis revealed,

181

however, that relationship-oriented leadership and effective leadership related to
team communication at a much higher level than task-oriented leadership.

Hypothesis 4 This hypothesis proposed that within-team communication
frequency related positively to task-oriented leadership and team cohesiveness. It
also proposed that within-team communication related negatively to within-team age
dispersion and a team's level of functional diversity.

Table 26 indicates that within-team age dispersion and functional diversity
were significant predictors of within-team communication frequency. Task-oriented
leadership and team cohesiveness demonstrated no systematic relationship with
within-team communication frequency. The minimal amount of variance explained
by the four independent variables indicated that other factors should help explain
within-team communication frequency.

The data supported the predicted relationship between age dispersion and
within-team communication frequency. Teams with greater age dispersion, as
measured by the within-team standard deviation of team member ages,
communicated less frequently compared to a team whose members were closer in
age.

The direction of the relationship between functional diversity and within-team
communication frequency was opposite the original prediction. The statistical
significance of functional diversity is not due to team size. While larger teams often
had higher levels of functional diversity, no statistical relationship existed between

team size and within-team communication frequency.

182

Table 24
Hypothesis 3 Results
Hierarchical Regression Analysis
Accuracy of Team Communication

Dependent Variable: Accuracy of Team Communication

Cumulative SR

Variable: R Square: Square: Beta:

Initiating Task Structure (3-B) .0545 7 .0545 7 .23

Team Cohesiveness (3-A) .4833] .42874 . 71
Table 25

Hypothesis 3 Results
Hierarchical Regression Analysis
Openness of Team Communication
Dependent Variable: Openness of Team Communication

Cumulative SR

Variable: R Square: Square: Beta:

Initiating Task Structure (3-B) .14980 .14980 .39
Team Cohesiveness (3-A) .63979 .48999 .76

Sig.:

.01

Sig.:

2'8

183

Table 26
Hypothesis 4 Results
Hierarchical Regression Analysis
Within-Team Frequency of Communication

Dependent Variable: Within-Team Frequency of Communication

Cumulative SR

Variable: R Square: Square: Beta:
Age Dispersion (4-C) .06676 .06676 -.26
Functional Diversity (4-D) .15953 .09277 .31
Initiating Task Structure (4-A) .16886 .00953 -.10

Team Cohesiveness (4—B) .19365 .03412 .18

Sig.:

.01

.30

184

The original hypothesis proposed that cross-functional communication barriers
still exist that impede within-team communication. Greater functional diversity
lowers the probability that team members will be from the same functional discipline
all else equal. The original hypothesis, which proposed that members would not
overcome the barriers to cross-functional communication, particularly as functional
diversity increased, was clearly not supported.

The minimal variance explained by the two significant independent variables
indicated that additional variables probably exist that relate directly to
communication frequency.

Hypothesis 5 This hypothesis proposed that extemal-team communication
frequency was a function of a team's functional diversity and task-oriented leadership.
Hypothesis results revealed that only functional diversity related to external-team
communication frequency (Table 27). Task-oriented leadership demonstrated no
relationship with external-team communication frequency. Furthermore, functional
diversity explained only a minor proportion of the total variance associated with
extemal-team communication.

A major factor affecting a team's communication requirement is the team's
task. Differences among each team's task requirement and the intensity of those
requirements confounded our ability to identify specific predictors of communication
frequency or relate communication frequency to team effectiveness. Unfortunately,
segmenting teams based on task requirement or intensity was not possible within the

framework of this study.

185

Table 27
Hypothesis 5 Results
Hierarchical Regression Analysis
External-Team Frequency of Communication

Dependent Variable: ExtemaI-Team Frequency of Communication

Cumulative SR
Variable: R Square: Square: Beta: Sig.:

Functional Diversity (5-B) .07761 .07761 .28 .00
Initiating Task Structure (S-A) .07765 .00004 -.01 .95

186

Hypothesis 6 This hypothesis evaluated team-reported and extemally-
reported predictors of cross-functional sourcing team effectiveness. The team
effectiveness variable consisted of four separate performance dimensions.
Furthermore, each dimension included a self-reported and externally-reported
analysis of team effectiveness. This resulted in eight separate tables providing data
about the level of significance of each independent variable.

The findings in Tables 28-35 are generally consistent across each effectiveness
dimension and between externally-reported and team-reported effectiveness. While
this represents a large amount of data, Tables 28-35 report basically similar
conclusions. Team-reported and externally-reported data sources indicated that
generally the same set of predictors were critical to overall cross-functional sourcing
team effectiveness.

Minimal differentiation occurred between the separate dimensions of cross-
functional sourcing team effectiveness. For example, major findings that related to
innovation performance also related to team process performance. The original
research intention was to gain insight into variables unique to each effectiveness
dimension. The results of this study indicated that major differentiation across the
separate effectiveness dimensions generally did not occur.

A major finding of a priori data analysis was that external raters and team
members indicated that certain variables related consistently to cross-functional
sourcing team effectiveness. These variables included team leadership, organizational

resources, and team effort. Furthermore, both data sources agreed that

187
Table 28

Hypothesis 6 Results

Hierarchical Regression Analysis
Team-Reported Process Performance

Dependent Variable:

Variable:

Functional Diversity (6-I)
Cross-Functional

Experience (6-J)
Professional Tenure (6-K)
Consideration Leadership (6-C)
Initiating Task Structure (6-D)
Organizational Resources (6-B)
Team Effort (6-A)
Extemal-Team Frequency

of Communication (6-G)
Within-Team Frequency

of Communication (6-F)
Open Communication (645)
Accurate Communication (6-E)
Team Cohesiveness (6-H)

Cumulative SR

R Square:

.00071

.01275
.01928

.3404]
.3 7505
.4215 7
.64739

.64832

.64845

.6 7044

.67630
.67797

Square:
.00071

.01204
.00653
.32113
.03464
.04652
.22582

.00093

.00013
.02199
.00586
.00167

Team-Reported Process Performance

Beta:

.03

.11
.09
.57
.22
.22
.59

-.03

.02
.23
-.12
.10

Sig.:
.79

.27
.42

.02

.00

.00

.62

.85
.0]
.20
.49

188

Table 29
Hypothesis 6 Results
Hierarchical Regression Analysis
Externally-Reported Process Performance
Dependent Variable: Externally-Reported Process Performance

Cumulative SR

Variable: R Square: Square: Beta: Sig.:
Functional Diversity (6-I) .00380 .00380 .06 .53
Cross-Functional

Experience (6-J) .02251 .01871 .14 .17
Professional Tenure (6-K) .0586] .03610 .20 .05
Consideration Leadership (6-C) .08890 .03029 .18 .07
Initiating Task Structure (6-D) .14357 .05467 .27 .0]
Organizational Resources (6-B) .20873 .06516 .26 .0]
Team Effort (6-A) .61427 .40554 .72 .00
External-Team Frequency

of Communication (6-G) .62149 .00722 -.09 .18
Within-Team Frequency

of Communication (6-F) .62735 .00586 -.11 .23
Open Communication (6-E) .64534 .01799 .18 .03
Accurate Communication (6-E) .6670] .02167 -.24 .02
Team Cohesiveness (6-H) .66780 .00079 -.06 .64

 

189

Table 30
Hypothesis 6 Results
Hierarchical Regression Analysis
Team-Reported Innovation Performance
Dependent Variable: Team-Reported Innovation Performance

Cumulative SR

Variable: R Square: Square: Beta: Sig.:
Functional Diversity (6-1) .00192 .00192 .04 .66
Cross-Functional

Experience (6-J) .00192 .00000 .00 .98
Professional Tenure (6-K) .02235 .02043 .15 .15
Consideration Leadership (6-C) .17770 .15535 .40 .00
Initiating Task Structure (6-D) .20402 .02632 .19 .07
Organizational Resources (6-B) .24213 .0381] .20 .03
Team Efl‘ort (6-A) .34463 .10250 .40 .00
Extemal-Team Frequency

of Communication (6-G) .34504 .00041 -.02 .81
Within-Team Frequency

of Communication (6-F) .34529 .00025 .02 .85
Open Communication (6-E) .35029 .00500 .11 .40
Accurate Communication (6-E) .39932 .04903 -.36 .0]
Team Cohesiveness (647) .4259] .02659 .38 .04

190
Table 31

Hypothesis 6 Results

Hierarchical Regression Analysis

Externally-Reported Innovation Perfomrance

Dependent Variable:

Variable:

Functional Diversity (6-I)
Cross-Functional

Experience (6-J)
Professional Tenure (6-K)
Consideration Leadership (6-C)
Initiating Task Structure (6-D)
Organizational Resources (6-B)
Team Efl‘ort (6-A)
Extemal-Team Frequency

of Communication (6-G)
Within-Team Frequency

of Communication (6-F)
Open Communication (6-E)
Accurate Communication (6—E)
Team Cohesiveness (6-H)

Cumulative SR

R Square:

.00050

.00101
.04502
.04550
.06852
.07777
.2640]

.26529

.27866
.27885
.28321
.28341

Square:
.00050

.0005 1
.0440]
.00048
.02302
.00925
.18624

.00128

.01337
.00019
.00435
.00020

Externally-Reported Innovation Perfomrance

Beta:

-.02

.02
.23
-.02
.17
.11
.48

-.04

-.18
-.02
-.11
-.03

Sig:
.82

.82
.03
.82
.12
.32
.00

.68

.18
.87
.45
.87

191
Table 32

Hypothesis 6 Results

Hierarchical Regression Analysis

Team-Reported Time Reduction Performance

Dependent Variable:

Variable:

Functional Diversity (6-I)
Cross-Functional

Experience (6-J)
Professional Tenure (6-K)
Consideration Leadership (6-C)
Initiating Task Structure (6-D)
Organizational Resources (6-B)
Team Efl'ort (6-A)
Extemal-Team Frequency

of Communication (6-G)
Within-Team Frequency

of Communication (6-F)
Open Communication (6-E)
Accurate Communication (6-E)
Team Cohesiveness (6-H)

Cumulative SR

R Square:

.01876

.03678
.04108

.19130

.19169

.29890
.33040

.33125

.34002
.35547
.35969
.36013

Square:
.01876

.01802
.00430
.15022
.00039
.1072]
.03150

.00085

.00877
.01545
.00422
.00044

Team-Reported Time Reduction Performance

Beta:

-.14

.13
-.07
.39
.02
.33
.22
-.03

-.14
.20
-.11
.05

Sig.:

.17

E23823

.73

.27
.14

.80

192

Table 33
Hypothesis 6 Results
Hierarchical Regression Analysis
Externally-Reported Time Reduction Performance
Dependent Variable: Externally-Reported Time Reduction Performance

Cumulative SR

Variable: R Square: Square: Beta: Sig.:
Functional Diversity (6-I) .00289 .00289 -.05 .59
Cross-Functional

Experience (6-J) .06223 .05934 .24 .0]
Professional Tenure (6-K) .07883 .01660 .14 .18
Consideration Leadership (6-C) .11547 .03664 .19 .04
Initiating Task Structure (6-D) .11547 .00000 .00 .99
Organizational Resources (6-B) .22163 .10616 .39 .00
Team Efl'ort (6-A) .35314 .1315] .40 .00
Extemal-Team Frequency

of Communication (6-G) .35624 .00310 -.06 .50
Within-Team Frequency

of Communication (6-F) .36419 0.0795 -.14 .28
Open Communication (6-E) .37450 .01031 .14 .22
Accurate Communication (6-E) .37451 .00001 .00 .97
Team Cohesiveness (6-H) .37738 .00287 .11 .52

193

Table 34
Hypothesis 6 Results
Hierarchical Regression Analysis
Team-Reported Supply Base Management Performance

Dependent Variable: Team-Reported Supply Base Management Performance

Cumulative SR

Variable: R Square: Square: Beta: Sig.:
Functional Diversity (6-1) .01293 .01293 -.11 .25
Cross-Functional

Experience (6-J) .01984 .00691 .08 .40
Professional Tenure (6-K) .02117 .00133 -.04 .71
Consideration Leadership (6-C) .22952 .20835 .46 .00
Initiating Task Structure (6-D) .23742 .00790 '.10 .31
Organizational Resources (6-B) .4002] .16279 .4] .00
Team Efi'ort (6-A) .48697 .08676 .37 .00
Extemal-Team Frequency

of Communication (6-G) .48784 .00087 -.03 .69
Within-Team Frequency

of Communication (6-F) .49455 .00671 —.12 .27
Open Communication (6-E) . .49793 .00338 .09 .43
Accurate Communication (6-E) .50277 .00484 -.11 .35
Team Cohesiveness (6-H) .50806 .00589 -.17 .32

194

Table 35
Hypothesis 6 Results
Hierarchical Regression Analysis
Externally-Reported Supply Base Management Performance

Dependent Variable: Externally-Reported Supply Base Management Performance

Cumulative SR

Variable: R Square: Square: Beta: Sig.:
Functional Diversity (6-I) .00064 .00064 -.03 .80
Cross-Functional

Experience (6-J) .02253 .02189 .15 .14
Professional Tenure (6-K) .05338 .03085 .19 .08
Consideration Leadership (6-C) .07568 .02230 .15 .13 .
Initiating Task Structure (6-D) .08125 .00557 .09 .44
Organizational Resources (6-B) .29979 .21854 .47 .00
Team Efl'ort (6-A) .3603] .06052 .28 .00
External-Team Frequency

of Communication (6-G) .36763 .00732 -.09 .30
Within-Team Frequency

of Communication (6-F) .37168 .00405 -.10 .44
Open Communication (6-E) .39672 .02504 .2] .05
Accurate Communication (6-E) .39736 .00064 -.04 .76
Team Cohesiveness (6-H) .40472 .00736 -.18 .35

195

communication and cohesiveness explained a minimal amount of variance associated
with each dependent variable. Both data sources also agreed that within-team and
external-team communication frequency demonstrated no systematic relationship with
team effectiveness. Part of this resulted from the logical placement of these variables
later within the hierarchical regression analysis. Chapter V provides additional
insight and discussion concerning the communication and cohesive variables.

External raters and team members consistently agreed that no relationship
existed between functional diversity and cross-functional sourcing team effectiveness.
Except for a single significant relationship, both data sources also agreed that
previous cross-functional experience did not relate to team effectiveness. The single
significant relationship between cross-functional experience and externally-reported
time reduction performance could simply have occurred by chance.

External raters indicated that professional tenure demonstrated some
relationship with process and innovation performance. Team-reported performance
data failed to substantiate this finding. Furthermore, the amount of total variance
explained by professional tenure was relatively small. Thus, professional tenure was

not considered a major predictor of cross-functional sourcing team effectiveness.

POST HOC DATA ANALYSIS
Post hoc data analysis occurred because of a recognition that the development
of the original model and a priori hypotheses appeared not to fully explain the

variance associated with each dependent variable. In other words, there existed an

196

opportunity to account for variance beyond that explained by the a prior model by
evaluating additional independent variables.

Post hoc analysis used a combination of canonical correlation and regression
analysis and hierarchical multiple regression analysis to identify additional variables
relating significantly to cross-functional sourcing team effectiveness.2 Canonical
analysis was used to identify the correlation between a set of independent variables
and a set of dependent variables. The rational for this approach was to use a
powerful analytic method capable of identifying whether additional variables other
than those identified a prior significantly related to cross-functional team
effectiveness. Within this research, canonical analysis was used as an explorawry '
approach to identify potentially critical variables relating to team effectiveness not
originally specified during the development of a priori hypotheses.

Canonical correlation and regression analysis is a form of multiple regression
that identifies two sets of weights--one set for the independent variables and another
set for the dependent variables. The multiple dependent variables included the four
dimensions of team effectiveness identified during exploratory factor analysis. The
independent variables included each major variable measured during the study. '
When separate linear composites were found between the two sets of variables, they
represented maximum correlations between a standardized independent variable (a

covariate) and a canonical variable (variables created from combining multiple

 

For a more detailed discussion of canonical correlation and regression analysis, see
Cohen and Cohen 1983, pages 453-465.

197

dependent variables) (Cohen and Cohen 1983). No demographic variables were
included in the canonical analysis because a priori testing revealed an almost
complete lack of significance of those variables in this study. The significant findings
that did occur were weak and at a level expected by chance.

Once the key relationships between standardized covariates and the canonical
variables was identified, post hoc hierarchial regression analysis was used to partition
the unique variance explained by each independent variable. This approach helps
overcome a misleading aspect of interpreting the output from canonical analysis.
The substantive interpretation of a canonical variate by reference to the covariate
weights may be misleading because these weights are partial regression coefficients
subject to redundancy and suppression effects (Cohen and Cohen 1983). Thus,
hierarchical analysis supported identifying the unique contribution of each
independent variable suggested by the canonical. analysis.

The objective of the post hoc analysis was to employ a powerful exploratory
method to identify additional variables capable of explaining a greater amount of
total variance than that explained by the original a prior variables. Tables 36-40
present the key output from the canonical analysis.

Tables 36 and 38 present the correlation between the dependent variables and
the canonical variables or factors. The data indicated that the first canonical variable
related to the overall measure of team performance (i.e. team process performance).
The second canonical variable related most directly to supply base management

performance. The third canonical variable related to the time performance

198

dependent measure while the fourth canonical variable related to innovation
performance. The data were generally consistent between the team-reported and
externally-reported dependent measures.

Tables 37 and 39 present the correlation between the standardized covariates
(i,e, the 18 independent variables) and each canonical variable (separate tables for
team-reported and externally-reported dependent measures). The correlation
between a covariate and each canonical variable or root (Tables 39 and 40) is
conceptually similar to a factor loading score within exploratory factor analysis.
Thus, within each canonical variable, the canonical correlations that appear as
candidates for additional exploratory analysis are italicized. The covariates with an
italicized correlation within each canonical variable became the selected independent
variables evaluated during post hierarchical regression analysis.

Tables 40-45 present the results of post hoc hierarchical regression analysis
for the variables suggested by the canonical analysis presented in Tables 37 and 38.
As expected, post hoc analysis reconfirmed the a priori findings concerning the
significance of the relationship between (1) organizational resources, (2) team
leadership, and (3) team effort to the four dimensions of cross-functional sourcing
team effectiveness. In particular, leadership-related variables and effort explained
the majority of variance associated with the team-reported process performance
dependent variable (Table 40). Organizational resources appeared particularly
critical to team and externally-reported supply base management effectiveness

(Tables 41 and 45). Although open team communication demonstrated a significant

199

relationship with team-reported process performance, the amount of variance
uniquely attributable to open communication is small (Table 40).

More importantly, the post hoc analysis revealed the presence of additional
variables that relate directly to cross-functional sourcing team effectiveness. These
include team authority and supplier participation and involvement. Both extemally-
reported and team-reported data sources agreed that higher levels of internal team
process authority related directly to effective team process performance. Recall that
internal process authority is one of the four dimensions of team authority identified
during exploratory factor analysis.

Post hoc analysis also revealed that several dimensions of supplier
participation relate directly to cross-functional sourcing team effectiveness. Table 41
reveals that a significant relationship between supplier performance support and
team-reported supply base management performance exists. Externally-rated data
indicated that supplier meeting support related directly to team process performance
and supply base management performance (Tables 44 and 45). These additional
findings are critical to the overall conclusions of this research project. The post hoc
analysis provides evidence of the critical importance of team empowerment and
external supplier support. The analysis revealed no additional variables beyond

authority and supplier participation that significantly relate to team effectiveness.

200

Table 36
Post Hoc Analysis
Correlations Between Team-Reported Dependent Variables
and Canonical Variables

Canonical Variable:

Dependent Variable: 1 2 3 4
Process Performance .97 -.12 -.16 .11
Innovation Performance .77 -.21 -.29 -.52
Time Performance .52 .21 -.82 .10
Supply Base Management .71 .44 -.54 -.05
Performance
Table 37

Post Hoe Analysis
Correlations Between Covariates and Canonical Variables
Using Team-Reported Dependent Measures

Canonical Variables:

Covariate: 1 2 3 4
Team Effort .87 .03 -.05 .17
Organizational Resources .73 .29 -.40 .11
Open Communication .78 -.17 -.17 .05
Accurate Communication .64 .12 .01 .30
Team Cohesiveness .85 -.13 -.15 -.09
Supplier Performance Support .32 .39 -.04 -.04
Supplier New Product Support .30 .37 .14 -.12
Supplier Meeting Support .29 .36 -.29 -.08
External Decision Authority .15 -.06 -.44 -.05
Internal Team Process Authority .40 -.15 -.06 -.04
Selection Authority .00 .04 -.30 .03
Scheduling Authority .31 -.04 .03 -.00
Consideration Leadership .66 .04 -.05 -.11
Behavior
Initiating Structure Leadership .56 -.15 .13 -.35
Behavior
Resource/Commit. Attainment .81 -.17 -.17 .11
Performance Evaluation and .24 .40 -.16 .10
Reward System
Participation Rewards .07 .13 .20 .3]

Collectivism . 14 -.44 .05 .00

201

Table 38
Post Hoc Analysis
Correlations Between Externally-Reported Dependent Variables
and Canonical Variables

Canonical Variable

Dependent Variable: 1 2 3
Process Performance .99 -.09 .07
Innovation Performance .48 .15 .15
Time Performance .34 -.65 .57
Supply Base Management .50 -.71 .15
Performance
Table 39

Post Hoc Analysis
Correlations Between Covariates and Canonical Variables
Using Externally-Reported Dependent Measures

Canonical Variables:

Covariate: 1 2 3
Team Effort .84 -.05 .34
Organizational Resources .16 -.73 -.16
Open Communication .39 -.40 .11
Accurate Communication .21 -.36 .26
Team Cohesiveness .35 -.32 .34
Supplier Performance Support .19 -.13 -.12
Supplier New Product Support .06 -.07 —.19
Supplier Meeting Support .32 -.38 -.20
External Decision Authority -.01 -.19 .30
Internal Team Process Authority .22 -.14 .09
Selection Authority -.03 -.48 .22
Scheduling Authority .19 .05 .02
Consideration Leadership .19 -.35 .16
Behavior
Initiating Structure Leadership .41 .00 .00
Behavior
Resource/Commit. Attainment .33 -.33 .10
Performance Evaluation and .09 -.40 .21
Reward System
Participation Rewards .16 -.27 .11
Collectivism .03 .23 .24

.12
.85
.37
.48

.11
.18
-.12
-.09
-.17
.10
.11
-.19
.13
.07
.06
-.60
-.15

-.05

-.15
-.41

-.28
.19

202

Table 40
Post Hoc Analysis
Hierarchical Regression Analysis
Team-Reported Process Performance

Dependent Variable: Team-Reported Process Performance

Cumulative SR

Variable: R Square: Square: Beta: Sig.:
Internal Team Process Authority .1328] .1328] .36 .00
Resource/Commit. Attainment .53153 .39872 .65 .00
Initiating Task Structure .53192 .00039 .03 .77
Consideration Leadership .53755 .00563 .10 .27
Organizational Resources .56167 .02412 .21 .02
Team Efl'ort .68254 .12087 .48 .00
Open Communication .69758 .01234 .19 .03
Accurate Communication .-70586 .00828 -.15 .10
Team Cohesiveness .70824 .00238 .11 .38
Table 41

Post Hoc Analysis
Hierarchical Regression Analysis
Team-Reported Supply Base Management Performance
Dependent Variable: Team-Reported Supply Base Management Performance

Cumulative SR

Variable: R Square: Square: Beta: Sig.:
Organizational Resources .42922 .42922 .66 .00
Performance Evaluation

and Reward System .43190 .00268 .06 .53
Team Member Collectivism .43277 .00087 -.03 .72
Supplier Performance Support .46169 .02899 .18 .04
Supplier New Product Support .47379 .01210 .15 .17
Supplier Meeting Support .49181 .01802 .17 .09

203

Table 42
Post Hoc Analysis
Hierarchical Regression Analysis
Team-Reported Time Reduction Performance

Dependent Variable: Team-Reported Time Reduction Performance

Cumulative SR

Variable: R Square: Square: Beta:

Organizational Resources .25908 .25908 .5]

Supplier Meeting Support .3012] .04213 .2]

External Decision Authority .30692 .00571 .08

Selection Authority .30753 .00061 .03
Table 43

Post Hoc Analysis

Hierarchical Regression Analysis

Team-Reported Innovation Performance

Dependent Variable: Team-Reported Innovation Performance

Cumulative SR

Variable: R Square: Square: Beta:

Performance Evaluation .00385 .00385 -.066
and Reward System

Initiating Task Leadership .13050 .12655 .36

Accurate Communication .16546 .03496 .19

Sig.:

.52

204

Table 44
Post Hoc Analysis
Hierarchical Regression Analysis

Externally-Reported Process Performance

Dependent Variable: Externally-Reported Process Performance

Cumulative SR

Variable: R Square: Square: Beta: Sig.:
Internal Team Process Authority .04556 .04556 .2] .04
Supplier Meeting Support .08193 .03637 .20 .05
Resource/Commit. Attainment .1247] .04278 .22 .04
Initiating Structure Leadership .15550 .03079 .22 .05
Team Efl'ort .51640 .36090 .67 .00
Open Communication .52841 .01201 .15 .14
Team Cohesiveness .53786 . .00945 -.20 .18
Table 45
Post Hoc Analysis

Hierarchical Regression Analysis
Externally-Reported Supply Base Management Performance

Dependent Variable: Externally-Reported Supply Base Management Performance

Cumulative SR

Variable: R Square: Square: Beta: Sig:
Performance Evaluation .00621 .00621 .08 .45
and Reward
Selection Authority .02793 .02172 .15 .16
Supplier Meeting Support .07424 .0463] .22 .04
Resource/Commit. Attainment .11654 .04230 .22 .04
Consideration Leadership .13564 .01910 -.20 .17
Organizational Resources .32830 .19266 .47 .00

V. RESEARCH DISCUSSION

The analysis reported in Chapter IV revealed that a set of variables exist that
relate directly to cross-functional sourcing team effectiveness. This chapter presents
an in-depth evaluation of the variables identified as relating directly to cross-
functional sourcing team effectiveness along with the relationships those variables
may have had with other research items. This presentation also includes a discussion
of additional research issues and questions associated with each variable. The critical
variables discussed include (1) organizational resources, (2) supplier participation,
(3) team authority, (4) team effort, and (5) team leadership.

This chapter also briefly discusses variables included as part of the research
but not identified as critical to the cross-functional sourcing teams participating in
this study. This includes (1) demographic indicators, (2) performance evaluation and
reward systems, (3) team cohesiveness, and (4) team communication. The chapter

concludes with formal discussions of research extensions and limitatiOns.

DISCUSSION OF CRITICAL RESEARCH FINDINGS
Organizational Resources

The presence or absence of organizational resources related directly to cross-
functional sourcing team effectiveness. As Peters and O'Connor (1980) have argued,
a set of potentially important, but often overlooked, variables include aspects of the
immediate work situation (referred to as organizational resource constraints) that can

interfere with the translation of abilities and motivation into effective performance.

205

206

Furthermore, Guzzo (1986), in his critique of group decision making and
effectiveness, argued that resources in the organizational environment are an
important determinant of team effectiveness. He also maintained that organizational
resources play an important role in affecting a team's collective belief that the team
can be effective. A key factor influencing that belief is the extent to which an
organization makes available the support required to accomplish a team's task
assignment (Guzzo 1986). This section examines the critical relationship between
organizational resources and cross-functional sourcing team effectiveness.

As the data results presented in Chapter IV revealed, a consistent and strong

relationship existed between a linearly combined scale of organizational resource

categories and cross-functional sourcing team effectiveness.‘ An analysis of .

individual resources revealed, however, that not all resource categories relate equally
to cross-functional sourcing team effectiveness. Certain resources appear to be more
important to team performance than other resources. Furthermore, not all resources
were available to teams at the same level. A team may have had adequate access to
one resource but not to another.

Table 46 presents findings about member-reported and externally-reported
availability of each resource category. While differing in the their assessment of the

overall availability of each resource category, external raters and team members

 

The ten resource categories evaluated during this research include (1) job-related
information, (2) tools and equipment, (3) materials and supplies, (4) budgetary
support, (5) required services and help from others, (6) team member task
preparation, (7) time availability, (8) work environment and (9) executive
management commitment, and (10) supplier participation.

.“".'I d‘. 1

 

207

generally agreed about the order of resources available to teams. The time to pursue
team task assignments, required services and help from others, and budgetary
support were rated the least available of all resources. As will be discussed, these
resources also related the strongest to cross-functional sourcing team effectiveness.

Research results suggested that inadequate time to support team assignments
appears to be a critical team issue. Over 15 percent of all team members and over
25 percent of all team leaders indicated that a lack of time has severely hampered
team performance. Furthermore, team members indicated they committed 15
percent of their time on average to team-related activities. They further indicated
they should commit at least 20 percent of their time on average to support team
goals and objectives. This represents a 33 percent increase over the present time
commitment. Team leaders indicated they committed almost 27 percent of their
time on average to team-related activities. They further indicated they should
commit at least 35 percent of their time on average to support adequately team-
related activities.

The data also indicated it was not unusual for a team member or leader to
be simultaneously part of more than one cross-functional sourcing team. This may
further increase the potentially negative impact that time constraints can have on
team effectiveness. This issue may be a concern because time availability related
strongly to team effort and effectiveness. Its availability as an organizational

resource, however, was well below other resource categories.

208

Table 46
Externally and Team-Reported
Organizational Resource Levels

Member-Reported Resource: Availability:
Materials and Supplies 5.7
Work Environment 5.6
Supplier Participation 5.6
Team Member Task Preparation 5.5
Job-Related Information 5.3
Tools and Equipment 5.3
Executive Management Commitment 5.2
Budgetary Support 4.9
Required Services and

Help From Others 4.9
Time Availability 4.1

Externally-Reported Resource: Availability:
Materials and Supplies 6.1
Work Environment 6.0
Team Member Task Preparation 5.9
Supplier Participation 5.8
Job-Related Information 5.7
Executive Management Commitment 5 .7
Tools and Equipment 5.7
Budgetary Support 5.4
Required Services and

Help From Others 5 .2
Time Availability 4.4
Scale: Completely Unavailable

1 =
4 = Moderately Available
7 = Completely Available

209

All organizations included in this study added cross-functional sourcing team
assignments to existing employee job assignments. Furthermore, and as discussed
elsewhere, relatively few team members received formal recognition and/or reward
for team participation. Over 50 percent of team members indicated they received
no formal evaluation or appraisal for team participation. These findings suggested
that added team assignments with little or no recognition for participation can
combine to make a lack of time availability a potentially serious barrier to effective
team interaction.

Certain organizational resources related systematically to cross-functional
sourcing team effectiveness in this study. Table 47 details the team-reported
relationship between each organizational resource category and the two major team
effectiveness dimensions evaluated during the study--process performance
effectiveness and supply base management effectiveness. Table 48 details the
externally-reported relationship between each resource category and the same two
effectiveness dimensions.

General agreement existed between the two data sources about the resources
that relate the highest with team effectiveness. External raters indicated, however,
that effective teams received greater executive support and commitnrent compared
to less effective teams. External raters were usually members of executive
management several organizational levels removed from the team. This may help

explain why team members stressed the importance of adequate tools and

210

Table 47
Correlations Between Team-Reported Resources
and Team-Reported Effectiveness

Effectiveness Dimension: Process Performance

Budgetary Support .52
Time Availability .49
Tools and Equipment .44
Team Member Task Preparation .42
Required Services and Help From Others .39
Executive Management Commitment .38
Job-Related Information .34
Materials and Supplies .30
Supplier Participation .28
Work Environment .23

Effectiveness Dimension: Supply Base Management

Performance
Supplier Participation .60
Required Services and Help From Others .59
Time Availability .48
Budgetary Support .46
Team Member Task Preparation .40
J ob-Related Information .38
Tools and Equipment .38
Executive Management Commitment .29
Work Environment .18

Materials and Supplies .16

211

Table 48
Correlations Between Externally-Reported Resources
and Externally-Reported Effectiveness

Effectiveness Dimension: Process Performance

Team Member Task Preparation .37
Time Availability .36
Required Services and Help From Others .26
Executive Management Commitment .22
Budgetary Support .21
Supplier Participation .12
Work Environment .06
Job-Related Information .05
Tools and Equipment .00
Materials and Supplies .00

Effectiveness Dimension: Supply Base Management

Performance
Supplier Participation .43
Required Services and Help From Others .43
Budgetary Support .37
Time Availability .35
Executive Management Commitment .33
Work Environment .25
Team Member Task Preparation .24
Job-Related Information .17
Materials and Supplies .16

Tools and Equipment .15

212

equipment for team effectiveness while external raters emphasized executive
management support and commitment.

Tables 47 and 48 provide insight into the organizational resources that related
directly to effective cross-functional sourcing teams. The key point of these tables
is that certain resource categories related systematically to team effectiveness at
higher levels.

Further analysis identified which resources relate directly to specific
effectiveness items. Recall that each effectiveness dimension (except innovation
performance) is a scale consisting of multiple items. It is not unusual for
management to form cross-functional sourcing teams with a focused or single task
or objective. For example, the focus of a team could be to reduce purchase item
cost or develop supplier performance capabilities. The resources identified as most
critical for the broader team effectiveness dimensions may not always be the
resources critical for specific task assignments.

The data indicated that not all resource categories related equally to each
effectiveness item. Tables 49 and 50 segment process and supply base management
performance indicators by the resource(s) demonstrating the strongest positive
relationship with individual performance items. In other words, for effective cross-
functional sourcing teams, which organizational resource(s) related highest to a
specific effectiveness item? Note that not all of the 28 effectiveness items evaluated
during this research had a resource category(s) identified as more critical compared

to other resources.

213

Tables 49 and 50 provide information about the relationship between
resources and specific effectiveness items. The number in the parentheses next to
each resource is the zero-order correlation between the corresponding effectiveness
item and the specific organizational resource category. The resources presented in
these tables demonstrated the highest correlation at the individual effectiveness item
level. Although the same type of analysis using externally-reported data is possible,
this analysis used team-reported data. As Peters and O'Connor have argued (1985),
the most appropriate source of information about resource availability or absence is
the job incumbent (i.e. the individual team members). It is the job incumbent who
interacts with various task demands and resources on a continuous basis and has
first-hand knowledge regarding the presence or absence of a resource. Invariably,
both external raters and team members provided subjective evaluations of
organizational resource availability.

Table 49 details the resources that related directly to individual process
performance indicators. This table reveals that certain organizational resources
demonstrated stronger relationships with various process performance indicators.
For example, teams rated highly in their ability to introduce innovative ideas
indicated that the availability of time to pursue team assignments was a critical
resource. Overall, table 49 reveals that budgetary support consistently related to
effectiveness for different process performance indicators.

Table 50 presents the strongest correlations between organizational resources

and individual supply base management performance items. Of the 18 unique

214

Table 49
Critical Relationships Between Process Performance
Items and Organizational Resource Categories

Have Access to the Following

Teams Rated Higher In: Resources at a Higher Level:
I Quantity of work 0 Budgetary support (.48)
I Quality of work 0 Budgetary support (.47)

° Team member task preparation (.45)
0 Tools and equipment (.36)

I Innovative team ideas ' Time availability (.45)
I Efficiency 0 Time availability (.48)
0 Budgetary support (.48)
0 Tools and equipment (.41)
0 Required help from others (.40)

Job-related information (.36)

I Team morale Time availability (.51)

Budgetary support (.43)

Tools and equipment (.42)

Executive commitment (.42)

Team member task preparation (.38)

lob-related information (.37)

Budgetary support (.48)

Team member task preparation (.45)
Required help from others (.45)
Tools and equipment (.44)

Time availability (.43)

Job-related information (.40)

I Ability to communicate
and coordinate across
functional boundaries

215

Table 49 (continued)
Critical Relationships Between Process Performance
Indicators and Organizational Resource Categories

Have Access to the Following
Teams Rated Higher In: Resources at a Higher Level:

I Meeting timing schedules 0 Time availability (.39)
° Budgetary support (.37)

I Meeting executive ' Budgetary support (.42)
management expectations 0 Time availability (.42)
0 Executive commitment (.40)

I Ability to work with
individuals and groups
outside the team

Required help from others (.47)
Team member task preparation (.44)
Budgetary support (.44)

Time availability (.40)

Tools and equipment (.38)

216

Table 50

Critical Relationships Between Supply Base Performance
Items and Organizational Resource Categories

Teams Rated Higher
in Their Ability to:

I Reduce purchase cost

I Improve purchase quality

I Support reduced new
product development
cycle time

I Improve supplier
delivery performance

I Support increased use of
supplier technical
capabilities

I Reduce material ordering
cycle time

I Develop sourcing strategies
that support business unit
strategies

Have Access to the Following
Resources at a Higher Level:

' Time availability (.44)
0 Required help from others (.42)
Team member task preparation (.35)

Time availability (.51)

Required help from others (.41)
Supplier participation (.37)
Budgetary support (.36)

- Required help fi'om others (.49)
0 Time availability (.43)

° Supplier participation (.52)
0 Required help from others (.51)
0 Time availability (.35)

° Supplier participation (.58)

° Supplier participation (.44)
0 Required help form others (.38)

° Supplier participation (.47)
0 Required help from others (.40)
° Budgetary support (.38)

217

Table 50 (continued)
Critical Relationships Between Supply Base Performance
Indicators and Organizational Resource Categories

Teams Rated Higher Have Access to the Following
in Their Ability to: Resources at a Higher Level:
I Support new product 0 Supplier participation (.43)
performance targets 0 Required help from others (.41)
I Achieve best-in-class 0 Time availability (.59)
supplier selection ° Required help from others (.50)
° Supplier participation (.49)
° Job-related information (.45)
' Budgetary support (.39)
0 Team member task preparation (.38)
0 Tools and equipment (.36)
I Improve supply base 0 Supplier participation (.68)
responsiveness 0 Required help from others (.51)
0 Time availability (.45)
0 Budgetary support (.45)
0 Team member task preparation (.41)
0 Tools and equipment (.39)

Job-related information (.38)

I Support inventory reductions ° Supplier participation (.50)

° Job-related information (.36)

0 Required help from others (.36)
I Develop supplier capabilities Required help from others (.53)
Supplier participation (.47)
Budgetary support (.44)
Required tools and equipment (.37)
Time availability (.36)
Job-related information (.36)

218

Table 50 (continued)
Critical Relationships Between Supply Base Performance
Indicators and Organizational Resource Categories

Teams Rated Higher Have Access to the Following
in Their Ability to: Resources at a Higher Level:

I Support early sourcing and ' Required help from others (.50)
supplier participation during 0 Supplier participation (.45)
product design 0 Time availability (.38)

° Budgetary support (.38)

I Foster development of new 0 Required help from others (.47)
technology by suppliers for 0 Supplier participation (.38)
company use

I Establish strategic ° Supplier participation (.57)
relationships with suppliers ' Required help from others (.54)

° Time availability (.42)
' Budgetary support (.42)

Team member task preparation (.36)

219

performance items assessed during the study, 15 related strongly to certain resource
category(s). The following section discusses the organizational resources rated
especially critical to team effectiveness at the individual item level.

Supplier Participation Many cross-functional sourcing team tasks involve
working directly with or managing external suppliers. The direct relationship
between supplier participation and team effectiveness for individual performance
items is an important research finding. The degree to which suppliers directly
supported a team‘s task may be critical for tasks which can benefit from supplier
involvement. The data indicated that supplier participation and involvement relate
directly to most of the individual supply base management performance items
evaluated during this research. Teams rated more effective on most supply base
management effectiveness items indicated they had access to supplier participation
at a higher level. A later section discusses this topic in greater detail.

Required Services and Help From Others Data analysis indicated that a second
major resource relating directly to specific effectiveness items was required services
and help from others. We can speculate about why effective cross-functional
sourcing teams rely on external assistance at higher levels. First, a team could
consist of relatively few members and lack direct support in certain functional areas.
A team may require expert knowledge from functions or personnel not represented
on the team. Secondly, a team may require external assistance to implement or
review the team's decisions. The importance of required services and help from

others, like supplier participation and involvement, demonstrates that teams

 

220

necessarily are not independent entities. Cross-functional sourcing teams must still
work across external team boundaries. Third, required services and help from others
may actually be related to the supplier participation resource measure. Supplier
participation is a form of required services and help from others. Teams indicated
that the availability of services and help from others relates directly to each item
detailed in Table 50.

Time Availability The amount of time that team members committed to team
assignments also correlates highly with individual supply base management
effectiveness items. As mentioned, team members indicated this resource is
consistently the least available of all organizational resources. As such, time
availability may act as a constraint rather than a resource. It is difficult to imagine
an effective team that lacked the time to pursue a meaningful task assignment.
While this makes intuitive sense, the evidence gathered in this study strongly
suggested a critical relationship between time availability and cross-functional
sourcing team effectiveness.

Budgetary Support The availability of adequate budget also related directly to
teams rated more effective on many supply base management effectiveness items.
This particular resource may be especially critical to the 40 percent of the teams in
this study that had members located flying distance apart. Budget support is critical
for travel authorization and face-to-face meetings.

Segmenting resources by individual effectiveness item provided additional

insight into those resources that relate most directly to cross-functional sourcing team

 

221

tasks. The data suggested that allocating organizational resources differentially
depending on a team's task focus may affect team effectiveness. Matching critical
resources to team task, as suggested by the results of this study, may increase the
probability of allocating properly valuable organizational resources.

Research Issues and Questions This research identified a strong relationship
between various organizational resources and cross-functional sourcing team
effectiveness. It is likely that given the proper circumstances, a lack of any resource
category could affect team performance. The data clearly suggested, however, that
not all resources relate equally to team effectiveness. In particular, (1) required
services and help from others, (2) supplier participation, (3) time availability, (4)
team-member task preparation, and (5) budgetary support appeared most critical
overall to team effectiveness.

The importance of supplier participation and required services and help from
others indicated that cross-functional sourcing teams often require external
assistance. Additional research is required in this area to further our understanding
about how teams gain access to external assistance. While the data suggested that
teams with access to external assistance are more effective than team with less access,
the cross-sectional nature of this study precludes making a causal connection between
external assistance and team effectiveness.

Both external raters and team members agreed that more effective teams had
adequate time to pursue team assignments. The design of this study also prohibits

concluding that a causal link existed between time and effectiveness. However, the

222

data suggested that a serious factor limiting team performance potential may be a
lack of time available for team assignments. Future research should address how
serious an effect a lack of time for cross-functional sourcing team assignments may
have on team performance potential.

There are at least three reasons why time availability may be a serious issue
confronting cross-functional sourcing teams. First, time availability related directly
to the effort a team put forth on its assignments. A reduction in time availability
may mean a reduction in team effort. Second, team participation is usually an added
task assignment with limited or no recognition or reward provided for participation.
Third, cross-functional sourcing teams are responsible for tasks historically associated
with purchasing. The effective use of these teams, however, requires a time
commitment from nonpurchasing members. There is a risk, as evidenced from data
gathered during the study, that functional groups will focus on functional tasks when
confronted with professional time constraints.

Extended research in this area can evaluate a number of questions designed
to address the important issue of cross-functional sourcing team time constraints:

(1) Can management use a performance evaluation and reward system to .

encourage team member participation and commitment? Should
teams be responsible for and measured against specified team goals

and objectives?

(2) Does securing a member's verbal support towards the team process
affect that member's commitment to the team?

(3) Can limiting the number of teams that a member belongs increase the
professional time committed to the team?

 

(4)

(5)

(6)

(7)

(8)

(9)

223

Can larger teams be reduced in size? For example, can the use of a
core team with external support reduce the potential for time
constraints among team members?

Is it possible that smaller teams actually create time constraints by
placing additional workloads on team members?

Do teams have a life cycle that may allow the elimination or
discontinuance of certain teams?

What is the effect of executive management commitment on the time
a team (and team member) allocates to its assignments? If executive
management commitment affects team member time commitment,
what actions can executive management take to express its support
towards the cross-functional sourcing team process?

Is team member rotation a viable option for members unable to
provide adequate support to team assignments?

Can teams be structured to minimize task assignment overlap? If
overlap exists, can team builders combine team tasks to increase team
efficiency and reduce total time requirements? Also, is an
organization using a team just because it believes it needs to use
teams?

The results of this study suggested that without adequate time, the opportunity

for a cross-functional sourcing team to achieve its theoretical performance potential

likely diminishes. Adequate time to support team assignments, however, does not

guarantee team success. In fact, one can make the argument that a poorly composed

team with ample time for team assignments may actually create greater

organizational harm. Future research under more controlled conditions should

investigate the actual causal effect that time availability may have on cross-functional

sourcing team performance.

This research revealed that a set of potentially critical organizational resource

categories existed that relate directly to cross-functional sourcing team effectiveness.

224

Future research can focus on these resource categories specifically to evaluate their
effect on team performance. Insight into how organizations provide or transfer
critical resources to the cross-functional sourcing team may also prove valuable

towards furthering our understanding in this area.

Supplier Participation

The importance of supplier participation was evident throughout this study.
Post hoc data analysis revealed that the availability of supplier participation and
support relate directly to cross-functional sourcing team effectiveness, particularly the
effectiveness dimension of supply base management performance. Furthermore,
teams which indicated greater reliance on suppliers to support directly the team's
goals and objectives were significantly more effective than teams relying on supplier
reliance to a lesser degree. Table 51 details the relationship between various
dimensions of supplier participation and involvement and team-reported
effectiveness.

As an organizational resource, supplier participation was also critical for
supporting most individual indicators of supply base management performance (refer
to Table 50). Recall that supplier participation as an organizational resource
demonstrated the highest correlation with externally and team-reported supply base
management effectiveness compared to all other organizational resource categories.

The a priori research hypotheses did not include supplier involvement as a key

 

225

Table 51
Correlations Between Team-Reported Effectiveness
and Supplier Involvement Indicators

Team Effectiveness Dimension:

Process Innovation Time Supply Base

Performance Performance Reduction Mamgement
Supplier Involvement

Indicator:

Performance improvement

support .34** .24* .28“ .38”
New product development

support .24“ .08 .19 .25*
Team meeting support .32“ .29“ . .31” .38*-*
Overall reliance on suppliers .32“ .17 .35M .45"

* Significant at .05 level
** Significant at .01 level

226

predictor of team effectiveness. Subsequent data analysis, however, indicated this
variable relates directly to cross-functional sourcing team performance.

No previous research was found that examined the potential impact of
supplier participation on cross-functional sourcing teams. However, it is now
apparent that reliance on supplier participation and performance contribution makes
cross-functional sourcing teams unique compared to other organizational work teams.
The data clearly indicated that supplier participation, when required by a cross-
functional sourcing team, relates directly to team effectiveness.

This research evaluated supplier participation and involvement in four
separate sections of the survey. First, team members, leaders, and external raters
evaluated the presence of supplier participation as an organizational resource.
Second, team leaders evaluated three dimensions of supplier involvement: supplier
performance improvement support, support at team meetings, and supplier new
product development support. Third, the team leader evaluated the degree to which
the team relied on external supplier input to support directly the team's goals and
objectives. Finally, the team leader indicated whether suppliers were formal
members of the cross-functional sourcing team. Each variable and data source
agreed that supplier participation and involvement, both formally and informally,
relates directly to team performance.

Both external raters and team members also agreed that teams which included
suppliers as formal team members were, on average, more effective compared to

teams which did not include suppliers as formal members. In particular, teams with

 

227

formal supplier participation were more effective in the performance areas of time
reduction and supply base management. Overall, teams which relied on supplier
input and involvement both formally and informally reported significantly greater
effectiveness in the performance areas of time reduction and supply base
management. This finding was consistent across team-reported and extemally-
reported effectiveness measures. A positive link existed between supplier
involvement and participation and cross-functional sourcing team effectiveness.

The data also indicated a strong relationship existed between supplier
participation and the quality of information exchange between the team and key
suppliers. This suggested that perhaps supplier involvement with team activities
promotes the establishment of necessary communication channels between the team
and its suppliers. Ideally, these channels can promote effective two-way
communication between organizations. Future research can assess the impact of the
quality of information exchange between a team and its suppliers and team
performance.

The data indicated a number of other positive relationships between supplier
participation and cross-functional sourcing team activities. Teams which included
suppliers as formal members had (1) greater satisfaction concerning the quality of
information exchange between the team and its key suppliers, (2) a higher reliance
on suppliers to support directly the team's goals and objectives, (3) greater supply
base management effectiveness, and (4) greater supplier contribution in the areas of

performance and new product support. Supplier performance support includes

 

228

providing cost reduction ideas, providing quality improvement ideas, supporting
actions to improve material delivery, offering process technology ideas, and
supporting material ordering cycle time reductions. Supplier new product support
includes offering new product ideas, performing product design activities, and
supporting new product development cycle time reductions. Table 52 details the
bivariate correlations between formal supplier participation and other study variables.

The finding of a link between formal participation and cross-functional
sourcing team effectiveness is critical. Only 31 percent of the teams in this study
included suppliers as formal members. Findings concerning the significance of formal
participation, therefore, reveal the power of participation to affect the cross-
functional sourcing team process and performance.

Interestingly, the data indicated that teams including suppliers as formal
members also had greater authority to select members and manage internal team
processes. We can speculate that teams with greater member selection and internal
process authority use this authority to promote formal supplier participation. One
way that teams may link authority and team effectiveness is by actively including
suppliers as formal team members. The results of the study indicated that a link
existed between member selection/internal team process authority, formal supplier
team participation, and cross-functional sourcing team effectiveness.

One issue is whether teams with formal supplier participation were different
from teams without formal supplier participation. In other words, could other

variable(s) have mediated the relationship between formal supplier participation and

229
Table 52

Correlations Between Formal Supplier Team Membership
and Other Key Study Variables

Indicator: Correlation:

Quality of information exchange between the team .24*
and key suppliers

Team reliance on suppliers to support team .32**
goals and objectives

Supply base management effectiveness .25*
(team-reported)

Receive supplier performance improvement .25*

support
Receive supplier new product support .35“
Receive supplier support at team meetings .38”

Note: This is the correlation between a dichotomous variable that evaluated whether
suppliers were formal team members and the statements listed in the table.
For example, the second statement is interpreted as teams which included
suppliers as formal team members also tended to rely on supplies to support
team goals and objectives at a higher level.

* Significant at .05 level
** Significant at .01 level

230

team effectiveness? Examination of the relationships between formal participation
and other study variables revealed no significant relationships. Teams with formal
supplier participation did not demonstrate greater cohesiveness, leadership
effectiveness, effort, communication etc. Formal participation related directly only
with team authority, supplier participation and involvement variables, and team
effectiveness.

Another indication of supplier participation and involvement concerned the
degree to which suppliers met with and provided support to a team during meetings.
Supplier participation at meetings was sometimes, but not always, related to formal
supplier involvement. In other words, suppliers did not have to be formal team
members to attend and provide support at team meetings. The data indicated that
teams reporting that suppliers provided support to the team by attending team
meetings (1) put forth greater overall effort on tasks, (2) achieved generally higher
levels of externally and team-reported effectiveness, (3) had a higher degree of
satisfaction concerning the quality of information exchange between the team and
key suppliers, (4) experienced fewer problems coordinating work activity between the
team and key suppliers, and (5) reported particularly higher supplier performance
improvement and new product contribution. The data suggested that supplier
involvement at team meetings may be one way to realize the benefits that can result
from closer buyer-seller interaction. Table 53 presents the correlations between
teams receiving higher level of support from suppliers at team meetings and other

items evaluated during the study.

231
Table 53

Correlations Between Supplier Support at Team Meetings

and Other Key Study Variables

Indicator:

Quality of information exchange between the team and
key suppliers

Team-reported effort

Externally-reported effort

Team-reported process performance effectiveness
Team-reported supply base management effectiveness
Externally-reported process performance effectiveness
Externally-reported supply base management effectiveness

Ability to coordinate work activity between the team
and key suppliers

Overall reliance on supplier support
Receive supplier performance improvement contribution

Receive supplier new product development support

Note: This data represents the correlations between higher levels of support
provided by suppliers at team meetings and higher levels of each item listed
in the table. For example, teams that received higher levels of support from
suppliers at team meetings indicated higher quality of information exchange

between the team and key suppliers.

* Significant at .05 level
** Significant at .01 level

Supplier Support
at Team Meetings:

.34”

.36"
.22*
.32"
.39“
.24*
25*

.283338

.37**
.56**

.39**

232

Research Issues and Questions The consistent finding across multiple study
variables of the relationship between supplier involvement and cross-functional
sourcing team performance suggested that future research may provide additional
insight into this area. For example, what tasks can benefit from supplier involvement
and at what point of a team's life can supplier involvement become a realistic option?
Clearly, not all cross-functional sourcing teams require direct supplier involvement
and participation. The degree or extent of supplier involvement depends upon a
team's task. A team with responsibility for evaluating and selecting new suppliers is
not likely to include suppliers as part-of the team. However, teams responsible for
developing supplier capabilities or improving supplier responsiveness may be more
likely .to include suppliers over time. When supplier selection has already occurred,
supplier participation may become a viable option. Almost all. team members (95
percent) indicated that supplier participation was possible at the point of the team's
life which this research captured. Further research can help provide additional
insight into how firms can best create performance benefits from supplier team
involvement.

Research conducted by Monczka, Trent, and Gay (1992) found that over 60
percent of firms plan to emphasize buyer-seller performance improvement teams
during the next three years. The design of these teams includes supplier
participation by design. Almost 80 percent of firms plan to emphasize cross-
functional teams for sourcing decisions during the next three years. The teams

included in this study more often than not exclude suppliers as formal participants.

233

Almost all firms recognized the potential benefit, however, of closer
interorganizational relationships (Monczka, Trent, and Gay 1992). Including
suppliers as a formal part of the team structure may promote positive interaction and
relationships between buyers and sellers.

Previous research concluded that obstacles often exist prohibiting the use of
suppliers as part of a cross-functional team (Monzcka, Trent, and Gay 1992). The
Monczka et. al research found that over 70 percent of the firms studied indicated
that factors existed which limited supplier participation in product design and
development. Furthermore, almost half of all firms surveyed further indicated that
confidentiality is the major linriting factor to direct supplier participation and early
supplier involvement. A secondary obstacle included the use of suppliers who were
not leaders'in product or technology development. Teams that support new product
development or other assignments which involve sensitive information may face
obstacles to supplier team participation. Future research in this area can focus on
ways that firms can control or minimize the obstacles to closer buyer-seller team

interaction.

Team Authority
The critical relationship between cross-functional sourcing team authority and

team effectiveness is an important research finding.2 Surprisingly, minimal research

 

Different sources refer to team authority in several ways. For our discussion,
authority, autonomy, and empowerment each refer to the same concept.

234

has explored the link between various levels of team authority and team
effectiveness. Unfortunately, research hypothesis development failed to include team
authority as a key predictor of cross-functional sourcing team effectiveness. Several
theoretical models of group/team effectiveness, however, have addressed the
importance of task team autonomy. Guzzo (1986) noted that the socio-technical
perspective of group effectiveness regards autonomy as crucial to the creation of
effective work groups or teams. Autonomy enables a team to control its own task
performance procedures and boundary transactions. Similarly, Hackman (1987) has
viewed autonomy as important because autonomy stimulates effort and greater use
of member skills and abilities. Furthermore, teams with higher autonomy tend to
adopt appropriate task performance strategies.

This research moved beyond a general discussion of autonomy by identifying
through factor analysis four independent dimensions of team authority. In this study,
task autonomy, authority, or empowerment referred to the degree to which a team
has the ability to (1) exercise scheduling authority, (2) select new members and/or
the team leader, (3) control internal team processes, and (4) make decisions affecting
others external to the team. These four dimensions of authority each represent a
different aspect of team empowerment.

The first dimension, scheduling authority, represents a team's ability to
schedule its meetings without others approving the scheduling decision. Almost all
teams in this study had scheduling authority. The next dimension, membership

selection authority, is the ability of a team to select its leader and new team members

235

as required. Fewer teams had total authority to exercise this type of activity as
compared to scheduling authority.

Internal team process authority, the third dimension, is the ability of a team
to allocate budget and material resources to support team activities, determine team
performance goals and objectives, make timing decisions for the completion of team
activities and assignments, and request non-team members to support team
assignments as required. Internal team process authority represents generally a
higher level of empowerment compared to scheduling and selection authority. Total
internal authority empowers a team to manage its internal processes without
executive management approval. The final dimension, external decision authority,
allows a team to make supplier selection decisions, establish purchase prices, and
develop sourcing and negotiating strategies without external approval. This is
conceptually a higher dimension of authority because the team has the power to
make decisions that externally bind or commit an organization.

Segmenting team authority by these dimensions is important. Post hoc
analysis along with analysis of bivariate correlations revealed that teams differing by
their level of empowerment demonstrated significantly different performance results.
The ability to control internal team processes and to make external decisions, for
example, related directly to team effectiveness. Table 54 presents data supporting
these conclusions between the four authority dimensions and cross-functional

sourcing team effectiveness.

236

Table 54
Correlations Between Team Effectiveness
and Team Authority Dimensions

Externally-Reported Effectiveness:
Process Innovation Time Supply Base

Performance Performance Reduction Mamgement
Team Authority Dimension:

Scheduling Authority .12 .05 .04 .08
Selection Authority .05 -.04 .27“ .17
Internal Process Authority .21* .09 .26" .19
External Decision Authority .07 .14 .29** .21*

Team-Reported Effectiveness:

Process Innovation Time Supply Base

Performance Performance Reduction Mamgernent
Team Authority Dimension:

Scheduling Authority .13 .16 .06 .10
Selection Authority .07 .00 .18 .14
Internal Process Authority .36“ .27** .20* .23*
External Decision Authority .22* .16 .24* .23*

* Significant at .05 level
** Significant at .01 level

237

Scheduling Authority Because almost all teams in this study had the authority
to schedule team meetings without external approval, this dimension provided
minimal differentiation among teams. If we can conclude that team empowerment
relates directly to greater effectiveness, then teams should have, at a minimum, the
authority to schedule team meetings. The data indicated that no significant
relationship existed between scheduling authority and team effectiveness.

Selection Authority Teams with the authority to select leaders and/or team
members also did not demonstrate any direct relationship with team effectiveness.
The data did indicate, however, that several important relationships existed between
teams with selection authority and other study variables.

Teams with higher levels of member selection authority were more likely, on
average, to include suppliers as formal members of the cross-functional sourcing
team. This is important because of the previously discussed relationship between
formal supplier involvement and team effectiveness. This finding suggested that
teams, when provided the authority to do so, appeared to include suppliers as formal
team members more often than teams that lacked selection authority. Additional
investigation should verify the statistical relationship between selection authority and
formal supplier involvement.

Teams with member selection authority also had, on average, team members
with greater cross-functional team experience. This could suggest that teams recruit
members with greater team experience. Speculatively, a more likely explanation is

that previous cross-functional team experience reflects an organization's historical

238

involvement with cross-functional teams.3 One hypothesis is that organizations
experienced with cross-functional involvement likely empower their teams at higher
levels. A direct relationship also existed between a team's previous cross-functional
team member experience and internal team process authority. The data suggested
that greater team empowerment likely evolves as organizations become increasingly
familiar with the cross-functional team process. As cross-functional teams become
part of a firm's formal structure, the granting of higher team authority probably
results. Future research can investigate the evolution of authority and cross-
functional teams within organizations more thoroughly.

Internal Team Process Authority As indicated in Table 54, teams granted
internal process authority demonstrated higher levels of process performance
effectiveness when compared to the other dimensions of team authority. This was
consistent for externally-reported and team-reported process performance ratings.
This suggested a link between the ability of a team to control its internal processes
and its performance.

Internal process authority also related positively to other aspects of team
interaction, particularly when compared to the other authority dimensions. Other
direct relationships involving teams with greater internal process authority included:

(1) Greater accuracy of within-team communication

(2) Higher externally and team-reported effort

 

No relationship existed between team member cross-functional experience and
age. This eliminates age as a possible predictor of team experience.

239

(3) Greater satisfaction with the methods of information exchange
between team members

(4) Greater satisfaction with the usefulness of formal meetings

The data suggested that teams with internal process authority may achieve
many of the performance and process outcomes sought by firms. The relationship
between internal process authority and effectiveness/process outcomes indicated that
internal empowerment likely contributes to higher performance. Further
investigation is required to establish the causal link between internal process
authority and the performance indicators cited here. Table 55 presents the
correlations between the above cited items and internal team authority.

External Decision Authority External decision authority allows a team to make
decisions involving supplier selection and management. Total external decision
empowerment is critical because it represents a transfer of decision-making authority
from an existing decision-maker to a cross-functional sourcing team. With total
external decision authority, a team now has empowerment to make decisions that
were formally the responsibility of managers. Across the entire study, teams were
granted external decision authority at levels comparable to internal decision
authority. It is not necessarily true, however, that teams with external authority
necessarily had a comparable level of internal authority (and vice versa). A
correlation of .36 existed between a team's level of internal and external decision-
making authority. (A stronger correlation of .61 exists between internal process

authority and selection authority).

240

Table 55
Correlations Between Internal Team Process Authority
and Other Study Items

Correlation With
Internal Process

Item: Authority
Accuracy level of within-team communications .29
Externally-reported team effort .27
Team-reported team effort .33
Satisfaction with the methods of information .31

exchange between team members

Satisfaction with the usefulness of formal .35
team meetings

Note: All correlations significant at the .01 level

241

The data presented in Table 54 indicated that teams with external decision
authority received, on average, slightly higher externally-reported ratings for time
reduction and supply base management performance as compared to the other
dimensions. This was consistentacross externally-reported and team-reported data.
This may suggest that teams empowered to make external sourcing decisions are able
to channel this empowerment into greater performance results.

External raters indicated that teams with external decision authority received
greater budget support compared to the other authority dimensions. This suggested
a possible link between team importance (reflected in the team's empowerment) and
executive management's recognition of this importance through budgetary support.
It is also possible, however, that geographic distance influenced the relationship
between external decision authority and budgetary support. Teams with greater
external decision authority in this study often included members separated by greater
geographic distances. Furthermore, teams separated by greater geographic distances
were often teams coordinated at the corporate level. These teams required
budgetary support (particularly travel support) to conduct formal meetings. The link
between external decision authority and budgetary support is not totally clear.
Future research should investigate and clarify the relationship between external
decision authority and other key study variables.

Research Issues and Questions Post hoc hierarchical regression analysis and
analysis of bivariate correlational data revealed that team decision authority relates

directly to cross-functional sourcing team effectiveness. While the amount of

242

variance explained by the team authority variables was not as great as other critical
variables, the finding that team authority significantly related to team performance
is important. In addition, the data indicated that greater team empowerment relates
directly to other positive team outcomes. This research established that a significant
relationship involving authority and effectiveness exists. Future research can focus
on establishing a causal linkage between team empowerment and effectiveness.
Also, future research can perhaps manipulate the various authority dimensions to
identify the effect that authority may potentially have on team performance.

The results of this study further suggested that a hierarchy of authority or
empowerment exists. The lowest level of authority consists of scheduling authority,
which almost all teams in this study had. The next levels consist of member selection
authority, internal process authority, and eventually external decision authority. Both .
external raters and team members agreed that teams with higher internal process and
external decision authority demonstrated greater effectiveness. Further research into
this area can focus in detail on the proposition that a hierarchy of team authority
exists.

Evidence gathered during the study (both quantitative and qualitative)
suggested that firms likely initiate the use of cross-functional teams by establishing
teams with lower authority levels. As organizations gain experience with the cross-
functional team approach, team empowerment likely increases. Organizations that

routinely used cross-functional teams appeared to provide team empowerment at

243

higher levels. The evolution of cross-functional team empowerment within

organizations offers excellent future research opportunities.

Team Effort

The finding that team effort related directly to externally and team-reported
cross-functional sourcing team effectiveness is not surprising. It is difficult to
conceive of an effective team that does not exert an adequate effort towards its task
assignment. However, the creation and use of a cross-functional sourcing team does
not guarantee that team members will exert the effort required to complete
successfully team assignments.

Several previous models of team effectiveness have theorized that a team's
effort relates directly to performance (Hackman 1987, Cummins 1981). Hackman's
normative model of group effectiveness includes team effort as one of three variables
critically impacting performance. According to Hackman, effort does not result
simply from team formation. He argued that a team will work especially hard if it
has the autonomy to pursue a meaningful task assignment with visible performance
outcomes. Furthermore, a team's task should require members to use higher level
skills with regular feedback about team performance. If a team's task satisfies these
criteria, members will regard the assignment as important and feel collectively
responsible for the team's output.

A meaningful task assignment can provide a built-in motivation for a team to

work hard and perform well (Hackman 1987). This study-found, for example, that

244

teams with higher internal authority exhibited, on average, greater effort compared
to teams with lower internal authority. This supports one of Hackman's key task
conditions (autonomy) required for promoting team effort.

It is not enough to conclude that members should work harder or that teams
only require a meaningful task assignment for members to demonstrate the necessary
effort and commitment. A further understanding of the. factors relating to team
effort is necessary to create an environment which actively promotes team member
participation. The data indicated that a meaningful task assignment appeared to be
a necessary but not sufficient condition for promoting team effort. Data analysis
indicated that other factors relate directly to the effort a team puts forth on its
assignment.

Research Issues and Questions External raters and team members agreed
that a critical relationship existed between team effort and cross-functional sourcing
team effectiveness. The testing of the a priori hypothesis focusing on team effort
found that only leadership and cohesiveness related significantly to team and
externally-reported team effort. As mentioned earlier, analysis of study data
indicated that certain factors may work against cross-functional sourcing team
members putting forth the effort required to support a team's task.

By extending work conducted by Hackman (1987), we can propose that one
way to promote member contribution is to design task assignments which encourage
team member participation. This means the team's task (1) requires members to use

a variety of higher level skills, (2) is meaningful with visible outcomes, (3) provides

245

the team with autonomy and ownership, and (4) has significant positive consequences
on others within the organization (Hackman 1987). Foushee (1984) concluded that
team builders must structure tasks in a way that require coordinated performance
among all team members. This will require increased member effort by necessity
and produce real organizational change.

Further data analysis revealed that an additional set of variables other than
those proposed a priori relate directly to team effort. Future research can investigate
whether manipulating these variables can promote team effort and member
involvement at a higher level. The remainder of this section discusses the variables
that the data indicated related significantly to team effort which offer a possible basis
for future research.

The data revealed that external raters and team members agreed that teams
with greater internal decision authority exerted, on average, greater effort on team
assignments. Previous discussion already addressed the critical relationship between
team authority and effectiveness. The finding of a link between internal decision
authority and team effort is also important. Hackman (1987) argued strongly that
providing the authority to control a team's internal processes can motivate a team to
purse actively its task assignment. A critical proposition in his group effectiveness
model, for example, is that a team will work especially hard on its assignment if it
has the autonomy to decide how to perform its work. The relationship between
external decision authority and team effort was not as strong as the link between

internal decision authority and effort.

246

Supplier involvement also related significantly to the effort a team put forth
on its assignments. In particular, team members, and to a lesser extent external
raters, indicated that a direct relationship existed between team effort and suppliers
who provided support during formal team meetings. Perhaps the presence of
external suppliers encourages team members to work harder and commit themselves
to the team's task assignment. The presence of suppliers at team meetings
demonstrated a positive statistical relationship with team effort.

A critical relationship also existed between effective leadership and team
effort. In fact, effective leadership accounted for the greatest proportion of variance
of any variable for team-reported effort. The team leadership section discusses the
critical issues and implications concerning leadership in greater depth.

The data also suggested that a positive relationship existed between the
amount of time a member actually committed to a team and team-reported effort.
While the amount of time a member actually commits is partly a function of a team's
task or the member's role on the team, it is possible the actual time committed by
members reflects a team's effort. Data analysis also indicated that members whose
performance appraisal included a larger component for team participation and/or
performance committed a greater amount of time to team assignments. Future
research can investigate if the use of a performance evaluation and reward system
can positively impact team effort and performance.

External raters and team members agreed that the availability of the

combined organizational resource scale demonstrated a direct relationship with team

247

effort. Furthermore, the data indicated that certain resource categories relate to
team effort at a higher level. These resources included (1) time availability, (2) team
member task preparation, (3) executive management commitment, and (4) budgetary
support. Future research can evaluate whether providing access to certain resources
can create an environment which causes active team participation.

A strong relationship existed between time availability and externally and
team-reported effort. Logically, it would appear difficult to put forth a greater effort
under conditions of time constraint. Understanding how organizations can provide
the time required for team assignments provides another area of research
opportunity.

As a resource category, executive management commitment also related
directly to cross-functional sourcing team effort. Additional investigation into this
area should focus on how executive management can demonstrate its commitment
to the cross-functional team process. Also, can certain actions taken by executive
management make teams more aware of the importance of the cross-functional
sourcing team process? Will these actions have a positive effect on a team's effort
and commitment? How critical is it for executive managers from nonpurchasing
functions to demonstrate support and commitment to the cross-functional process?
Understanding the role of executive management within the team process is another
area of future research opportunity.

Team member task preparation also related strongly to team effort. Task

preparation is the personal preparation and experience of team members, through

248

previous education, formal company training, and relevant job experience, required
to perform the team's assignments (Peter's and O'Connor 1985). Previous research
concluded that members with the proper task preparation will be more active on a
team and make a greater contribution to the team's attempts to complete its task
(Shaw 1981). Future research should investigate more thoroughly the possible causal
relationship between member task preparation and cross-functional sourcing team

effort.

Team Leadership

Likert (1960) noted that while different. characteristics of highly effective
groups exist, the role of effective leadership is particularly important. Stodgill (1981)
argued that an effective team leader can positively influence a team's effort,
cohesion, goal selection, and goal attainment. The team leader provides group
stimulation, discourages member complacency, provides a source of enthusiasm for
team goals, and sees to it that the task of the team is important and challenging.
Only a formal team leader can perform many of the responsibilities associated with
leadership (Likert 1960). Extending this discussion to cross-functional sourcing teams
highlights the importance of selecting competent persons for team leadership
positions during the team building process. This study found a strong relationship
between leadership and team effectiveness. Study results also indicated a critical
linkage between leadership and (1) team effort and (2) organizational resource

availability.

249

This research assessed three dimensions of team leadership: task-oriented
leadership, relationship-oriented leadership, and resource/commitment attainment.
Task-oriented and relationship-oriented leadership assessment used a modified
version of the Leader Behavior Description Questionnaire (LBDQ). The assessment
of resource/commitment attainment used a nine-item scale to evaluate a team
leader's performance against specific leadership tasks.‘

The data indicated that team leadership probably does not contribute to the
highest potential level of cross-functional sourcing team effectiveness. This study
asked members to compare their team leader to other organizational leaders with
whom they have familiarity or knowledge. Overall, members ranked their team
leader as follows: (1) 1.5 percent of all team members ranked their team leader in
the bottom 10 percent of all leaders with whom they have familiarity or knowledge,
(2) 7.3 percent ranked their leader in the bottom third, (3) 32.2 percent ranked their
leader in the middle, (4) 38.7 percent ranked their leader in the top third, and (5)
20.3 percent ranked their leader in the top ten percent. More revealing, almost 70
percent of team leaders received an effectiveness score of less than five on a seven

point scale with no leader receiving a "totally effective" rating. Although team

 

The nine-item resource/commitment attainment scale appeared in Chapter 4. To
reiterate, the scale assesses the team leader's ability to (1) secure member
involvement, (2) manage internal conflict, (3) maintain team focus, (4) secure
required resources, (5) prevent team domination by a member or function, (6) deal
with obstacles, (7) coordinate multiple tasks, (8) establish specific team goals, and
(9) clarify and define each member's role.

250

leadership appears critical to team effort and success, study results indicated the
impact of leadership has probably not been as strong as it could be.

Of the three leadership dimensions evaluated during this study,
resou roe/commitment attainment demonstrated the strongest overall relationship with
team-reported and externally-reported effectiveness. Table 56 compares the
correlations between the four dimensions of team effectiveness and the three
leadership dimensions. Some inconsistency exists between the importance of task-
oriented and relationship-oriented leadership styles. Task-oriented leadership
demonstrated a higher relationship with externally-reported effectiveness compared
to relationship-oriented leadership. Team members indicated that relationship-
oriented leadership, however, relates slightly higher to team-reported effectiveness.
This discrepancy resulted from the different perceptual ratings that external raters
and team members had concerning team effectiveness. Both team members and
external members agreed, however, that task-oriented and relationship-oriented
leadership relates directly to team effectiveness.

Referring to a discussion by McGrath (1964), effective and task-oriented
leadership should satisfy a team's requirements when the team pursues a structured
task. This assumes, of course, that interpersonal relationships within the team
remain positive. Cross-functional sourcing teams with loosely structured task
assignments will most likely benefit from a relationship-oriented leadership style.

Members and external raters agreed that the ability of the team leader relates

directly to organizational resource availability. The strongest relationship existed

251

Table 56
Correlations Between Effectiveness Dimensions
and Cross-Functional Sourcing Team Leadership

Externally-Reported Effectiveness

Leadership Dimensions:

Task- Relationship-
Performance Dimensions: Oriented Oriented
Process Performance .33" .18
Innovation Performance .17 .03
Time Reduction Performance .12 .20*
Supply Base Management .27" .15

Performance

Team-Reported Effectiveness

Leadership Dimensions:

Task- Relationship-
Performance Dimensions: Oriented Oriented
Process Performance .44” ' .54“
Innovation Performance .36“ .35"
Time Reduction Performance .18 .36”
Supply Base Management .27** .42M

Performance

* Significant at .05 level
** Significant at .01 level

Resource/
Commitment
Attainment

.30"
.05

.33"
.27"

Resource/
Commitment
Attainment

.7o**
.43**
.47**
.50**

252

between a leader's resource/commitment attainment ability and team budgetary
support. Future research can investigate more thoroughly the relationship between
a leader's ability to secure resources and member commitment and the team's access
to critical team resources.

Team members and external raters also agreed that teams with leaders adept
at resource/commitment attainment demonstrated greater effort. Furthermore, task-
oriented leadership related directly to team effort at a level higher than relationship-
oriented leadership. As George (1977) has argued, the cross-functional team leader
must be skillful at bringing together members from a variety of functional areas to
pursue a team's assignment. This study found an important relationship between
task-oriented leadership and cross-functional sourcing team effort.

The data also indicated that a direct relationship existed between relationship-
oriented and resource/commitment attainment and a team's ability to coordinate its
work activity. This refers to the coordination of work activity between team
members, with others external to the team, and with key suppliers. Relationship-
oriented and resource/commitment attainment also related directly to higher team
satisfaction with (1) the usefulness of team meetings, (2) the quality of information
exchange between team members, between the team and others external to the team,
and between the team and key suppliers, and (3) the methods of communication
between team members. Relationship-oriented and resource/commitment attainment
also related directly to higher levels of open and accurate within-team

communication. This finding supported Halpin and Winer's (1957) argument that

253

a primary role of relationship-oriented leadership is to establish positive patterns and
channels of communication.

The results of this study indicated that leadership can potentially impact many
different aspects of the cross-functional sourcing team process. The results also
indicated that both task-oriented and relationship-oriented leadership styles relate
directly to various aspects of team interaction and performance. Tables 57-59
present the bivariate correlations between the different dimensions of team
leadership and key process and performance team indicators.

The positive relationship between team leadership and team performance in
this study was clear. While past research demonstrated the importance of leadership
for nonsourcing teams, this research established a strong linkage between leadership
and cross-functional sourcing team effectiveness. This research also demonstrated
important relationships between team leadership and other factors, particularly team
effort, resource availability, the coordination of work activity, and communication
satisfaction.

Research Issues and Questions The finding that team leadership related
directly to team performance encourages additional research questions in this area.
First, a key question concerns which organizational function(s) should lead a cross-
functional sourcing team? While the teams in this study included members from at
least fifteen separate organizational functions, over 55 percent of team leaders were
from purchasing. The most common functional areas of nonpurchasing leadership

included engineering and operations. Engineering representation included product,

254

Table 57
Correlations Between Team Leadership
and Cross-Functional Sourcing Team Effort

Leadership Dimensions:

Task- Relationship-
Oriented Oriented
Externally-Rated Team Effort .35 .27
Team-Reported Team Effort .52 .32

Note: All correlations significant at .01 level

Table 58
Correlations Between Team Leadership
and Coordination of Work Activity

Leadership Dimensions:

Task- Relationship-
Coordination: Oriented Oriented
Among all team members .29" .39"
With individuals or groups -.05 .21“
outside the team
With key suppliers .07 .23*

4:

Significant at .05 level
Significant at .01 level

*4!

Resource/
Commitment
Attainment

.37
.61

Resource/
Commitment
Attainment

.54”
.24*

.28”

255

Table 59
Correlations Between Team Leadership
and Team Communication Satisfaction

Leadership Dimensions:

Resource/
Task- Relationship- Canmitment
Satisfaction Area: Oriented Oriented Attainment
Usefulness of formal team meetings .46 .55 .54
Quality of information exchange .40 .54 .56
between team members -
Methods of communication between .36 .55 .59
team members
Quality of information exchange .26 .44 .48

between the team and others
outside the team

Quality of information exchange .28 .39 .48
between the team and key suppliers

Note: All correlations significant at the .01 level

256

process, and quality engineers. The data indicated that teams with greater functional
diversity were more likely, on average, to have nonpurchasing team leadership. As
team functional diversity increased, the probability that a nonpurchasing member
assumed the position of a team leader appeared to also increase.

An analysis of team leadership indicated no relationship between purchasing
leadership and cross-functional sourcing team effectiveness. The results of this study
indicated no loss of performance in any key area when the leader was from a
nonpurchasing function. This finding was consistent across team-reported and
externally-reported effectiveness.

The finding that purchasing led a majority of the teams in this study raises
critical research issues. For example, does the probability that purchasing will lead
a cross-functional sourcing team create a perception that these teams are purchasing
controlled activities relying simply on cross-functional contribution? Presumably, a
challenge facing team builders is to develop commitment by functional members to
the team the same as members commit to their full-time position. This assumes, of
course, the cross-functional sourcing team requires the commitment of each member
at that level. Can assigning or rotating leadership among different functions promote
greater interaction and commitment by team members?

An interesting finding in this study was that members committed a higher
proportion of their professional time to their team when the team had nonpurchasing
leadership. While speculative, this could indicate greater commitment to the cross-

functional sourcing team process when nonpurchasing members are able to assume

257

leadership. Future research might verify this finding and perhaps eliminate other
possible causes for the relationship between nonpurchasing leadership and team
member time commitment.

A qualitative observation made during the study is that team leadership was
not necessarily a formal position on a cross-functional sourcing team. In particular,
smaller teams (i.e. three to four members) sometimes used shared leadership, rotated
leadership responsibilities, or had no formal leadership position. The data indicated
that smaller teams in this study were likely to be walking distance apart or co-located
together. Geographic proximity may minimize the need for a leader to act as a
formal coordinator of team activities for smaller teams. Furthermore, the data
revealed an inverse relationship between team size and member stability. Smaller
team membership was more likely to remain stable over time compared to the
membership of larger teams. While speculative, it is possible that team member
stability can further reduce the need fora formal leader to coordinate a smaller
team's activities over time. Future research can assess the relationship between
formal team leadership and team size.

Future research questions also concern the methods team builders use when
structuring the formal cross-functional sourcing team leadership position. While it
appeared the most common method for leadership selection in this study was the
selection of a single member to act as the team leader, no data are available

concerning the leadership selection process. For example, did management select

258

the leader, the team select its leader, or the leader emerge as the team pursued its
assignment?

Further research in this area should also evaluate the value of shared
leadership. Does the use of shared leadership promote representation and
commitment from different functional areas to the team process? Does this
approach support a balanced approach towards the increased flow of information and
exchange between functions? Does the selection of a coordinator or facilitator to
manage routine team activities facilitate the team process, particularly for larger
teams? Furthering our understanding of role and importance of cross-functional
sourcing team leadership provides valuable research opportunities.

Team leadership is important for another reason. The data indicated that the
team leader committed a significantly greater amount of time directly to team
activities compared to the average team member. For example, only 13 percent of
team leaders indicated they committed 10 percent of their time or less to the team.
Almost 40 percent of team members committed 10 percent of their time or less to
the team. Team leaders committed, on average, 25 percent of their professional time
to the cross-functional sourcing team. Team members committed, on average, 15
percent of their professional time to the team. This indicated the team leader,
rightly or wrongly, committed a greater amount of time to team tasks compared to
team members.

Part of the time differential between leaders and members could be due to

a majority of team leaders being from purchasing. Cross-functional sourcing team

259

tasks are naturally part of the purchasing function. We therefore expect that
purchasing team leaders are able to commit a higher percentage of their professional
time to team-related assignments. Regardless, greater understanding is required
concerning time differential between team leaders and members. It is possible that
team leaders are assuming additional responsibilities that team members should
actually be responsible. Do team leaders legitimately commit greater time because
of the requirements of the leadership role? Or is it because of a lack of commitment
by team members to the cross-functional sourcing team process? The finding that
team leaders committed more time to team activities than team members suggested
opportunities for understanding the reasons for this time differential.

Critical research questions concerning cross-functional sourcing team
leadership remain unresolved. The discussion of research extensions proposes future

opportunities in this critical area.

DISCUSSION OF DEMOGRAPHIC INDICATORS

The results of this study indicated that demographic variables played a minor
or totally insignificant role in cross-functional sourcing team effectiveness. While
certain demographic variables related to externally-reported team effectiveness, the
overall strength of these relationships was weak. The probability of chance alone
explained the number of significant relationships between demographic indicators

and team effectiveness.

260

The following sections briefly discuss demographic team indicators and
relationships involving these variables. The demographic variables discussed include
(1) team size, (2) the functional diversity of the team, (3) the cross-functional and
professional experience of team members, (4) geographic diversity, (5) education, (6)

within-team age dispersion, and (7) team longevity.

Team Size

Team size demonstrated no significant relationship with any indicator of
cross-functional sourcing team performance. A slight relationship did exist between
externally-reported team innovation performance and team size. Other relationships
that existed with team size were not unexpected. For example, larger teams reported -
fewer impromptu team meetings, less stability of team membership, and the presence
of task-oriented leadership.

While no direct relationship existed between coordination/communication
variables and team size, there was an indication that larger teams in this study may
experience greater coordination problems. As team size increased, team leaders
tended to agree with the statement "the team has too many members to effectively
coordinate and accomplish its assignments." Leaders of larger teams tended to
disagree with the statement "team size is just right to effectively coordinate and
accomplish team assignments." Furthermore, larger teams were often geographically
diverse. This may also create additional coordination and communication challenges

for team leaders and members.

261

Previous research has concluded that team size is often a function of a team's
task (McGrath 1984). Future research can investigate the conditions under which
firms create larger (i.e. ten or more members) or smaller cross-functional sourcing
teams. A number of conditions may promote the creation of a larger team. These
conditions include (1) a large-scale task assignment, (2) a task assignment that affects
multiple departments or functions, or (3) a major decision that requires buy-in from
different locations and/or departments.

Most team leaders in this study reported satisfaction with team size and that
few coordination problems existed. This indicated that team builders may have
correctly matched team size and task requirements. The data indicated that within-
team coordination problems, when they existed, appeared to result more from

functional diversity than team size.

Functional Diversity

The degree of functional diversity demonstrated no direct relationship with
team-reported or externally-reported effectiveness. Functional diversity refers to the
number of unique functions represented on a cross-functional sourcing team. The

data indicated that a number of interesting relationships do exist involving functional

diversity:

r Teams with greater functional diversity indicated greater formal and
informal frequency of within-team communication and external-team
frequency of communication.

° A trade-off existed between functional diversity and satisfaction with

a team's ability to coordinate its work activity. Teams with greater

262

functional diversity indicated, on average, slightly greater problems
coordinating the team's work activity.

' Teams with greater functional diversity indicated they commit more of
their professional time to the team's activities over a typical time
period. However, teams with greater functional diversity also had a
greater chance of having a nonpurchasing team leader. The presence
of nonpurchasing leadership could explain increased team time
commitment.

Functional diversity offers the opportunity to bring together team members
with different task perspectives. Team members reported the most important benefit
from cross-functional interaction was "greater knowledge and skill brought together
at one time." The challenge for team builders is to assemble members with a
sufficient distribution of skills capable of addressing a required task (McGrath 1964).

Creating a functionally diverse team may be one way to assemble a
heterogeneous team with a sufficient distribution of task-relevant abilities. Future
research opportunities exist to evaluate the topic of functional diversity in greater
detail. Specifically, does a higher level of functional diversity actually promote the

benefits sought from cross-functional team interaction? Also, how do firms

determine which functions are necessary to a team's task assignment?

Professional and Cross-Functional Team Experience
The average professional/functional experience of a team's members
demonstrated no meaningful relationship with team effectiveness. Furthermore,

previous cross-functional team experience did not relate to team effectiveness.

263

External raters and team members agreed that professional and cross-
functional experience variables played a minor role in this study. The statistically
significant correlations between professional experience variables and team effort and
effectiveness indicators were slightly higher than expected by chance. However, team
members and external raters did not agree on which relationships were significant.
A lack of agreement between data sources and the probability of chance findings
revealed the lack of importance concerning the professional and cross-functional

team experience indicators evaluated during this study.

Geographic Diversity

Geographic distance between team members had no relationship with team
effectiveness. In fact, team members indicated "team members not located physically
close enough to allow effective team communication and interaction" was the lowest
rated drawback to the cross-functional sourcing team approach. Geographically
closer teams did indicate, however, greater frequency of team meetings and slightly
higher within-team communication compared to teams whose members were located
further apart.

It was not unusual for the teams in this study to include members from
geographically disperse locations. Almost 40 percent of participating teams had team
members located flying distance apart. Another 20 percent of the teams had

members located driving distance apart. The need to coordinate and communicate

264

across geographic distances is a factor that makes cross-functional sourcing teams

unique compared to other types of organizational teams.

Team Education

As a demographic indicator, a team's aggregate education showed no
relationship with team or externally-reported effectiveness. There existed minimal
differentiation between teams across the education variable. Over 71 percent of
team members were college graduates, 25 percent had some college or technical
school training, and only 4 percent had a high school diploma or less.

An inverse relationship existed between team age and education. Teams with
a lower average age generally had a higher aggregate level of education. This may
reflect the emphasis on continuing education for new and/or younger employees that
many firms stress. Secondly, nonpurchasing team members had, on average, a higher
educational level compared to purchasing personnel. This is partly a function of the
technical representation on cross-functional sourcing teams along with~ the lower
educational requirements historically associated with members of the purchasing

profession.

Within-Team Age Dispersion
No significant relationships existed between within-team age dispersion and
team effectiveness. There was a slight relationship between age dispersion and team

communication, however. Teams with greater age dispersion indicated slightly lower

265

formal and informal within-team communication. Overall, within-team age
dispersion demonstrated little or no relationship with other variables included in the

study.

Team Longevity

The amount of time a cross-functional sourcing team has worked together
demonstrated no significant relationship with team performance. This does not mean
that a team does not become more or less effective as it moves through a life cycle.
The cross-sectional nature of this study prohibited an analysis of team performance
over time. Furthermore, project oriented teams may only exist for four to six
months. Commodity management teams, on the other hand, may work together over
a period of years. Differences in team task assignments make comparisons based on
team longevity difficult.

A positive relationship did exist between team longevity and personal team
cohesiveness. The data indicated that as teams worked together over an extended
period of time, personal cohesiveness between members seemed to develop at a
higher level. Not unexpectedly, an inverse relationship existed between how long the
team has worked together and the stability of team membership.

A moderate relationship existed between team longevity and the degree to
which a cross-functional sourcing team relied on supplier support. This presents
interesting research questions. For example, do teams responsible for evaluating and

selecting suppliers (which was the primary task of sourcing teams in this study)

266

initially not rely on direct supplier support? Furthermore, once a team selects its
suppliers, should the team's focus shift from supplier selection to supply base
management and improvement? If it does shift, how should the team rely on
suppliers to support improvement efforts? The interaction between suppliers and

cross-functional sourcing teams over time offers interesting research possibilities.

DISCUSSION OF OTHER STUDY VARIABLES
Performance Evaluation and Reward Systems

No major research conclusions resulted from assessing the performance
evaluation and reward systems at participating firms. The performance benefit
derived from a properly designed evaluation and reward system, however, has been
the subject of recent empirical investigation (Pritchard et. al. 1988). Additionally,
a content analysis of participants' open-ended responses indicated the development
of a system which evaluated member contribution to a team's performance was a
frequently cited suggestion for improving the cross-functional sourcing team process.
Furthermore, data contained in Table 60 indicates that the performance evaluation
and reward systems at some participating firms may actually inhibit rather than
promote participation by team members. This topic warrants extended discussion
beyond that indicated by the results of this study.

The firms participating in this study most often (1) provided no formal
performance assessment or reward for participating on a cross-functional sourcing

team or (2) relied on a team member's immediate manager to evaluate a member's

267

Table 60
Performance Evaluation and Rewards

Percent in Percent in

Agreement Neutral Disag'eernent

Provides organizational rewards for devoting 15.2% 14.2% 70.6%
time and resources to team assignments

The performance appraisal system
used to evaluate my work:

Rewards superior total team effort 15.0% 16.3% 68.8%
and performance

Rewards superior individual team 15.5% 17.0% 67.5%
member effort and contribution

Penalizes a lax individual effort 12.2% 29.8% 58.0%
on team assignments

Is actually a barrier to participating 52.3% 30.3% 17.4%
on the sourcing team

Stresses my personal participation on 34.5% 17.3% 48.2%
the team very much . .
Stresses individual member effort over 44.3% 30.0% 25.7%
the team's combined performance

Encourages a total commitment to 39.0% 19.2% 41.8%

the team's assignments

N = 620 team members

268

individual contribution to a team. Previous research has warned of the negative
effects of recognizing and rewarding individual contribution rather than combined
team performance (McGrath 1964, Lawler 1981, Hackman and Oldham 1980).
' Hackman and Oldham (1980) warned of the consequences that can result when
organizations provide rewards differentially to individual team members based on
managerial perception of performance contribution. These consequences include
members perceiving they have been set off against each other in a competition for
rewards or members finding their work impaired by the fear of evaluation and
reward inequities among members (Hackman and Oldham 1980). Deming argued
that a continued emphasis on individual performance merit ratings (versus group or
team evaluation ratings) promotes short-term performance, does not support
teamwork, and nourishes rivalry and politics between team members and functional
groups.

The data revealed that few team members actually received a formal
evaluation based on the team's collective output and performance. :In fact, a majority
of team members indicated they generally did not receive any formal evaluation or
recognition for individual contribution to a team. In this sense, a cross-functional
sourcing team assignment is an unrecognized performance item. For example, 57
percent of team members indicated they received no formal evaluation for
participating on team activities. Tables 61 and 62 detail who has responsibility for
evaluating individual contribution to a team and how management determines a

performance evaluation. The uneven distribution of responses indicated that firms

269
Table 61

Responsibility For Evaluating Performance Contribution

52.5%

32.6%

5.2%

.8%

6.4%

2.5%

Do not receive an individual evaluation for participating on
the team

Immediate manager of the team member evaluates individual
contribution to the team

Team leader and/or other team members evaluate each
member's performance contribution

A single manager or executive who oversees the team
evaluates individual performance contribution

Individual team member evaluation is a joint activity
involving input from a number of personnel internal and

external to the team

Other

Table 62

Determination of Performance Evaluation For Team Participation

57.0%

7.2%

14.3%

19.1%

2.3%

Not Formally Assessed

Based solely on individual contribution and involves no
combined team performance rating

Based solely on the team's performance as a combined group
and involves no combined team performance rating

Based jointly on individual contribution to the team and
combined total team performance

Other

270

in this study generally did not include cross-functional sourcing team participation
as part of an employee's formal job assessment. This lack of distribution limited the
ability to compare teams based on responses to performance evaluation and reward
system measurement items.

The data did not support the hypothesized link between a firm's performance
evaluation and reward system and team effort. As mentioned, part of this resulted
from the unimodal distribution of responses concerning formal performance
evaluation. There was still evidence, however, of an indirect link between
performance evaluation and team - effort. This link occurred . between the
performance evaluation and reward system and the amount of time a team member
committed, on average, to team assignments.

The amount of time a team member committed to team activities relates
directly to (1) the degree to which organizational rewards were dependent upon the
member's individual contribution to team activities and/or overall team performance,
(2) the percent of the members' total performance appraisal or evaluation based on
team participation and/or the team's overall performance, and (3) the degree to
which a member's performance actually affected the member's overall performance
evaluation. These findings indicated that a team member whose performance
evaluation included a larger component for team participation committed more time,
on average, to team activities. In turn, the data revealed that the amount of time a
team committed to team activities relates directly to team effort. Team effort, in

turn, is a major predictor of team success.

271

Other data provided additional insight about performance evaluation and
reward systems. Teams with higher external decision authority indicated, on average,
the presence of an evaluation system that promoted team member participation. An
evaluation system that motivated team members to participate is one that
acknowledges or evaluates individual performance on the team, stresses personal
team member participation (either through formal review or rewards), and
encourages individual and team effort at level that supports meeting team goals and
objectives. It appeared that firms that have granted teams external decision
empowerment also tended to establish performance. evaluation and: reward systems
that motivate and reward team member participation. The evolution of and the
linkage between team decision authority and performance evaluation and reward
systems is a topic of future research interest.

Another observation from the data. was that a team whose members indicated
greater cross-functional team experience generally indicated the presence of a system
that tied organizational rewards (salary increases, promotions, bonuses, etc.) directly
to individual contribution and/or overall team performance. Although highly
speculative, previous team experience may be an indicator of a firm's historical usage
or degree of familiarity with the cross-functional team approach. Just as
empowerment may evolve over time as firms become increasingly familiar with the
use of teams, it also appeared that organizations may develop team evaluation and
reward systems over time. Understanding the development of team performance

evaluation and reward systems within firms also offers future research opportunities.

272

Research Issues and Questions The presence or absence of a team
performance and evaluation system raises a number of important research issues.
One major issue is whether a team member's job performance evaluation should
include a component for participating on a cross-functional sourcing team. Findings
from this study indicated that an employee's formal job evaluation should include a
component for team participation.’ Again, a strong relationship existed between the
percent of a member's job evaluation based on team participation and/or team
performance team and the time committed to team activities. Previous research in
this area concluded that team members will exert effort if they expect it to produce
rewards (Cummins 1981).

Two additional research issues include how to evaluate individual and team
performance and who has responsibility for determining team performance ratings.

A first critical research issue focuses on how to create a system that
encourages active cross-functional sourcing team participation. Previous work by
Hackman (1987) concluded that a progressive performance evaluation and reward
system should not base team member evaluation solely on each member's individual
contribution to the team. This can be a challenge for organizations where the -
reward system has traditionally recognized eXcellent individual performers and

ignored collective effort and participation (Hackman 1987).

 

5

This assumes that organizations evaluate employee job performance. Almost all
participants in this study indicated they received a performance evaluation.

273

Other researchers have emphasized the use of team measures as the, primary
basis for performance evaluation. Hackman and Oldham (1980) argued that when
organizations design work for teams, then rewards should be contingent on the
performance of the combined team and not the individual team member. A system
that recognizes and rewards excellent team performance can complement the
motivational incentives built directly into a team's task assignment. Pearce and
Ravlin (1987), however, have concluded that a mix of individual and team rewards
will prove most effective. Evaluating individual contribution helps secure a member's
personal commitment, effort, and involvement to the team. They further argued that
a team-based performance evaluation focusing on team goals and objectives supports
working together as a collective unit. A combination of individual and team
evaluation and rewards can establish commitment to individual task responsibilities
and the team's goals (Pearce and Ravlin 1987). Future research can address whether
firms should avoid focusing on individual contribution and instead move towards a
team-based evaluation or a combination of individual and team-based performance
measures.

Another research issue concerns who should have responsibility for evaluating
team effort and performance. Currently, that data indicated that if an employee
participated on a team and received an evaluation for participation, the employee's
immediate manager most likely performed the evaluation. This finding leads directly
to additional research questions. First, does an immediate manager have the

detailed insight into a team's interaction and performance required to evaluate

274

objectively a member's unique contribution? Secondly, what are the potential
consequences if team members put forth similar effort and contribution to a team
yet receive different performance ratings? For example, one member may have a
manager who evaluates contribution generously while another has a manager who
evaluates contribution strictly.

If future research concludes that individual managers evaluating team member
performance is not the best approach to cross-functional sourcing team performance
evaluation, then work in this area can evaluate the value of different performance
evaluation options. One option involves evaluation by internal team members only.
With this approach, the team members (including the team leader) evaluate each
member's professional contribution to the team. This may support evaluation by
those who have the closest insight into each member's unique contribution. This
approach, however, may not be the best approach for arriving at an overall
assessment of team performance. A second option involves a manager or executive
who oversees the team or receives the team's output evaluating individual
contribution and overall team performance. While a single executive may have
insight into overall team performance, he or she may lack the insight required to
assess individual contribution. With a third option, individual team member
evaluation becomes a joint activity involving input from a number of personnel
internal and external to the team. For example, team members can evaluate

individual contribution while an external executive evaluates overall team

275

performance. The third option may be most desirable if a member's performance
evaluation includes individual contribution as well as combined team performance.

Future research in this topic area should focus on whether firms can use an
evaluation and reward system to promote increased effort and effectiveness.
Another issue concerns how firms can develop performance evaluation systems that
promote team effort and effectiveness.

Research into cross-functional team performance and evaluation systems by
Ancona and Caldwell (1990) concluded that changing an obsolete system as. a
response to an inorease in the use of teams is clearly reactive. Instead, firms
promoting the active use of cross-functional teams must take a proactive approach.
Ancona and Caldwell (1990) maintained this requires a firm to redesign its
evaluation system: and policies to encourage cross-functional participation and to
guarantee that team members receive recognition and reward for active participation
and performance. They concluded firms must review their performance appraisal,
salary, and promotion policies to make sure they recognize and promote performance
and cooperation within and between teams. The work and conclusions by Ancona
and Caldwell may have application to future research concerning performance

evaluation and reward systems for cross-functional sourcing teams.

Team Cohesiveness
The a priori hypotheses proposed that a team's cohesiveness relates directly

to team communication and effectiveness. While demonstrating strong bivariate

276

correlations with team effectiveness, hierarchical analysis indicated that cohesiveness
actually explains minimal total effectiveness variance in this study. This is partly a
function of placing the cohesiveness variable last in the order of entry of independent
variables. However, this is the proper placement for this variable. For example, the
data revealed that team longevity relates positively to the personal dimension of team
cohesiveness. This indicated that team cohesiveness may evolve over time as
members become increasingly familiar and/or comfortable with each other.

Previous research has assessed the effect of cohesiveness on team
communication, effort, and effectiveness (Pearce and Ravlin 1987, Hoffman 1979,
Shaw 1981). This research provided minimal new insight about the relationship
between cohesiveness and team performance. Part of this is due to the unusually
high correlation between cohesiveness and other self-reported perceptual measures.
Teams indicated generally consistent (not necessarily high) ratings across all three
self-reported measures. . For. example, the correlation between professional
cohesiveness and open communication is .71. The correlation between professional
cohesiveness and member-reported effort is .78. Research conclusions relying on
these self-reported measures would be questionable.

There is a practical reason that team cohesiveness may not be the powerful
predictor or influence on team effectiveness as originally hypothesized. Almost 65
percent of the teams in this study indicated they met formally monthly, quarterly, or
not at all (27.3 percent, 27.7 percent, and 9.3 percent respectively). The average

frequency of formal meetings across the study is monthly. Given that the average

277

team in this study has worked together for 15 months, a typical team has had 15
formal meetings. One has to question, given the limited number of formal team
meetings, the geographic distance between many team members, and the limited
amount of time a typical team member committed to the cross-functional sourcing
team, how much of an impact can team cohesiveness actually have on team
performance?

Previous research has concluded that team builders should select members
capable of working together as a collective unit (Pearce and Ravlin 1987). However,
the ability to achieve performance gains through synergistic interaction may not be
as likely with this type of team as with other types of business teams characterized

by intensive interaction.

Team Communication

Communication fi'equency between members of a cross-functional sourcing
team exhibited no direct relationship with externally-reported or team-reported
effectiveness. This does not necessarily mean that communication frequency does
not relate to cross-functional sourcing team effectiveness. The frequency of
communication is partly a function of a team's task assignment. Certain tasks require
greater frequency of within and external team communication. Unfortunately, this
research was unable to segregate teams based on the intensity or communication
requirements of the task assignment. As discussed in the research extensions section,

the topic of communication frequency offers excellent future research opportunities.

278

While study results concerning communication frequency are inconclusive, other field
research has concluded that frequency of communication is a healthy aspect of a
cross-functional environment (Ancona and Caldwell 1990). Thus, it may be
beneficial to understand the factors that relate directly to communication frequency
within a cross-functional sourcing team.

The data indicated that a number of study variables relate directly to within-
team communication frequency. First, teams with greater functional diversity
indicated greater internal team communication frequency. Communication
frequency, however, demonstrated no relationship with team size. A direct
relationship also existed between geographic distance and internal communication
frequency. Team members located geographically closer indicated they
communicated more frequently compared to team members located greater distances
apart. Next, a team whose members had a wider dispersion of member ages
communicated less than teams whose ages were closer. Furthermore, as the average .
age of a team's members increased, the frequency of within-team communication
decreased. As expected, members that committed more time to team activities
indicated greater communication frequency within the team. Finally, a performance
appraisal system that included a larger component for team participation and/or team
performance relates directly to within-team communication frequency. Further
analysis revealed the variables just discussed combined to explain 41 percent of the

variance associated with within-team communication frequency.

279

The data also indicated that only two study variables explained a significant
amount of variance associated with total extemal-team communication frequency.
These variables included functional diversity (a positive relationship with external
communication frequency) and a performance appraisal system that included a larger
component for team participation and/or team performance (also a direct
relationship with external communication frequency). These variables combined to
explain only one-fifth of the total variance associated with external-team
communication frequency.

Research Issues and Questions Although this research did not reveal any
meaningful relationships between communication fi'equency and cross-functional
sourcing team effectiveness, future important research questions in this area still
remain. The finding that certain variables related directly to communication
frequency provides a basis to structure future research that evaluates cross-functional
sourcing team communication. For example, can team builders manipulate certain
aspects of a team's structure to encourage greater team communication? This
assumes that communication frequency between members is important to some
degree to successful cross-functional sourcing team interaction. Also, should team
builders consider the chemistry between members when forming teams? This
question can focus on the generational relationship between members that may affect
communication frequency. Next, should team builders select a wider range of
functions to support a team's task knowing that functionally diverse teams usually

communicate more frequently? What other variables besides those discussed impact

280

team communication frequency? Finally, how can team builders use performance
evaluation and rewards to promote team communication? Studying the factors that
promote cross-functional team communication may offer interesting future research
opportunities.

The openness and accuracy of team communication, as indicated by self-
reported measures, related directly to externally-reported and team-(reported .
effectiveness. However, the openness and accuracy of team communication variables
were subject to the same limitations as the self-reported team cohesiveness measure.
Almost all team members indicated that open communication existed to some degree
on their team. In fact, 87 percent of the teams in this study indicated some level of
agreement that open within-team communication erdsted. Fewer teams, however,
indicated the presence of accurate within-team communication. Only 50 percent of
the teams indicated that accurate within-team communication existed. Unfortunately,
the results of this study provided minimal insight into the conditions promoting open
and accurate team communication. Understanding the pattern of communication
within cross-functional sourcing teams also offers interesting future research

opportunities.

RESEARCH EXTENSIONS
The discussion throughout this chapter referred to different research issues
not specifically addressed by the data gathered during the study. This section

summarizes a number of key future research opportunities that became apparent

281

throughout the course of this research. This includes research areas which the study
did not stress explicitly or where study results are inconclusive or incomplete.
Certain topics will require in-depth analysis to further our understanding of the
relationship between the research area and cross-functional sourcing team
effectiveness. Furthermore, future research must focus specifically on establishing
the causal relationship between variables that this project was unable to establish
because of design limitations.

Understanding the Organizational Value of Cross-Functional Sourcing Teams
Perhaps the most basic future research question concerns whether there is a strategic
or performance advantage to using cross-functional sourcing teams. Future research
should address explicitly a comparison between a cross-functional sourcing team
approach and traditional approach to similar tasks. There has been a tendency
within the popular literature to accept most any. kind of team as a desirable
organizational structure. However, it is possible that the use of cross-functional
sourcing teams could create operating inefficiencies or reach inferior decisions,
particularly if the teams are poorly structured and managed.

A number of interesting questions arise from an investigation of the
organizational value of cross-functional sourcing teams. For example, is the cross-
functional sourcing team approach more effective than a nonteam approach? Is a
cross-functional sourcing team more effective than simply a sourcing team comprised
of purchasing and materials personnel? Are there other approaches to cross-

functional integration that may effectively accomplish the benefits sought from the

282

team approach? If the team approach does provide advantages over other
organizational approaches, what types of tasks or under what conditions is the team
approach most useful? Furthermore, should cross-functional sourcing teams be
established with finite time fi'ames or established with open-ended time frames?

This study accepted the team approach as an important organizational
structure because previous research indicated that almost 80 percent of-U.S. firms
plan to stress the use of cross-functional sourcing teams throughout the 19905
(Monczka, Trent, and Gay 1992). Whether the cross-functional sourcing team
approach actually provides performance advantages to firms or is better than other
approaches to organizational tasks is still an unresolved, yet important, research
issue.

Understanding the Team Building Process This study accepted each team's
composition and member/leader selection process as a given. Future research should
evaluate the process or approach firms use when forming cross-functional sourcing
teams. Pertinent research questions include understanding the team leader selection _
process as well as the selection of individual members. Also, how do firms decide
which functions to include for membership? How- do firms determine team size?
Can firms develop commitment and effort through the selection process? Should
team members have the opportunity to remove themselves if they cannot support a
team's objectives? A cross-sectional study with multiple firms invariably lacks insight

into important team formation and interaction issues.

283

The relationship between leadership and cross-functional sourcing team
effectiveness is an important research topic. The need to further our understanding
of the role and impact of the team leader supports the development of research
questions. For example, is the process used to select and develop (or not develop)
leadership important to team performance? Should team members or team
managers select the team leader? What is the role of the team leader? Should team
leadership be a rotating responsibility over the life of a team? What is the impact
of the majority of team leaders selected from the purchasing function? How does
this affect nonpurchasing member commitment and effort? What are the benefits
and drawbacks of dual or shared team leadership? Do smaller cross-functional
sourcing teams require a formal leader? Does the leadership position and its
requirements change depending on a team's task assignment? Does the requirements
of the leadership position change throughout the life of a team? Each of these
issues presented themselves during the study. The design of this study, however,
limited our ability to provide insight into these critical questions.

Determining the Potential Impact of Performance Recognition Systems As
mentioned, respected researchers have argued for performance evaluation and
reward systems that recognize and reward collective team performance rather than
individual contribution (Hackman 1987, McGrath 1964, Pearce and Ravlin 1987).
Unfortunately, the firms participating in this research generally did not recognize
cross-functional team participation or evaluate performance at the team level. It was

therefore difficult to assess the impact that a progressive performance evaluation

284

system could have on team effort and performance. While the results of this study
indicate a strong link between a performance appraisal for participating on a team
and the time a member commits to a team, overall findings remain inconclusive.

Future research should examine key questions in this area. For example, what
type of evaluation and reward system, if any, best motivates individual and combined
cross-functional sourcing team effort and performance? Under what conditions does
a performance evaluation and reward system motivate team members? Under what
conditions does a performance and evaluation system inhibit team . member
performance contribution? Can a firm use a pay for performance incentive system
to motivate team members? Is peer evaluation versus an executive management
evaluation beneficial? Should immediate managers evaluate individual performance
contribution to a team? Most firms lack direction in these critical areas. We simply
do not know the potential benefit, if any, of a. system that evaluates and rewards
individual member contribution and superior team performance.

Understanding the Frequency and Pattern of Cross-Functional Sourcing
Team Communication This project failed to demonstrate a link between the
frequency of team communication and cross-functional sourcing team effectiveness.
This does not mean, however, that communication frequency is not a critical
attribute of team success. The frequency of a team's communication is partly a
function of a team's task requirements. For example, one team's task may require
monthly communication between members while another team's task requires weekly

communication. External raters and team members could actually rate the team that

285

communicates monthly as more effective given its task assignment. Future research
should examine the impact of communication frequency, both within and external to
the team, for teams with similar task requirements. The teams in this research, while
all purchasing and materials related, still pursued task assignments that varied widely
in scope and communication frequency requirements.

Perhaps more important is the pattern or content of communication within
and external to the cross-functional sourcing team. Not all internal and external
team communication will be of equal value and quality. For example, a moderate
relationship existed between a performance evaluation that includes team
participation and the frequency of communication with others external to the team.
Why does this relationship exist? Recall that immediate managers are usually
responsible for team member performance when members receive an evaluation. Is
it possible that team members communicate externally with immediate managers
simply to keep managers informed about contributions to the team? While this
represents external communication, it is not communication of the type or quality
that helps a team support its task requirements.

This study, which attempted to equate frequency of communication to
effectiveness, failed to gain insight into the pattern and content of cross-functional
sourcing team communication. This topic, however, warrants further investigation.

Developing Cross-Functional Sourcing Team Performance Indicators An
observation made during this study is that few firms measure team performance

output at a meaningful or detailed level. A valuable research project includes

286

understanding the process that successful firm(s) undergo when developing systems
that measure objectively collective team output. This involves gaining an
understanding of how firms measure team output and individual performance
contribution.

Team measurement systems are critical for firms which currently rely on
individual employee performance measures and the performance measures of
operating units. Operating units include the firm, divisions, plants, production zones,
departments, etc. If a cross-functional team represents a legitimate organizational
structure, then greater understanding about how to measure the performance output ,
of that structure is necessary. Without this understanding, it is difficult to assess a
cross-functional sourcing team's performance contribution, measure performance
improvement that results from team interaction, or measure the performance

contribution of individual members or functions.

RESEARCH LIMITATIONS

While the contribution of this dissertation is significant, all research project
designs present limitations. The following issues represent either limitations or
potential hurdles that confronted the empirical investigation of cross-functional
sourcing team effectiveness.

Externally-Reported Versus Team-Reported Performance Ratings A major
research issue, and one that deserves attention, concerns the use of team-reported

versus externally-reported measures. There was concern that team-reported

287

performance measures would be subject to self-reported response bias. In other
words, team members would evaluate self-reported variables, including team
effectiveness, more highly while an external rater would: report results more
objectively. While reporting bias is an issue with self-reported performance
measures, several concerns also surround the use of externally-reported measures.

The first concern is that external raters were usually not active participants
within the team process. They were moSt often members of executive management
several steps removed from a team's operation. This can limit the external raters'
ability to evaluate team interaction and effectiveness based on direct knowledge or
observations of the team. Secondly, most firms did not measure cross-functional
sourcing team performance and effectiveness at the level of detail presented in the
measuring instrument. Simply stated, external raters may have lacked objective data
(along with the insight) to evaluate team performance. Third, most external raters
evaluated a single team. Because a lack of measurement often means a reliance on
relative (and subjective) rankings between teams, most raters did not have the
opportunity to compare one team against another. In contrast, the average team
member was often a member of several teams simultaneously and indicated
participation on an average of over four cross-functional teams during his or her
professional tenure.

It was also not unusual for the external rater to be the executive responsible
for the team's performance or who received the team's output. While this satisfies

Hackman's (1987) criteria that team output must satisfy the performance expectations

288

of those who receive and review the output, executive ownership and rating of the
team itself can introduce a source of measurement bias. When asked to evaluate
each team on a ten point scale, for example, external raters indicated that teams
average 7 out of 10. Team members, however, indicated teams average only 6.4 out
of 10.
Finally, the researcher did not control the selection of the external rater.
There were instances in which the selected-external rater did not have the ability to
evaluate a team and required replacement. A qualitative observation is that external
raters were sometimes selected to fulfill the research requirement of this position.
' A comparison of externally-reported and team-reported evaluations provides
interesting insights. External raters reported, on average, higher effectiveness across
the four effectiveness dimensions, higher availability of organizational resources, and
greater team effort. Furthermore, external raters rated team performance higher
than team-reported performance 65 percent of the time. To the external rater,
teams worked harder, had access to greater organizational resources, and were more
effective on average than what team members report.
Given the issues concerning team-reported and externally-reported measures,
a balanced perspective included models of team-reported and externally-reported
data. While unable to support this position conclusively, both sets of data presented
opportunities and limitations for analysis depending on the research topic under

consideration.

289

Lack of a Temporal Perspective Although researchers have indicated the
need to observe teams directly as they exist over time (Goodman, Ravlin, and Argote
1986), this research assessed the performance of cross-functional sourcing teams at
a single point in time. Because the research lacked a temporal perspective, this
prohibited making causal inferences. As a result, we can only refer to relationships
between variables rather than the cause and effect between variables. Furthermore,
this research cannot assess the effects of study variables over time. A cross-sectional
study simply does not provide insight into the dynamics of team interaction over
time.

The effect of this limitation on this study should not be underestimated. A
lack of a temporal perspective shifts the study from a discussion of causal effects to
one of relationships between variables in which causal directions are unknown.

Use of Subjective Measurement Scales Another limitation included the risks
associated with self-reported measures. Robinson, Athanasiou, and Head (1978)
have identified a number of concern or risks associated with self-administered scales.
Two risks in particular potentially relate to this research. The first risk is that a
participant arbitrarily checked a series of responses that did not represent true
answers. This contributes to error of measurement. A second risk is that "social
desirability" influenced self-reported responses. This occurs when a respondent
answers questions based on a preconceived image that he or she wants to convey to
the researcher. For example, a respondent may indicate a team is a cohesive, close-

knit group when actually the team interacts poorly together. Reporting that the team

290

does not work well together risks placing the team in a negative light. This is
something a "socially desirable" individual tries to avoid.

While each of these issues can affect research results, the use of hierarchical
regression analysis minimized the risk associated with self-reported measures.
Variables subject to self-report bias, such as team cohesiveness and open
communication, were entered later within the analysis. This helps avoid allocating
too large a proportion of variance to uncorroborated measures and nrinimized the
impact of self-reported measures on research results.

External Generalizability Selection of the firms included in this study
occurred from a nonrandom listing. Because of this, obvious concern exists about
the external generalizability of research results. While demographic characteristics
are unique to this research sample, the major findings of this study should be
applicable to firms not included as part of the study. There is no reason to conclude
the major study findings are not externally applicable to other cross-functional

sourcing teams.

Bibliography

fimis, DJ. 1990. "Semiconductor Manufacturing Team." In Groups That Work, ed.
1. Richard Hackman. San Francisco: Jossey-Bass Publishers.

Alavi, M., and P.G.W. Keen. "Business Teams in an Information Age." The
Information Society 6:179-195.

Allen, T., R. Katz, 1.]. Grady, and N. Slavin. 1988. "Project Team Aging and
Performance: The Roles of Project and Functional Managers." R&D Management
(UK) 18(4):295-308.

Altier, W.J. "Task Forces: An Effective Management Tool." Management Review
76(2):52-57.

Ancona, D.G., and DE. Caldwell. "Demography and Design: Predictors of New
Product Team Performance." Interim Report, Sponsored by the Center for Innovation
Management Studies, Lehigh University, 1989.

Argyris, C. "The Incompleteness of Social-Psychological Theory: Examples From
Small Group, Cognitive Consistency, and Attribution Research." The American
Psychologist 15:893-908.

Austin, 1.1., and P. Bobko. 1985. "Goal-Setting Theory: Unexplored Areas and
Future Research Needs," Journal of Applied Psychology 58:289-308.

Baker, RM. 1988. "Participation in Small Groups: Social, Physical, and Situational
Predictors." Small Group Behavior 19:3-18.

Barczak, G. and D. Wilemon. 1989. "Leadership Differences in New Product
Development Teams." Journal of Product Innovation Management 6(4):259-267.

Bamer, R.W. 1989. "The Right Tool for the Job." Training and Development Journal
43(7):46-51. '

Barr, Vilma. 1990. "Six Steps to Smoother Product Design." Mechanical Engineering
112(1):48-51.

291

292

Bart, CK. 1988. "Organizing for New Product Development." The Journal of Business
Strategy 9(4):34-38.

Beer, M. 1976. "The Technology of Organization Development." In Handbook of
Industrial and Organizational Psychology, ed. M.D. Dunnette. Chicago: Rand McNally.

Belbin, RM. 1981. Management Teams: Why They Succeed or Fail. New York: John
Wiley & Sons.

Bertcher, H.J., and FF. Maple. 1977. Creating Groups. Beverly Hills, California:
Sage.

Blake, R. R., and J. S. Mouten. 1969. Building a Dynamic Corporation Through Grid
Organization Development. Reading, Mass; Addison-Wesley.

Blake, RR, and J .S. Mouten. 1975. "Group and Organizational Team Building: A
Theoretical Model for Intervening." In Theories of Group Processes, ed. Cary L.
Cooper. New York: Wiley.

Bradford, J .P. "Getting Together: A Composite Case Study in Team Building."
Optimum 20(2):38-51.

Campbell, D.T. 1956. Leadership And Its Effects Upon The Group. Columbus, Ohio:
Bureau of Business Research: College of Commerce and Administration.

Cheney, AB. 1991. "Self-Managed Work Teams." Executive Excellance 8(2):]1-12.

Chun, K.T., S. Cobb, and RP French, Jr. 1978. Measures for Psychological
- Assessment. Ann Arbor, Michigan: Institute for Social Research-The University of'
Michigan.

Clark, MS, and H.T. Reiss. 1988. "Interpersonal Processes in Close Relationships."
Annual Review of Psychology 39:609-672.

Cohen, J. and P. Cohen. 1983. Applied Multiple Regression/ Correlation Analysis for

the Behavorial Sciences Hillsdale, New Jersey: Lawrence Erlbaum Associates,
Publishers.

Cook, T.D., and D.T. Campbell. 1979. Quasi-Experimentation; Design and Analysis
Issues for Field Settings. Boston: Houghton Mifflin Company.

Crittenden, V.L. 1991. "Crossing Functional Lines: Achieving Successful Strategy."
Industrial Management 33(1):26-28.

293

Cummings, T.G. 1978. "Self-Regulating Work Groups: A Socio-Technical Synthesis."
Academy of Management Review 3:625-634.

Cummings, T.G. 1981. "Designing Effective Work Groups." In Handbook of
Organizational Design, ed. P.C. Nystrom and W.H. Starbuck. New York: Oxford
University Press.

Davis, J. H. 1973. "Group Decision and Social Interaction: A Theory of Social
Decision Schemes. " Psychological Review 80: 97-125.

Dean, J .W. and GI. Susman. 1989. "Organizing for Manufacturable Design." Harvard
Business Review J anuary-Februaryz28-36.

Delbecq, A.L., A.H. Van de Ven, and DH. Gustafson. 1975. Group Techniques for
Program Planning. Glenview, 111.: Scott, Foresman.

Dowst, Somerby and Ernest Raia. 1990. "Design 1990: Teaming Up for the '905."
Purchasing 108(2):54-59.

Dyer, W.G. 1977. Team Building: Issues and Altematives. Reading, Mass: Addison-
Wesley.

Elmes, D. and Wilemon, D. 1988. "Organizational Culture and Project Leader
Effectiveness." Project Management Journal 19(4):54-63.

Fiorelli, J .S. 1988. "Power in Work Groups: Team Member's Perspectives." Human
Relations 41:1-12.

Foushee, HQ 1984. "Dyads and Triads at 35,000 F eet."American Psychologist 39:885-
893.

Friedlander, F., and L.D. Brown. 1974. "Organization Development." In Annual
Review of Psychology, ed. M.R. Rosenweig and L.W. Porter, Vol. 25. Palo Alto,
Calif: Annual Reviews.

Geber, B. 1989. "Speed: Where the People Fit In." Training 26(8):23-30.

George, W.W. 1977. "Task Teams for Rapid Growth." Harvard Business Review
55(2):71-80.

Gladstein, D. 1984. "Groups in Context: A Model of Task Group Effectiveness."
Administrative Science Quarterly 29(4):499-517.

294

Golembiewski, R.T. 1965. "Small Groups and Large Organizations." In Handbook of
Organizations, ed. J .G. March. Chicago:
Rand McNally.

Goodman, P.S. ed. 1986. Designing Eflective Work Groups. San Francisco: J ossey-Bass
Publishers.

Goodman, PS. 1986. "Impact of Task and Technology on Group Performance." In
Designing Efi‘ective Work Groups, ed. P.S. Goodman. San Francisco: J ossey-Bass
Publishers.

Goodman, P.S., E.C. Ravlin, and L. Argote. 1986B. "Current Thinking About
Groups: Setting the Stage for New Ideas." in Designing Effective Work Groups, ed.
P.S. Goodman. San Francisco: Jossey-Bass Publishers.

Guzzo, RA. 1986. "Group Decision Making and Group Effectiveness." In Designing
Effective Work Groups, ed. P.S. Goodman.
San Francisco: Jossey-Bass Publishers.

Grayson, D. 1991. "Self-Regulating Work Groups--An Aspect of Organizational
Change." Intemational Journal of Manpower 12:22-29.

Green, TE. 1975. "An Empirical Analysis of Nominal and Interacting Groups."
Academy of Management Journal 18:63-73.

Hackman, J .R. 1968. "Effects of Task Characteristics on Group Products." Journal
of Experimental Social Psychology 4:162-187.

Hackman, J .R. 1969. "Toward Understanding the Role of Tasks in Behavioral
Research." Acta Psychologica 31:97-128.

Hackman, J .R. 1976. "Group Influences on Individuals." In Handbook of Industrial
and Organizational Psychology, ed. M.D. Dunnette. Chicago: Rand McNally.

Hackman, J .R. 1983. A Normative Model of Work Team Effectiveness. Technical
Report No. 2, Research Program on Group Effectiveness, Yale School of
Organization and Management.

Hackman, J .R. 1984. "Psychological Contributions to Organizational Productivity: A
Commentary." In Productivity Research in the Behavioral Sciences, ed. A.P. Brief. New
York: Praeger.

295

Hackman, J .R. 1985. "Doing Research That Makes a Difference." In Doing Research
that is Useful for Theory and Practice, ed. E.E. Lawler. San Francisco: Jossey-Bass.

Hackman, J .R. 1987. "The Design of Work Teams." In Handbook of Organizational
Behavior, ed. J. Lorsch. Englewood Cliffs, New Jersey: Prentice Hall.

Hackman, J .R. 1990. Groups That Work--Creating Conditions for Effective Teamwork.
San Francisco: Jossey-Bass Publishers.

Hackman, J.R., K.R. Brousseau, and J.A. Weiss. .1976. "The Interaction of Task
Design and Group Performance Strategies in Determining Group Effectiveness."
Organizational Behavior and Human Performance 16:350-365.

Hackman, J .R., and GR. Oldham. 1980. Work Redesign. Reading, Mass.: Addison-
Wesley.

Hackman, J .R., and RE. Walton. 1986. "Leading Group's in Organizations." In
Designing Effective Work Groups, ed. P.S. Goodman. San Francisco: Jossey-Bass
Publishers.

Halpin, A.W. and B.J. Woner. 1957. "A Factorial Study of the Leader Behavior
Descriptions," in Leader Behavior: Its Description and Measurement. Columbus, Ohio:
The Bureau of Business Research: The Ohio State University.

Hare, AP. 1976. Handbook of Small Group Research. New York: The Free Press.

Hare, AP. 1982. Creativity in Small Groups. Beverly Hills, California: Sage
Publications.

Heise, DR. 1975. Causal Analysis. New York: John Wiley and Sons.

Hemphill, J .K. 1956. Group Dimensions: A Manual for Their Measurement.
Columbus, Ohio: Bureau of Business Research-College of Commerce and
Administration.

Herman, SM. and MD. Herman. 1989. "Camels and Orbiters: Special Teams,
Properly Used, Can Create Effective Solutions." PersonnelAdministrator 34(10):90-92.

Herold, D.M. 1978. "Improving the Performance Effectiveness of Groups Through
a Task-Contingent Selection of Intervention Strategies." Academy of Management
Review 3:315-25.

296

Hewett, T.T., G.E. O'Brien, and J. Homik. 1974. "The Effects of Work Organization,
Leadership Style, and Member Compatibility upon the Productivity of Small Groups
Working on a Manipulative Task." Organizational Behavior and Human Performance

11:283-301.

Hildebrand, D.K. and L. Ott. 1991. Statistical Thinking for Managers. Boston: PWS
Kent Publishing Company. .

Hirokawa, R.Y., and Johnston, D.D. "Toward a General Theory of Group Decision
Making: Development of an Integrated Model." Small Group Behavior 20:500-523.

Hoffman, LR. 1979. "Applying Experimental Research on Group problem Solving
to Organizations." Journal of Applied Behavioral Science 15:375-391.

Hollenbeck, J .R., H.K. Klein, A.M. O'Leary, and PW. Wright. 1989. "Investigation
of the Construct Validity of a Self-Report Measure of Goa] Commitment," Journal
of Applied Psychology 74:951-956.

Hunter, J .E. 1986. "Multiple Dependent Variables in Experimental Design,"
Presented at a workshop at the University of Iowa, October 12, 1986.

Hunter, J .E. 1991. Advanced Psychometric Seminar, Department of Psychology,
Michigan State University.

Hunter, J .E. and D.W. Gerbing. 1982. "Unidimensional Measurement, SecondOrder
Factor Analysis, and Causal Models." Research in Organizational Behavior 4:267-320.

Ilgen, DR, and Hollenbeck, J .R. 1989. "A Proposal for Team Research on Effective
Team Performance Under Stress and Normal Conditions." A research grant proposal
submitted to the Office of Naval Research.

James, LR. 1980. "The Unmeasured Variables Problem in Path Analysis." Journal
of Applied Psychology 65:415-421.

James, L.R., S.A. Mulaik, and J .M. Brett. 1982. CausalAnalysis: Assumptions, Models,
and Data. Beverly Hills, California: Sage Publications.

Janis, LL. 1982. Groupthink: Psychological Studies of Policy Decisions and F iascoes.
Boston: Houghton Mifflin.

Kaplan, RE. 1979. "The Conspicuous Absence of Evidence That Process

Consultation Enhances Task Performance." Journal of Applied Behavioral Science
15:346-360. >

297

Katz, R. 1982. "The Effects of Group Longevity on Project Communication and
Performance." Administrative Science Quarterly 27:81-104.

Kerlinger, EN. 1986. Foundations of Behavorial Research Fort Worth, Texas: Holt,
Rinehart, and Winston, Inc. ‘

Kolodny,H., and M. Kiggundu. 1980. "Towards the Development of a Sociotechnical
Systems Model in Woodlands Mechanical Harvesting." Human Relations 33:623-645.

Kumar, S. and Y.P. Gupta. 1991. "Cross-Functional Teams Improve Manufacturing
at Motorola's Austin Plant." Industrial Engineering 23(5):32-36.

Larson, C. 1988. "Team Tactics Can Cut Product Development Costs." The Journal -
of Business Strategy Sep/Octz22-25.

Latane, B. 1986. "Responsibility and Effort in Organizations." In Designing Effective
Work Groups, ed. P.S. Goodman. San Francisco: Jossey-Bass Publishers.

Latane, B., K. Williams, and S. Harkin. 1979. "Many Hands Make Light the Work."
Journal of Personality and Social Psychology 37: 822-832.

Lawler, EB. 1981. Pay and Organization Development. Reading, Mass: Addison-
Wesley.

Lawley, D.N. and A.E. Maxwell. 1971. Factor Analysis as a Statistical Method,
London: Butterworth.

Leavitt, H.J. 1975. "Suppose We Took Groups Seriously." In Man and Work in
Society, ed. E.L. Cass and PG. Zimmer. New York: Van Nostrand Reinhold.

Lefton, RE, and V.R. Buzzotta. 1987. "Teams and Teamwork: A Study .of
Executive-Level Teams." National Productivity Review 7:7-19.

Leiva, WA. and J .W. Obermayer. 1989. "Commonsense Product Development."
Business Marketing 74(8):44-48.

Levine, J .L., and R.L. Moreland. 1990. "Progress in Small Group Research."Annual
Review of Psychology 41:585-634.

Likert, R. 1961. New Patterns of Management. New York: McGraw-Hill Book Co.

Lippitt, G.L., and E. Seashore. 1980. Group Effectiveness: A Looking-Into-Leadership
Monograph, Fairfax, VA: Leadership Resources, Inc.

298

Litwin, G.H. and RA. Stringer, Jr. 1968. Motivation and Organizational Climate.
Boston: Division of Research-Harvard University.

Loehr, L. 1991. "Between Silence and Voice: Communicating in Cross-Functional
Project Teams." IEEE Transactions on Professional Communications 34(1):51-56.

Maier, N.R.F. 1963. Problem Solving Discussions and Conferences: Leadership
Methods and Skills. New York: McGraw-Hill Book Co.

Maier, N.R.F. 1967. "Assets and Liabilities in Group Problem Solving." Psychological
Review 74:239-249.

March, J .G., and HA. Simon. 1958. Organizations. New York: Wiley.

Margerison, OJ. and D. McCann. 1989. "How to Improve Team Management."
Leadership and Organizational Development Journal (UK) 10(5):4-42.

Massinrilian, RD. and L. Pedro. 1990. "Back From the Future: Gearing Up for the
Productivity Challenge." Management Review 79(2):41-43.

McGrath, J .E. 1986. "Studying Groups at Work: Ten Critical Needs for Theory and
Practice." In Designing Effective Work Groups, ed. P.S. Goodman. San Francisco:
Jossey-Bass Publishers.

McGrath, J .E. 1984. Groups: Interaction and Performance. Englewood Cliffs, New
Jersey: Prentice-Hall. .

McGrath, J .E. 1964. Social Psychology--A Brief Introduction. New York: Holt,
Rinehart, and Winston.

McGrath, J .E. and I. Altman. 1966. Small Group Research: A Synthesis and Critique
of the Field. New York: Holt.

McGrath, J .E. and DA. Kravitz, 1982. "Group Research." Annual Review of
Psychology 33:195-230.

McGregor, Douglas. 1960. The Human Side of the Enterprise. New York: McGraw
Hill Book Company.

McKeown, J .J . 1990. "New Products From New Technologies." Journal of Business
and Industrial Marketing 5:67-72.

Messick, D.M., and D.M. Mackie. 1989. "Intergroup Relations." Annual Review of
Psychology 40:45-81.

299

Meyers, D.G., and H. Lamm. 1976. "The Group Polarization Phenomenon."
Psychological Bulletin 83:602-617.

Meyers, P.W. and D. Wilemon. 1989. "Learning in New Technology Development
Teams." Journal of Product Innovation Management 6(2):79-88.

McPhee, RD. and A. Barrow. 1987. "Causal Modeling in Communication Research:
Use Disuse, and Misuse." Communication Monographs 54:345-366.

Miner, CF. 1984. "Group Versus Individual Decision Making: An Investigation of
Performance Measures, Decision Strategies, and Process Losses/Gains."
Organizational Behavior and Human Performance 33:112-124.

Mink, O., and B. Mink. 1975. Developing Effective Work Groups. Austin, Texas:
OHRD Associates.

Mink, O.G., B.P. Mink, and KO. Owen. 1987. Groups at Work. Englewood Cliffs,
New Jersey: Educational Technology Publications.

Mitsch, RA. 1990. "Three Roads to Innovation." Journal of Business Strategy
11(5):18-21.

Monczka, RM. 1989. "Integrated Procurement Strategy" study (unpublished), The
Eli Broad Graduate School of Management, Michigan State University.

Monczka, R.M., R.J. Trent, and GR. Gay. "Purchasing and Sourcing Strategy:
Trends and Implications," Purchasing White Paper, August 1992.

Morgan, J .P. 1991. "Purchasing 2000: Spell the Function with an "S" for Supply."
Purchasing 1 10(2):48-51.

Morris, RM. 1991. "Motives, Selection, and Support: What Management Can Do to
Make Sure a Team Fails." Industrial Management 33(2):2-3.

Mountain Bell Team Marketing DevelOpment Project. 1982. Graduate School of
Business at Columbia University, New York, New York.

Murray, A. I. 1989. "Top Management Group Heterogeneity and Firm Performance."
Strategic Management Journal (UK) 10(Summer):125-141.

Musselwhite, WC. 1990. "Time-Based Innovation: The New Competitive Advantage."
Training and Development Journal 44(1):53-56

300

Mower, J ., and D. Wilemon. 1989. "'Team Building in a Technical Environment." In
Handbook of Technology Management, ed. D. Kocaoglu. New York: John Wiley.

Nagao, D.H., D.A. Vollrath, and J .H. Davis. 1978. "Origins and Current Status of
Group Decision Making." In Dynamics of Group Decisions, ed. H. Brandstatter, J .H.
Davis, and HG Schuler. Beverly Hills, Calif: Sage.

Nieva, V.F., E.A. Fleishman, and A. Rieck. 1978. "Team Dimensions: Their Identity,
Their Measurement, and Their Relationships." Final Technical Report Washington
DC: Advanced Research Resources Organizations.

Nunnally, J .C. 1978. Psychometric Theory New York: McGraw-Hill Book Company

O'Dell, J .W. 1968. "Group Size and Emotional Interaction." Journal of Personality and
Social Psychology 8:75-78.

O'Leary, A.M. 1990. Motivation in Groups: A Control Theory Model. Unpublished
Ph.D. Dissertation. Michigan State University.

Osborn, A. F. 1957. Applied Imagination. New York: Scribner's.

Parker, G. M. 1990. Team Players and Teamwork The New Competitive Business
Strategy. San Francisco. J ossey-Bass Publishers.

Pearce, J. A., and E. C. Ravlin. 1987. "The Design and Activation of Self-Regulating
Work Groups." Human Relations. 40: 751- 782.

Pearson A.W., and HP. Gunz. 1981. "Projects Groups." In Groups At Work, ed. R.
Payne and CL. Cooper. New York: Wiley.

Peters, L.H. and O'Connor, E.J. 1980. "Situational Constraints and Work Outcomes:

The Influences of a Frequently Overlooked Construct." Academy of Management
Review 5:391-397.

Peters, Tom. 1988. "Restoring American Competitiveness: Looking for New Models
of Organizations." Academy of Management Excellance 2(2):103-109.

Pinto, MB. and J.K. Pinto. 1991. "Determinants of Cross-Functional Cooperation in
the Project Implementation Process." Project Management Journal 22(2):13-20.

Pinto, MB. and J.K. Pinto. 1990. "Project Team Communication and Cross-
Functional Cooperation in New Program Development." Journal of Product
Innovation and Management 7(3):200-212.

301

Pitts, CE. 1990. "For Project Managers: An Inquiry into the Delicate Art and
Science of Influencing Others." Project Management Journal 21(1):21-23,42.

Porter, A.M. 1991. "Quality 1991-Buying Quality: Learning to "Walk the Talk"."
Purchasing 110(1):62-71.

Poza, E.J., and ML. Markus. 1980. "Success Story--The Team Approach to Work
Restructuring." Organizational Dynamics (Winter):3-25.

Price, J .L., and CW. Mueller. 1986. Handbook of Organizational Measurement.
Marshfield, Massachusetts: Pitman Publishing, Inc.

Raia, Ernest. 1991. "Taking Time Out of Product Design." Purchasing 110(6):36-39.

Rao, Ashok. 1989. "Manufacturing Professionals of the 1990s: How Should They Be
Prepared?" Production and Inventory Management 30(4):64-67.

Reich, RB. 1987. "Entrepreneurship Reconsidered: The Team as Hero." Harvard
Business Review 3:77-85. ‘ ,

Robbins, HA. 1988. "An Alternative to Turf Wars." Executive Excellence 5(7):9-10.

Rosen, N. 1989. Teamwork and the Bottom Line: Groups Make a Difference. Hillsdale,
New Jersey: Lawrence Erlbaum Associates.

Rubin, I., M. Plovnick, and R. Fry. 1977. Task-Oriented Team Development. New
York: McGraw-Hill Book Co.

Rubin, J.Z. 1984. "Introduction to Group Decision Making." In Group Decision
Making, ed. C.F. Swap. Beverly Hills, Calif: Sage Publications.

Russell, RD. 1990. "Innovation in Organizations: Toward an Integrated Model,"
Review of Business 12(2):19-26.

Sample, J .A., and Wilson, TR. 1965. "Leader Behavior, Group Productivity, and
Rating of Least Preferred Co-Worker." Journal of Personality and Social Psychology
3:266-269.

Seers, Anson. 1989. "Team-Member Exchange Quality: A New Construct for Role-
Making Research." Organizational Behavior and Human Decision Processes 43:118-
135.

Schein, EH. 1988. Process Consultation. Reading, Mass.: Addison-Wesley.

302

Schwab, D.P. "Construct Validity in Organizational Behavior." Research in
Organizational Behavior 2:3-43.

Shaw, ME. 1981. Group Dynamics. New York: McGraw-Hill Book Co.

Shiflett, SC. 1972. "Group Performance as a Function of Task Difficulty and

Organizational Interdependence." Organizational Behavior and Human Performance
7:442-456.

Shiflett, SC. 1979. "Toward a General Model of Small Group Productivity."
Psychological Bulletin 86:67-79.

Simmons, John. 1989. "Partnering Pulls Everything Together." Journal for Quality and
Participation J une:12-16.

Smock, D. 1982. "How to Stem the Tide of Shoddy Materials," Purchasing, 92(9):51-
57.

Stalk, G. 1988. "Time-The Next Source of Competitive Advantage," Harvard Business
Review 66(4):41-51.

Stasser, G., and J .H. Davis. 1981. "Group Decision Making and Social Influence: A
Social Interaction Sequence Model." Psychological Review 88:523-551.

Stein, M.I. 1975. Stimulating Creativity. New York: Academic Press.
Steiner, ID. 1972. Group Process and Productivity. New York: Academic Press.

Stogdill, RM. and A.E. Coons, eds. 1957. Leader Behavior: Its Description and
Measurement. Columbus, Ohio: The Bureau of Business Research: The Ohio State
University.

Stogdill, RM. 1981. "Leaders and Their Immediate Groups." Handbook of
Leadership. Chapter 24. New York: Free Press.

Stumpf, S.A., D.E. Zand, and RD. Freedman. 1979. "Designing Groups for
J udgmental Decisions." Academy of Management Review 4:589-600.

Survey Research Center. 1975. The Michigan Organizational Assessment Package-
Progress Report [1. Ann Arbor, MI: Institute for Social Research.

Swap, C.F. ed. 1984. Group Decision Making. Beverly Hills, Calif: Sage Publications.

303

Tindale, SR. 1989. "Group versus Individual Information Processing: The Effects of
Outcome Feedback on Decision Making." Organizational Behavior and Human
Decision Processes 44:454-473.

Thamhain, H.J. and D.L. Wilemon. 1987. "Building High Performing Engineering
Project Teams." IEEE Transactions on Engineering Management 34(3):130-137.

Thomas, E.J. and Fink, GR 1961. "Models of Group Problem Solving." Journal of
Abnormal and Social Psychology 63:53-63.

Thomas, E.J. and Fink, GR 1963. "Effects of Group Size." Psychological Bulletin
60:371-384.

Tjosvold, D. 1991. Team Organization: An Enduring Competitive Advantage. New
York: John Wiley & Sons.

Tolle, ER 1988. "Management Team Building: Yes But!" Engineering Management
International 4(4): 277-285.

Tuckman, B.W. "Developmental Sequence in Small Groups." Psychological Bulletin
63:384-399. - ‘

Tziner, Ahanor and Eden, Dov. 1985. "Effects of Crew Composition on Crew
Performance--Does the Whole Equal the Sum of Its Parts?" Journal of Applied
Psychology 70:85-93.

Van de Ven, AH. 1974. Group Decision Making and Effectiveness: An Experimental
Study. Kent, Ohio: Kent State University Press.

Van De Ven, A.H., and D.L. Ferry. 1980. Measuring and Assessing Organizations.
New York: John Wiley & Sons.

Van de Ven, AH. and G. Walker. 1984. "The Dynamics of Interorganizational
Coordination." Administrative Science Quarterly 29:598-621.

Versteeg, A. 1990. "Self-Directed Work Teams Yield Long-Term Benefits." The
Journal of Business Strategy Nov/Decz9-12.

Welter, TR. 1990. "How to Build and Operate a Product-Design Team." Industry
Week April 16:35-38.

Whyte, W.F. 1955. Money and Motivation: An Analysis of Incentives in Industry. New
York: Harper.

304

Wicker, A.W., S.L. Kirrneyer, L. Hanson, and D. Alexander. 1976. "Effects of
Manning Levels on Subjective Experiences, Performance, and Verbal Interaction in
Groups." Organizational Behavior and Human Performance 17:251-274.

Woodman, R.W., and JJ. Sherwood. "The Role of Team Development in
Organizational Effectiveness: A Critical Review." Psychological Bulletin 88:166-186.

Zander, AF. 1971. Motives and Goals in Groups. New York: Academic Press.

Zander, AF. 1977. Groups at Work: Unresolved Issues in the Study of Organizations.
San Francisco: Jossey-Bass.

Zander, AF. 1980. "The Origins and Consequences of Group Goals." In
Retrospections on Social Psychology, ed. Leon Festinger. New York: Oxford University
Press.

Zenger, TR. and BS. Lawrence. 1989. "Organizational Demography: The
Differential Effects of Age and Tenure Distributions on Technical Communication."
Academy of Management Journal 32: 353-376.

Ziller, RC. 1957. "Group Size: A Determinant of the Quality and Stability of Group
Decisions." Sociometry 20:165-173.

APPENDIX A

Information Packet Forwarded to Potentially Participating Companies

BROAD

THE ELI BROAD
GRADUATE SCHOOL
OF MANAGEMENT
Robert M. Monczka. 'th

Protector of Strategic Sourcing
management and The National
Association ot Purchasing
Ianaoelnont Protessor

Department 0! Management
Mrdllgan State UnivetSlty
206 anley Center

£351 1.303109 Mlcmw
48824-1121
517/336-3503

m. 5177'339-0778

KSU 7: an strum-1mm
flatworm; mam

305

MICHIGAN STATE

UNIVERSITY

 

April 27, 1992

1"
2-
3-
4-
5-

Dear 6':

We are undertaking a research project in The Eli Broad Graduate School of
Management at Michigan State University that focuses on cross-functional
sourcing (purchasing) team effectiveness. Although firms and executive ’4";
management are increasingly relying on cross-functional sourcing teams to
support quality, cost, technology, delivery, and concept-to-customer timing
improvements. little is actually known about what determines team success or
failure. The Cross-Functional Sourcing Team Effectiveness Research Project
will provide research-based answers for applied use in improving cross-
functional sourcing-team effectiveness and competitive performance. This
research will evaluate in depth the effects of certain variables on sourcing team
effectiveness. These variables include organizational resources. team
leadership, team effort, the organizational reward system, communication. team
experience, and team cohesiveness.

We are requesting that 3', because of your use of cross-functional sourcing
teams, participate in the study. Company participation will require identifying
cross-functional sourcing teams, team leaders, and an external rater capable of
assessing team effectiveness who will complete the sourcing team effectiveness
survey. We also ask that you identify a liaison or coordinator at your
organization who will serve as a primary contact between our research staff
and your firm. Details are discussed more fully later in this letter.

We define a cross-functional sourcing team as consisting of personnel from at
least three different functions brought together to achieve a purchasing or
material related task(s) where purchasing/sourcing goals or decisions involving
supply base management must be considered by the team. The task can be a
specified project or consist of an ongoing assignment. Examples of different
tasks or assignments of cross-functional sourcing teams can include purchased

306

1'
April 27, 1992
Page 2

product cost reduction, commodity management, Special purchasing and materials
projects, value analysis or new product design and development sourcing support.

A questionnaire will be used to gather data to assess cross-functional sourcing team
effectiveness. Each cross-functional sourcing team member, leader, and external rater
will respond to items individually about the possible determinants of sourcing team
effectiveness. The total time required for survey completion should not be more than 1-2
total hours per team member and leader and can be completed anytime before the
scheduled return date. The survey does not require team leaders or members to gather
data or respond to open-ended questions. In addition, a shorter questionnaire requiring
20-30 minutes must be completed by a manager or executive who, as an external team
rater, can reasonably evaluate team performance and respond to questions across a
number of team performance categories.

This research includes three versions of a cross-functional sourcing team effectiveness
questionnaire:

(1) Each team member who is not a team leader will complete the
questionnaire.

(2) The team leader will complete a slightly different version of the
questionnaire. The team leader's survey includes most of the same
questions as the team member's survey along with questions the leader can
answer for the entire team.

(3) An external rater evaluates team performance from a number of critical
perspectives. .

At least one of the team members must be from a purchasing or materials related
function. The team must have worked together over a period of time long enough to
assess the team's capability against your performance expectations. If the sourcing team
is project oriented, the team should have a significant portion (i.e. 50 to 75 percent or
more) of the project or work effort complete. We can include teams in this study that
are no longer active if they have worked together recently and an external rater can
objectively evaluate the team's performance.

Team members do not have to be located in the same general physical proximity. For
example. a cost reduction team may include functional members from separate business
units located in different geographic regions. Cross-functional sourcing teams which are
also cross-locational teams are candidates for study because they allow us to assess the
effect of geographic distance on team performance.

307

1'
April 27, 1992
Page 3

All survey information will be held in the strictest confidence. Team members are not
identified during data analysis. We will be able to identify results, however, across
different industries and companies. While industries will be identified during data
analysis, participating companies will not.

In addition to gaining insight into cross-functional sourcing team effectiveness, we will be
able to benchmark responses at an industry and organizational level. Participating
companies will be large, multi-national firms in the aerospace, appliance, automotive,
computer, electronic, pharmaceutical, pulp and paper, telecommunication, and tractor
and farm implement industries plus selected other firms from both a process and
assembly environment. Fifteen to twenty companies are being asked to provide 10-15
cross-functional sourcing teams for survey completion. Of course, if 10-15 teams are not
available, a lesser amount is also appreciated.

It is critical that we study teams which have had varying degrees of success. A study that
evaluates only the most successful cross-functional sourcing teams within an organization
will provide minimal insight about team effectiveness. If possible, we are asking firms to
identify cross-functional sourcing teams perceived to have different degrees of success.

We believe the time to complete the questionnaire will be worth your firm's effort. In
return for your participation, we will provide your organization with an in-depth
evaluation of the research results above and beyond those to be reported in business
journals. This includes a detailed white paper of our research findings and conclusions.
Comparative data will also be provided. We are also willing to present our research
findings directly to key managers within your organization. Our goal is to gain a better
understanding of cross-functional sourcing team effectiveness so your firm can effectively
and efficiently manage these critical teams.

Executive management support and commitment is critical for this project. Once
organizations indicate their willingness to participate in this research and have identified
teams and a coordinator, we will move quickly to forward questionnaires to team
members, leaders, and external raters. Questionnaires will be mailed directly from
Michigan State University to team members, leaders and raters during early May. We
will request that each research participant complete, seal and return the questionnaire
directly to Michigan State University in a postmarked enveIOpe we provide within ten to
fourteen days after questionnaire receipt.

Enclosed information outlines the planned research process, an executive
acknowledgement indicting your support, control sheets to provide team data, a brief
description of the external rater's position, and a sample letter that will accompany
questionnaires sent directly to team personnel within your film. Also enclosed is a recent
writing by Peter Drucker in the Wall Street Journal for your review. Enclosure is not
meant to suggest agreement or disagreement but merely to further support the

308

1-
April 27, 1992
Page 4

importance of this research topic.

We would greatly appreciate your participation in a project that will strengthen our
knowledge in this important area. Please complete the enclosed Executive
Acknowledgement which indicates your willingness to support this project. If you agree
to participate, please have your film complete the team record sheets and forward the
sheets directly to me. We also hope you will be able to gain the support of other
nonpurchasing members of the cross-functional sourcing teams. If you believe it is
appropriate for another business unit within your firm to also participate, please advise us
by fax or letter and we will forward the information to that business unit.

My project coordinator for this research is Robert J. Trent. Mr. Trent is a Ph.D.
candidate at Michigan State University. He is combining this project with his Ph.D.
dissertation research. Again, thank you for your time and consideration. Please call me
if you have any questions (517/353-5415 or 517339-4456).

Sincerely,

Robert M. Monczka, Ph.D.

Professor of Strategic Sourcing Management
and The National Association of Purchasing
Management Professor

jkt
chS

309

EXTERNAL TEAM RATER
DESCRIPTION SHEET

I Theselection ofthe external sourcingtamrateriscriticaltothis research project.
This individual must be able to evaluate objectively the cross-functional sourcing m
from a number of critical perspectives

I Theexternalratermustassess:

Sourcing team level of effort
Availability of organizational resources
Sourcing team authority and autonomy

Sourcing team performance

I Sourcing team performance areas requiring objective evaluation include:

Quantity and quality of team output

Ability of the team to generate innovate ideas

Team reputation for work excellence

Team efficiency

Team morale

Timing and task schedule targets

Communication

Team ability to meet organizational expectations or objectives in a number of
purchasing and sourcing performance areas (i.e. cost reduction targets,

delivery improvement, etc.)

Overall sourcing team performance

I The external rater can consult with others outside the team if required to provide an
objective assessment for a particular item

The at M Gender-e School DIM-Iguassu The Cruel-Fused“ Sauna; Tear-l ”New Research Played

310

0-QMMuEZG§¢2>~t MQQ-6~ZQ HRS: mwwmnfi~<m2hmm
gwgafiz waQfima >2? >Q~<~§Mm

2. nuance.) BIB-III: E8..- i W. . 9.5% 885...- 85‘. 6..- G. Del-SEER- neneaaoa .585 .e .33

4 Quinn—.38. . . §.§E§§I §.§B&3§L3§
0, £353.32...- ,. . o EEBEEEI
guns. Eu 0 Oil-FER. 8812! Ex. .eteaeopeq .28..
seen-331.82g f «Sea-8.3.3.588 sienna-rage?
, IE. 3.. SEN . HIIIHV 885% can-13:93. PHHV ‘ 325. 829—8
43.1 E" an 2.1— 3.; also. .52 3 I: 73

4.80. a..." we. 25.. waéo

7. 4.33 5.309.... Es. 2.x— ofioap— G. ”883: an: 3.58 803.32: en 98 E. U..- .eLw-m. Bx— 5863.59.

gialSBuiEa-BE flags-Ee-
-9:8I. on} ea 38.! 42.8 a...“ .53;

4.81 .52 9.3:.-
ronea 98595.”... grungy 8

2:53 we: c2335 .HV t”

4.31 a..." Z. 388. 38.3.. 3
Foo _ ‘

 

BI? pea—pvt 8 130.75.. :38
were: I. leans—S. 3.0.
.782 .52 0988 13?! ea ecu—out «9:13
.3... m: ma!— 0% «98. ea 22.-n88.
7:255 «Ea c.5335
3. M35 088..
ml. Eu. 2. 33?:2

1.2.5.3 .35 392:

 

 

 

gnu-igfgillli urogfiigflq'gnagig

311

fithhtfiQZGNqQZg hQanHZQ HM): Nhfim2~<h2~wmm gmgafifi fibeMfiu.

a; god:

’4‘.
085523». game—en 288 8:653 53 «:8. new 80.. 8:3...» .83 wen—cane E :5 «83. 2:. 3233 5: mien—3:2. 3. 333:0 8

A33 30.3.3.

9:624"

 

235 2 083.38 on 48::

 

2.30

062532.». Ceca—mans:

 

d:&>a§.flh§5=1fi.8§=n

 

4.83 Eon

 

 

4.8:. 2232.2

3

 

 

3

 

 

CV

 

 

3

 

 

9

 

 

3

 

 

 

3

 

T83». H85 ~38:

 

 

H85.“ v1.25 H8.» 9. tapas—=2...
.230 v3.8 «5:9 .9: Home. :8 fen—8.. 4.3092. 53:58“ 8 Seizes—c“

Zefi =. we: 89:8 £332.»— 680. 28.6 £62. n 8338 «.52

iuigi‘kii

 

at. 908.!!! in Hall gall;- Dagr is:

APPENDIX B

Information Letter Forwarded to Individual Study Participants

BROAD

THE ELI BROAD
GRADUATE SCHOOL
OF MANAGEMENT
Robert M. Monczka. PhD

Protessor ot Strategic Sourcing
Management and The Itatlonal
Association at Purchasing
Management Protessor

Department at Management
Mrcnrgan State Unlvthlfy
206 Eaaley Center

East Lansmg Mncmgan
4882mm

517/335-3503

FAX Sui/3390778

MSU Is an JMWI'WfOfl
cow Wendy nsvrwan

312

MICHIGAN STATE

UNIVERSITY

 

September 16, 1992

1-
2-
3-
4-
5-

Dear 6':

The Eli Broad Graduate School of Management at Michigan State University is
undertaking a project titled The Cross-Funcrional Sourcing Team Effectiveness Research
Project. ‘lhis research focuses on cross-functional sourcing team effectiveness for teams
involved with purchasing and materials-related acrivities. The executive management of
your firm has agreed to support our research efl‘ort. As a result, we would greatly
appreciate your completing and returning the enclosed questionnaire. The data you
provide will help us analyze and identify the factors that contnbute to cross-funCtional
sourcing team effecriveness. I

Your management indicates you are 7' of a cross-funcrional team which appears on the
enclosed team listing identification sheet. All questions and answers refer to your
experiences with the team listed on the enclosed identification sheet only and no other
team you may have been involved or are currently involved with.

Please consider the following when answering the questionnaire:

0 Respond to each quesrion as openly and honestly as possible. There are no right
or wrong annvers per so. However. it is critical that your responses accurately
reflect your experiences with the team identified in this packet only.

0 If you are a member. leader, or external rater of more than one team included in
the Study. you will receive a questionnaire for each team you are associated. We
would greatly appreciate your completing each queSIionnaire. It is critical that we
receive questionnaires from all members of a team. Your experiences with each
individual team and its members are important to this study.

0 If your team has more than one designated leader. please respond to any
leadership questions based on your eitperience with the team leadership in general
or your experience with the leader with which you are most familiar.

313

1-
September 16, 1992
Page 2

0 Completion of the questionnaire is an individual effort Please do not discuss your
answers with other team members.

0 If the team you are providing answers about is no longer active. please respond to the
survey as if the questions are phrased in the past tense. Also, please provide answers
which best reflect when the team was actively pursuing its assigned task(s).

The information you provide remains confidential at all times. While we request your name on
the survey. it is not keypunched or recorded. We require your name simply as a means of
verifying the progress of our data colleCtion for each team and to match teams and members
correctly. When a team has completed and returned its questionnaires, we will keypunch the
data for analysis. Only the project directors will ever see the completed surveys.

We believe the time commitment to complete the quesrionnairc will be worth your firm's effort.
While the questionnaire contains numerous items covering a range of topics. you do not have to
collecr external data to complete the questionnaire. In return for your participation, we are
providing your organization with an in-depth evaluation of the research results above and beyond
those reported in business journals. This includes a detailed white paper of our research findings
and conclusions. We will also possrbly present our research findings directly to your company.
Our goal is to gain a better understanding of sourcing team effectiveness so firms such as yours
can develop and manage these critical teams for maximum performance.

We greatly appreciate your participation in a project that will strengthen our knowledge in this
important area. Please place the completed questionnaire in the provided envelope and forward
it direcrly to Michigan State University. We request that you complete and forward the
queStionnaire no later than fourteen days after receipt. Again, thank you for your support.

If you have any questions. please feel free to contact Robert Trent, the research project
coordinator, at (517) 485-7859.

Sincerely.

Robert M. Monczka, Ph.D.

Project Director

Professor of Strategic Sourcing Management
and The National Association of Purchasing
Management Professor

jltt
enc

APPENDIX C

Follow-Up Letter Sent to Participants With Missing Data

BROAD

THE ELI BROAD
GRADUATE SCHOOL
OF MANAGEMENT
Robert is. htonczta. Pno

Protessor ot Strategic Sourcing
flanagement and The National
Association ot Purchasing
Management Protessor

Department at Menagerient
Micmqan Slate Unive'ssly
206 Eooley Center

East Lansing. Micmgan
4882mm
517/336-3503

FAX 517/339-0778

A641 rs an min-1cm
ems/W mama

314

MICHIGAN STATE

UNIVERSITY

 

September 15. 1992

l'
2-
3-
4-
5-

Dear 6':

As you may know, your firm has agreed to participate in the Michigan State
University Cross-Functional Sourcing Team Effectiveness Research ProjeCt. Our
records show that as of September 14. 1992 we have not yet received your completed
survey. We are well along in the data collection effort and have just completed a
review of "missing" team members. leaders, and external raters. To successfully
complete this project, we musr have whole sets of team represented. We would
greatly appreciate your completing and returning the survey as soon as possible.

We believe the time commitment to complete the questionnaire will be worth your
firm's effort. In return for your participation, we are providing your organization
with an in-depth evaluation of the research results above and beyond those reported
in business journals or those available to nonparticipating companies. Our goal is to
gain a better understanding of sourcing team effectiveness so firms such as yours can
develop and manage these critical teams for maximum performance.

We apprecrate your participation in a project that will Strengthen our knowledge in
this important area. If you have any quesrions or require an additional survey. please
contact Robert Trent, the research project coordinator. at (517) 485-7859.

Sincerely.

Robert M. Monczka, Ph.D.

Project Director

Professor of Strategic Sourcing Management
and The National Association of Purchasing
Management Professor

jkt

APPENDIX D

Requests for Measuring Instrument

Please forward requests for a copy of the team member, leader, or external rater
survey to:

Robert J. Trent
Department of Management
Rauch Business Center

621 Taylor Street

Lehigh University
Bethlehem, PA 18015-3117

315

APPENDIX E

Frequency Reporting of Survey Questions

316

Gm :82 Z
.623— EEouxm

wvé

mm;

3."

on;

2.—

oo. ~

he.“

"NA

54
on.“

ow.n

SN

wwN

86
N56.

mod

Ema

wvfi

mod

Kw.

Ham.

0:”

brim

mod
wmrm

wwv

oov

owv

owe

owe
3v

ca...

03.

03.

owe
03.
09.

8v

av

03.
Re

No. H

on;

no."

we;
no.

on;

$4
NA
34
3;

v.4

me. fl
3. a

86
3.0

3d

mad
and
3d
00%.
8.0

mad

and
3N

oocw< bweohmnr 6298.5 hweohmflfi

Gm :32 Z

con—:22

Gm :32 Z

hO—ug

no“

mo“

we

mod

we“
we

wo—

mg

mg

we
won
me“

no”

we

we
we

E3. £5 .33 waste? 98.... 2
once E EoEEwimo Eco. .«o 25 En 30E? 388 2:03 _

E3. «2. mo EeroE .050 So: c2356.?
283an ©0332 _ cos? 88: me ..er=: a no :55 c3 _

Eco. £5 :0 .8380
c2385.: ego—2E3. no omeanoxo some 98 229—50

E3. £5
:0 eosaoieEEoo o. autumn “50.55% E8 03 22F

vo>anm on 2:8 955
2.: .«o EeroE wcoEa common co_.anE_ .0 58:8“ 2:.

Eco. m2. .«0 EonEoE .050 2 “=3 e. 0393.qu x can _

0:. Eat 9580.. .95 :oEEEfi:
05 Eggnog: .0: cc 9650 35 .3.“ 85368 _

EonEoE
Eco. cofio an ozoooc _ eozaEa85 on. go .6838
05 385 one zone on 9 E .8 E8802. coco a .—

some
bo> “a Eco. «.2. mo EonEoE 05 5223 eozaoieEEoU

BoEEooE
coco a §EoE Eco. .350 E9... ozoooc _ eozoEcoufi 2F

Eco. v.2. uo EonEoE 05 55, ano 58 9 Web as .—
Eoo. of. co boo...“ $5.08 :05 Booms—5E8 :3 EonEoE

350 coco
9 cosoanHE oEoE:EE8 9 :8 code EopEoE E89.

EonEoE Eco.
cofo 53> v.32 :05 Seems—EEO“. beoao EonEoE Ewoh

EopEoE Eco. .850 9 cozoEeefi
memos—E: 02.65 Eco. aw: co fleocmzoem oEom

EonEoE E3. .550 .82—8 9 2356 mm x 4805» 5

magic» 525,—.
:82on

317

00.. no.0. 3.

mm... afim 06.

N0. mw.m mo.

Gm 30.). Z
00.3. 3:08....

no. .

o...

o...
0N.

2..
8..
n...
8..
S...
8..
0.0..
8..
on.
NN.
N...
D...
Gm

m3. 8..
8.0 8..
2.... 8..
:0 a.
8... n...
8.0 8..
8.0 8..
.2 8..
0...... .2.
NS 8..
8.0 8..
.3 8..
00.0 8..
$4 8..
3.0 a...
9.... 0:.
2... 08
.302 Z

35:05.

no. .

w...

no.
hm. .

N0.

00.

o...

0N.

no.

a.
on.

Gm

9:0. we.
wvd mo.
8.0. we.
mm. mo.
00.0. mo.
00:0. we.
own no.
wfiN mo.
05m we.
mwfl mo.
00sz we.
3N no.
and mo.
5... mo.
own,“ mo.
8.0. mo.
no... we.
:82 2
EDGE

00500.30 .83 030» 0.. .02: 0. .03.. 0.103 Eco. 0......

0.3. 020%
0.Ec0. 05 3.0030 0.... .00.. .0: 0.. Eco. 05 .0 EBEoE
05 .0 0EO0 000.30 0. .. :03 00E: 33E 03 03:...

3:25.93 :30. :0..0E003 0. €0.00 .0 30.0 .3.»
c 5.0.. .00 0. m:....3 0. .000.» c 3 E30. 0.... 30.0 0. ..

0.5000
:0..0 Eco. 30 .0 0.00E0E wcoEc 85:00 02.08.80

000.» 03009.00 c 0. Eco. .50

:00EOE 530 0. 033003 03 35
0:25.00 .0. 50300 EopEoE Eco. 60.3 0.00.0000 :25»

00.03. :05
8000.00 5.0500000 80:. 0. .050 2000 0.0.. EonEoE Eco F

00.cEEco. .35
352 0.0...00E00 0.... 0.08 03 :30. 0.5 .0 flopEoE 0......

00200 on. 0. .000: 35 0.5.00.0 ...cB=E..
3 E05 0300.00 EonEoE Eco. 60.3 E03000 :055

03053000 .050 E0... 0.00E0E
Eco. 00.0....0 :0...0 305300.. 0:0 80... fionEoE Eco ...

00500008 .03
3:03. .c:0.000..0.0 0.0050 :08 053000. 0.00EOE Eco F

0:00E0E
Eco. 0050 05 :. .0:... .23 00:00.0:00 nosE 03... .0: 00 .

:30. 05 w:oEc 0.008 003...... 2.8.0.... <

0.008... .0. 0c :03 .3
.050 05 .0. .00 00.00. 0:0 :30 35 305. 0.0..E0E Eco .0.

as. 05 .o :8 £8. .5 . _8. _

.050w0.
02823.0. w:.....03 .000» c 0.3... Eco. 0.5 .0 0.2.EOE 2....

000..» .3083 c 0. Eco. 02...

5500.5

318

mm.

5m. .
.5. .

00..
Gm

.80“. .0505...

NNd mo.
0.50. no.
ofim mo.
2N no.
Go... 50.
wmd 50.
ohm mo.
mm... no.
3.1m mo.
3:0. no.
50.5 no.
55...” no.
:82 z

8. .

9...

50.

5N.
av.—

mm.

om. .

.5.

R. .

an...

X...

on.

2.0 8..
00.0 8..
00.... 8..
a... 8..
00... a...
8.0 000
00.0 a:
0.0 00..
8.0 00..
8.0 8..
.00 .3
.0... 8..
00.0 8..
3.0 .3
00.0 $0
2.0 8..
:00: Z
.0:—:02

no.

N"...

on.

on.
on.

mm.

55..

S...
.0.

05..

we.—

Gm

«NW. 8.
ohm 50.
N5.N 5.
8.0. 8.
3.0. S.
0.50. we.
wvd we.
3N mo.
3.5 we.
09m we.
mwd we.
50.5 we.
ohm mg
9.0. no.
. .5 we.
55...” we.
:82 z
.0080

800. 05 .0 2.00 05 .0. 000.....000
8.08 0. 00.5.3 2. 5.09.0 800. 05 .0 0.05808 05.

085 05 :0 0.050 5.3 0:5..03 :05 .050.

:30 ..05 :0 03.5 05 800. .8. .0 0.05808 0053 .. 0.... .
0.050 5.3 3.03 0. 0.00.. . 0.053 com 0 05 :05 .050. 000.0
3.03 :00 . 0.053 50.. 0 05 .050. 5.003 . 60.2.0 05 :0>.0
000.0 0.53

0. :05 .050. 800. 0.5 :0 0.050 5.3 0:03 0. .80.: .
.005 05 0.. 0. 0>..5 05 00.. 800. 0.5 5005 05... 0.00.. .
0.0580098 00:08.0..00 0505008090 .00

05000.8 .0 0.008 .05 .96. 0 .0 03.03 0.0.5.505 800. 05....
950500... 0. 800. 0.5 .3 5.0. .00 E050 x0. 0:...
0.050.898

00:08.0..00 080800808 .008 0. 0. 5.09.0 ~£00. ..

.33 =2. .95. 0. as. 0.5 .0 £8. a. :8.» .o .26. 2:.
50.0098 05 :00 00 5.0: 00 3.03 0.05808 800...
0808:3000 800.80:

000.0: .050. 5.003 0.30.0 0.05808 800. 0800 8083.000
800.80: 0 5:0 808:0.000 800. 0 503.05 00.9.0 0 :0>.0

8085.000
0.800. 0.5 .0050 500:0 50:.00:00 .0: 0.0 0.05808 800 h.

3.03 05 .0 0.050 .5... .0505. 0005 0:0...03
..05808 :80. 5000 .0 52.000. 00.00. 05 85.800 00.. :05?

8:02.000
0.. 0.:. ..0....0 0.08 30.80.0500 .00 :00 800. 0......

030.0080 05:... .:080w0:08
0.03 :. 0.585.000 0.. 8.0..00 0. 5.0.. 00.53 800. 0......

800.
0.5 .0 0.05808 05 b. ..0...0 50.003 0: 3.00:? 0. 0.05....

0800.00 550000000 0 05.030.
0.0..0 ..05 0.05.5:00 .0: 05 800. 0.5 :0 0.005.255 080m

0050000

319

80.308. 3080.0... u b .5... .< 80:008. .07.".

Gm 000$. 7.
.30.. .0500...—

55..

oo. .
.0.

NN.
8..

50.

50.
mo. H

mm.

8..

No.

..w.

0...
Gm

2....

«0.0.

5%.

.30.
Eur.

0.5.0.

wow.
50%.

:5

Eur.

000 8..
0.... 8..
00.0 8..
000 8..
000.). 7.

.05802

85

005

OS

03.
wwv

mm...

ca.
005

85

85

um. . mm...
o... 50.0.
3. no.5
8.. 55¢.
00. mm...
cm. 55.0.
mm. 50.0.
00. 0.0.0.
o... one
50. 3.0.

0... $0. 8.
00.. 00.0 8.
00. 00.0 B.
00. .00 8.
Gm 0005. Z
.0500;

wo.

wo.

mo.

8.
wo.

wo.

wo.
wo.

wo.

wo.

080803000 800.

:0 5.0. .00 ..0..0 05 05.030. 5:0 00.00.05 .05. 80.0.0 <
50.0 «30.00500. 8080.000

.085 ..0»505 .0... 0.085.000 800. .0 0050.088

05. ..00000 0. 000.0000. .0:0..00.00w.0 .0 35.50530 0....

00002500500
0..--.05.0w0. 5.03 0. 0.003 5:0 0.00.0.0. 800. 05. :03 30:

800. 05 00 8.08

.0... 0.05808 005.350. 05. .0 000050000 .000.000.0.0 05...
05.05050 00:08.0..00 800. 3.5 .0 00008.0 05...

0.00. 0.800. 05.

o. .08.... 8:80 000.20... .58 =00 05 008.39... 2F

800. 05. >5 503050.

05 0. 0030.050 00:08.0..00 .0 000.0 00500 50.0.0 5.00.0
0.50.0500. 800. 0>..00..m.

00.58.50 500

0.00» 0.800. 05. 0. 80858800 500 .0 m:.500.0.0500 :<

.00. 0.0.00. 05.
505080000 0. 5.0. .00 ..0..0 .0 .26. 0..05808 800. 500m.

Noah—00:):

mag—U3. mwMDUDm dog

:30 ..05. 00 .00 0w:.5. 5.03 0. mg... .0 500.00. .050 5000
5.3 0.0.00000 0. .005 ..0... 05 5.0050 800. 0.5. .0 0.05803.
. 05 0. .003 ..:05 >05. $0.5.

05 0. 0305 0. ”0.00 00850800 0.0 .05. 800. 0.5. .0 ..00 0
05 0. w0.0w 0.0 >05. .. .05. 0.0.30 0508 05 5.0050 0.05802
.003 .05. .053 .0» 0. w:.0w

0003.0 .00 0.0 .9... .05. 00.5000. 5.0050 0.05808 0.00...
0.053 0 00 800. 05. .0

00.00 05. .0. 000.....000 00.08 0. 0.00.. 0. w0.0w 0.0 00850800
.05. .05 00.500. 5.0050 800. 0.5. .0 0.05808 05...

0050000

320

00500.08. 0.800. .00 80.. 05000. .05 5.0005 .0..08
0 08000.00. 80.. n 0 80500.08. 0.800. .00 80.. 05000. .0... .50005 0 .00 0. 80:".

Q5 000.). Z
.08.. .0000.me

55..
55..
.5..
0.5. .
05..

55..

0.0..
....
0.5.
....
55.
05..

on.
3..
Gm

2.0.
55.0.

50.5

05.5

00.... 0:.
0.... 00..
00.0 8..
8.0 0.0..
00.0 8..
8.0 00.
2.0 8..
.00 0%
0005. Z
.05802

00..
00..
00..
00..
00..
00..

555

55..

.5..

..N. .

00...

mm.

55..

0.5..

55. .

00.

50.

00.

0...

50.0.
05.0.

no.0.

00.5

50.0 0.0.
55.0. 50.
50.5 50.

0. ..5 55.
00.5 55.
N60. 55.
010.5 55.
55.0. 5c.

:82 z
008..

50.

5o.

55.

5o.

55.

no.

no.

00008.0..00 800. 50000.00. 0. 00.055800
.05. 0.05808 0.00. 0003.05 80505050. .0 00000 <

005005.056 .00 0.5.3 0808:0005 80.0...5
.00.... .05 080.580 .0 0050.000. 500 0050055005. .2000

0.05.05 00500808800
.0808..0005..08. 500050. 0. 005 00.80.000.800 506.08.

800. 0... 00
0050800000. .0005000. 80.0...5 .0 0000005 005008056
.00 8055095 9.0.0.005 800. .0 0050808208. .300...

05.00.. 00.0.005 506.08.

00.00. 50505.0 8.0.00
0. 050508 500 0003 300 0.5.0600. .0 80800.08

0.00. 5005.000 0.0.0800
.0 80.590 0 2500 0. 505000. 08.. 05 0. 0050050. <

ZOE—DEE 40.7.0575..—
.5506 20%... mEm—me Elm—HO.—

b.85.0.0 05.00.5005 000.0 0.5.3 50.000. 0.05808 800...

800. 05 05.0.00 0000.5 .0
0.005.550. 5.3 05005000 00508.0.0. 0.0.0000 500 508....

0.05808
0.00. 0003.05 0500598 00508.0.0. 0.0.0000 500 0.08....

00000.0
0.00. 05 05.030. 808050008 05000.8 .0 ..00000 05...

800. 05 00 00.50.0000 .0 0005000.
880...... 05 0003.05 0.000050 00500808800 0000

08000.00 0.00. .0.. 506.000
808050008 805.3 80.0.85 0.00. 00.08 0. 80500.. 05...

00008.0..00
805.550. .2500 800. .0050 0.005500. 50.5003.
800. 05 00 000.0 .05808 5000 0.0. 0... 50.500.03.00:

0050000

321

$365383ng 8 95303 58 a .33 2a Sufi. do? E9, ..ofiowo. x53 03 u h
”mwficqfiflovasmmfi go @803th Eofiflom 2m 82F .350; E9, uofiowou x53 oBHfi

Gm :82 z
.822 BEoiH

Wm. g

mm. u
mm."
Q..—
3.;
on;
mvé
3.4
mm.“
"NA
NA
ea;
3..—
SA
mu."
om. fl
an. g

and RV am.” 5N
02w< zfiaobmuh ”09%,“me bwaobmufi
36 no" on.“ 56
wow. 3; m: 36
N3 2. :2 av
90v mo 5.." 3.6
mmé NE on; vmé
£4. a: :2 c3.
NS. m2 we; 86
8... a: :3 3.6
own SN 24 can
end mum MNA mnm
can «8 84 «aw.
8+ m2 8; 5m
em... 02 m2 3w.
Rm. 0% m2 ova
v3. won 84 8w.
mfim mmfl ma. wv.m
$6. 94 no." Sun

rd;

a.” N5. 8?
:4 one. w?
«.2 a? ma.
Gm :32 Z
“3.52

:4 new. 8“
s; as we
92 ns.. 2:
Gm :32 z
.533

R:

«a
ma
3
man
8
on
on
on
8
00
3.
on
mv
mm.
we
on
3

87.8 .5 358563 9:83..qu .8 8508 E0650
ca .0: g 585% ES: memes» 35:88-30; 25.

mag—34.19 44:230.—
BBOEBP ouca>v<

3832.5 .83...

wc=bm

mg

cognizan—

wctoocmmcm ems—8m

E08833 98 £838.:
matey—.82

mctoocmmcm goongctsfiflzcaz
9.1005ow .269.—

mccoocacm case

w:==:o8<

005:5

wet—.8355:

.928 5:303:30
Eon—owns: 2.2.032

wcfianea

Egg—(amn— u ZOE—DEE 59—.

3:8 8 new 8538 .26th .5303
..80 «820586 3080328 Bow 5 EoEoSEE =<

08: 060 an

USomS .5995 E“? 98 £23889”: Juno—BOS— 8.35

cocfimcawuo 50 £55, £050

9 032908 was.» E 358563 88. Ho EoEnm=Q808<

553:0

322

Gm :32 Z
.83. .0508...

8.. s... a...
a... a... a...
9.. .2 w...
9... 3;. a...
on. 8.. a...
3.. N... .3
8.. R. a...
.n. a... a...
9.. on. a...
m... a... a...
.0. $0 5..
c... a. 5..
8. 3.. .3.
an. 5.. 8..
am :32 Z
.3805.

3... mad co.
3... vwd so.
Nm. . ..NN no.
3.. 8..“ no.
.N. mvd no.
3.. .w& 8.
um. and 00.
F... EN 8.
mm... Sun 8.
mo. nmgm no.
cm. va 8.
mm. . SN co.
3.. wt. no.
5.. SN 8.
Gm :35. Z
hO—ug

0.2.808 83. 503.2.

03.0. .0 085.98 :000 0... 0.8.... 300.26.. :02... E3. 2..
:0 02:03.00. 0.: 0.0.6. 8.05:: 80:38:90 80.0....0
b.0070 00.02800 .0 00.23.03 .0: m. 0.2.808

83. .0 8:08.300: 88.00 ..0 :0..:£..:00 .0 80:. 2....

80.0 .0: 0.: E3. 0.... .0 002.0030 .0:: 0.00» 25.

302.000.. 00:08.0...0: ..0... 03:08.. .0>0 «88.8.00
3.3.00 0... 03 :03 30.. 08. ...w...:. 0...: 00.. 83. .00

3:08.300: .0 80:08.. c0030: 00:3... 85:38:30
0.88:0 0. 0...: 0:00 0.0.. :0..0a.0.:. 180.858-0020

83. .80..0:....mmo.0 0.... ..0 33.00.30 .0
0.00» 0... toga... .0: 0000 .0... 8.30808 0 00.. 83. .00

08080.33. 00:08.0..2. «.83. .00 3.00:... .0... .05. a .0
.0: 0. 30.003. 0:: 8088.80 ..0 808...:800 808035.).

..:0 02.02.00 0 08000.. 0. .0» 0.8.. 03 ”8.00. 0.0.30.
:0 82.03 0.2.808 .3232. .0 :0..00=00 a m. 83. 2....

00.8.0.0“... 850:8 0.8.00. 03 :0...» E3.
.00 .5008 0. 80.030. .0 30.0 0... 0.2.808 83.802

80.0.8.0 .088 8.08 0. b.5500 .0 .030: 80.08
80.0.00: .3. 0: 0.... .003. E3. 2.. .0 E3. 0... 5.5500
86.00.. 0.... 0.8.. 3.00.00 83. 2.. 00.0.00 82.5

.88
9.0583 «20.0.35 EC...“ muons—~08 ENG. CoguOD «maumma

E3. .00 ..0 0.2.808
50.20.. .008 0.0.8... :0..3.::8800 8:030:30005

0:0... 09:03 0.2.808 E3. .. :0... 80.0.00:
8.03 .0 02.... 3.0800 03 82.00.08. E3. 500:;

E3. 0.... :0 0.. 0. 000000.... 0.: >0... 8...?
.30.... 0003.50 0... 8:08.300: 820...... 2.. 80.. 0.2.808
83. 80.00.88 0.0. :. 3.03. :0..0:.0.:. 82.83-0090

8.33:0

323

Gm :82 Z
033— .aEflxm

no.

85.8 aggaaooum ”83,8 .2 2 82".

mm. .
.mv. .
ow. .
or...

on.
Gm

av. vwv cw. mod
on.“ mwv 3. 3d
3am .3. mm. 3....”
wrxm .3. mm. 2...,”
$6 8v ow. cod

end 93 NR... wad
wo.m ewv .fi. mfim
SN NS. om. cod
.N.m va NV... mvd
mud vwv 09. mod

gamut :2 2 “oz".

San wwv mm... and
cod awv 3... m3”
moN wwv mm... owN
NV...” owv 3.. Had
wwd awv or... EN

:32 Z Gm :32 7.

00:50.). 0030\—

wo.

wo.
wo.
wo.
we.

00.

so.
00.
0o.
no.

no.

no.

no.

no.

no.

30.39% >0”.
was :50. 05 c0922. 0wcmzox0 8.38.5.5 .0 5.13.0 0......

E3. 05 0.0.330 munch» 8 30:22.05
new 83. 05 503.2. Human—.38 8.38.0.5 .0 >526 2F

0.09808 :80. c0052. consume—.8800 ..o 8050:. RF
303808 :80. c0932. 0wc£0x0 5.38.8.5 .0 5.73.0 0......

$5.008 :80. .083. .0 3053.8: 2F

”as, aofiéaémfiuofieéam

«Enos—5.0mm Emu. $38.0 .5 E0302. 00.0.0.
800. 0....003 028 o. $5.008 3.53:8:3 532558.
c. 329:. :80. So» a. .0922,“ :0 52.05.00... 30...

2:085 on
o. .00—302.8 0.5 Eon—:08 88. ..a £023 a. "9500:. Emu.
.558 .0380 E3. 50.. 80.0 amp—02w :0 5:53.00... >62

:30. .50» .0 02,050 3:9.» .0 £322.05 5.3
9.3808 :80. ..a wcoE<
309808 88. 0050 .98 can :0» 50.3.8.

”8808 .8 83m?— aoztz, .0 3:03.000...
ZOF<U~ZD§OU Em:

82.3.00 800.

.9550 8 095008 :30. 0.05806 E05808 E8. 50:00
3:059me E3. £2000 £0339

Egg—=88 ..o taco ..o “.00. a .325 £09808 E8...
8.820.... .05 cozaoisfiEg E8. 03.00....0 30:0

9 520:0 08.0 2.3.3.3 00.80. .0: 0.8 €2.82: 88%
80300.0 E3. 00:02.5 5

83.2.8 :80. .9300 o. 380:0 :50. 05 023.5 flown—5.2

$585.80
$5.6. 0.... £085.00? comma o. 0020:0on .0 ...2m
630.165. 0... 0.00. flonEOE :80. 0.... .0 080m .30. .<

5.60:0

324

co. 8%. 5:
mo; coma moH
mo. 006 m9
mm. H ova. mow
vo. mow. mo—

24 NN.r. m9
R: and 2:

No. mud NS
3. mud No"
ma. flaw. m8

880.00% “SF 00 Emma—0:05.. 30,—." N. 60583“ 35. 00 00080.1 30F":

Gm 8002 z
0003— BESxm

on; SM 9%» FJ N30 cow

8. _ 9.6 cm». Eu _ 09m co"

8&< bwaobmn b “00093.5 Ewaohmufi

(«A ovum bow mm. a mod no:
mm; 36 3V RA vfiw me
an; 86. 0; 3. Now. we"
we." N3. mmv SA wfiv we
no; mmfi. ewv m: 9.6. no—
84 awé owv an. — vwé wow
$.— 86 $6 3." 26. :2
84 $6. wmv a. mod Nod
24 3%. omv 00.: 3.6. vo—
BA and ewe «NA 0N6. we"

:2 w: wwv no: «.5. so:
a: mwgn wwv m: 36 no:
swans :2 2 “oz":
K. Sum mwv mm. mud m9
8. Sam ewv No. end we:
3.. 09m owv 5. Sam mo:
mu. mud awv 2.. wmsm mo:
0.80305 02" m $80305 Emogawmmnfi
am :82 z Gm 882 2

000802 000004

3:08..&_mma MES: 05 0. 80838800 :32 a 83500?»

800:
mi. :0 0880800: x8 0.0296 .0 0300—30500 .0: 809

EUR—Mg 23mm
977.. ZOF<DA<>H Egcamm

5:080:30 002005

toga 80809808 0>::0owm
80880880 0:03

Essa: 25

00380005 0102 000808 800%

8050 80c 20: 05 000308 00::00m
toga bflowgm

802003 :00 £010.02

8080300 93 £00k

00:08.85 02200.03
KHz—30mm“— A<ZOFH<N~Z<O-O

«00:08:89.8 .0 «8030.5 020.0. 800. 5000
800. £5 no 800808 .050 5? 3:08.88 ..0 b.0883
2008:8800 02—8080: :0» 00 6w80>a :0 5:00:00... 30:

«0.808%5.“ 00 080305 008.00
800. 8000 802 05 028:0 £500» .0 30:23.08 55
2008:8800 £80800 :0» 0.0 dug—0.5 :0 0:80:09.“ 302

 

3.83 E .33

802 0:0» 00 023:0 8:00» ..0 £32208 55’
800808 800. :0 $58.4.

800808 800. 0050 80 0:0 :0» 0003.3—

”m80305 0030800000

558:0

325

Gm 0002 Z
00:3- .00.:0:xm

«0.: 3d «0:. we: 23 8:
80:00 0:680:80": m80:3: :02":

3.: 8:. 8:. an: :0... 8:
00: 8.». 8:0 2.: 0:0 8:
a: 8.0. 8:. 8.: 8.0 8:
8.: a... 8:. on: 8.0 8:
00: :00 N8 8.: 8.0 8:
8.: S... 8:. 8.: $0 8:
8.: 8.0 v8 3.: 02 8:
3.: 000 N8 3.: a... 8:
8.: 3.0 00:. 8.: 0.0 8:
8.: 0:0 80 8...: 000 8:
9.: 0:0 :8 NS :0 8:
3.: 0:. :8 8.: 8.0 8:
00: a... 8:. 8.: 3.0 8:
Gm 0002 Z Gm 0002 Z
000802 000004

00:08:83: 800: =0:0>0
:2000 003380 800: 0: 00:35:00 70:22:00: 50» :00:
80.00000: 0:0 00:030.: 8000008090 0023 0: 8&0: 2:.

80:00.30 :00 0:00” 800: ::0 w0_:008 0:000:00
:00: _0>0_ 0 :0 ::0.:.:0 800: .000 8:22:00: 00w0::000m

800: w::0::00 :0:0_:0::.:

000:0 0E: 00 b90200 3:080:30: 0: 80:00 0 5.0200 0:
08085000 800: :0 ::0.:.:0 70:22:00: x0_ 0 80:000.:
0:::0:::0 8000008090 800: w0_0::00

30000080090 0:: .:0 80:00: 0:: 0:3 0:0: :0 ::0 0:
00:8 b0> 800: 02: 00 000005000 3:00:00 8 00000::m
080803000 808.000 0: 00:09:00 00 08085000

0.800: 00: 00 0000080 :0 £303 :3: 0:: 00:50::
00:08:83: 0003800 0.800: 0:: :96 00:08:83:

:0 ::0.:.:0 :3808 800: 3:23:00: 00000::0 >_:0_::m

0:0080w_000

0.800: 02: 0: 0005000: 000 08:: >8 :0 :0_ 0 $00.60
:0: 00000805 :0 0000000 0000080: .9200 0: 00006::—
0050008 00:08:83:

88: 05 08.20:: 8 808388 20:: 2:: 80:02:50
00:08:83: 000 00:00 800:

:0:0: :0_:00:0 :0: 80:03 :0 00.08000: 0:830: 8030::

:0_::£::000 :00 ::0:.:0 :00808 800: 8:00:00:
:0_:00:0 :0: 80:00: :0 0038000: .0::030: 00:30::

::0.:.:0 800:
:0 00:22:00: :00 5:3 3:80: E00: :00 0:0 0:3 800: 0::
02080 3:03:08 :3 080800003 000083.00: :0 00:3:

000088.000 800:
:0 8:03:00: :0 0:200:00: 02:00.38 00 00:8 00: 00:3:

00::00:O

326

3.: 80:: 8:
«.0: 3.0 8:
:0: 8.: 8:
:0: :3 8:
00: 8.: 8:
:0: $0 8:
:0: 8.0 8:
:0: 0.8 8:
on: 8.0 8:
Gm 000:2 Z

00:00: 80:00:”:

0w0:0>< 0>00< 8:": :0w0:0><um :0w0:0>< 30:0m :00":

00:

0.0.:
No.0
00.:
mm. 0
3.. :
on: 0
NY:

20000. 00:00:80“:

5.:

00:

00mm

0&8

Gm 0002 Z
000802

$6.

0N0

:00
3.0
00.0.
~00
no.0
00.6
0.00

8.0.
808050 0:58.":

00:0

0.9:

9.0

mwv
30
03.
owe
vwv
0.9:
0.9:

mm».

0:0

0004: 30

008 wwv

wma:
00::
mm..—
003 0
F: 0
3::
am. 0

8.:

mm.

:0.

.08 0.8 8:

.08 .08 8:

am 000$: Z
:00000

and
mmd
2.0
00.0
30
8.0
0.80

3.0.

H 5 ”0:00:83 N :

000

3.0

we:

mo:

mo:
8:
wo:
wo:
wo:
mo:
8:

5:

mo:

”00:08800 :0Zn:

we:

0080:0003 0208800": ”0080:0003 :02 on":

0:090: 0800000 0:00: 000 m0:8:: :008 0: b:_:0<

800: 00: 02080
0:80 05 20880:: 5:3 0:83 2 88: 2:. :o 802

8:300:00 80:00:: 0088 00:02:00
0:080:08 000 2008:8800 0: 800: 00: :0 .0:—:00:

:00000:0: 800: :0 30:02

0000828 800: :0 80:00:...“

0000:0000 0:03 :0: 000383:

800: 00: >0 0000098: 0000: 080.608 :0 :00800 00:.
03:00:: 0:03 :0 808000 :0 0000:: 00:.

03:00:: 0:03 :0 8:080 :0 3:800: 00:.

0:900”: 00:08:83:

02:00— mmfiZm—EEH 25:.

0000: 80000030080 0:0: :0 0:0
00: :26 80800808 800: :0 b:::00:0 00: 0:00:00: 0000::

0:00» 0:: 000 800: 0:0: 0: 80808800
80» 0:02.:0: :80 :00: 0000:00: 00: 0:000 0000::

800: 88800 8000003890 0:0: 00 0:0:
:00» 000:0:000: :0: 00:03 0: 00:»00 00: 0:00:00: 0000::

000380.30 000 0:00“ 0.800:
0:0: 00:80 0: 00:03:00 00:00:80.0: 0: 0:960 08000
«a.» :00: :0» 00 08:: 00:20:03. :00» :0 800:0: :003

 

000::0: 08:: 88:00 0 :0>0 00:80:00 0.800:
0:0: 0: 0:030 :0: 00 08:: 00:0:0:.00_. 80: :0 800:0: :00?»

mZO—Pmflbo 3“:sz
00::0000

327

ohm

vim

$.N

and

SN

wwd

Qua

SN

98

mwfi

OWN

MEN

co;

EN

EN

hm.“
EN

8 803585 venomoxm Sm u 0 $0280.30 ho maocfioomxm $22 9 03.25 >=SOHHH

mo. m

Gm :82 Z
..83— 353.3“

8%.
wnm
3.0
on”.
moi.
no.0
No.0

8.0

Nod.
36

:3

m2
mo—
no“

v.2

:6

3N

96

mod

3d

36

:6

$.N

ow.~

mod

2N

waN
NYN

QWN

mod

HNN
Rd

NW;

am :32 Z
.5952

Saw

mod
:6
cow.

8.0

w:

m3. 0 fin
53. God
n9. 3d
no». r. fim
o: and
m: mN.m
v: m fin
on". mud.
at. mod
a: oo.n
w: 3.N
02. mod
a; «fin
owv on.~
RV m3.“
mwv mod
mwv cod
8260.30

5.. _

mm :32 Z
.5334

m—d
"7
V“

mem
86
mwfi
ohm

3...”.
and
vflé

96

09

8—

we
we
we

52
co"
2:
be

2:

we

we“
we

we

cod
we
2:

R:

03 FEES
.8 9.3—«Ea .3 .60—canon: 30: Ho EoEQEoBv “0.8m

3260.5
35?: 5:28 use: v3.32?— uo cozgvhmvcsm togam

F980 836a
mam—.6 cosaamgtaa .3158 v5 9:958 230 roamsm

$093 cos—38:8 33 bang 9.622
82388 085:8,“qu ..ozaazm @2260

$2928. 05 2 888 96: floanEoo 282.
3205.8. 8685 van .2605 >6: 2 9808 025E

82.262 bo.co>£ toaasm
8653888.. 33 3&3 265::
count—8 how—n53 850.5-.8n 2620?.

gums. 3:8 3 :80 63m .2293 .od $098
35858th .2695 30: mo 80896an 05 toga—am

8382? $5 80539 toga
.3026 35 8&326 «52:8 50803099 8.969

.3396
can 3.95 on. 5922. 08: 0.950 unto—Bo 3:89: cos—vow

83.333 .3238. 8:053 ..o 8: 3329: 2: roamsm

9.3—Qua
72:25 go 3525 So... 09885“qu bog—0c 26th—

$260K
Bu: “8 08: 29A... EoEQBoBU .2695 3269 9.95%

5:26 80: vows—~23 963::

.80 Eu: 33:82. 00.60%

”9 ban? Macao?

82.88qu
082.8th macoenuwacmfi 02.398 808 o. 5:34.

538:0

328

m...—

on.
8..
Gm

s$000.3. .505." b $0.063. 02 u .

0055.6 .062 820,0..."5 3.0555 02 8630.5".

cod mo. 2... $0 owe
8.0 No. 3... NW6 2...
and no. Nd and 3..

0002 Z Gm 0002 Z

00.3— .0000.me 0000.02

8.. 3%.
$4 Rf.
3N cod
3d 36.
05.. ovsr.
2.... man
mo. ngv.
mod wmfi.
SN am...
no. no...
3.. 3}.
mm... min
3... Sam
3.— no...

0... fl
3N

Om

3.0

2.0
8.0

0002 Z
.6004

3
mm

no
mo
wo

ow
om
mm
vo
mo
mm.
55
no
55

mo.

no.
8.

0.0 $0080.63.

.50—00.. Jag-.5328 3.00835 5.86 00000.3.

30.00000. 3 003 5666 .2605 80...... .2. «00.086 8.02
«90.0030 600 «.00» 8008.885 «.83. 0.... 0000.800

832.00
E3. .3550... 0. 80.008. 3.0.00. 60.. .0600 0.30:...

.68. 0.3. o... .023

«000.086 00.828 3:553 8.02

mZO—m—Um—G EOE... 5MP

80:060. 0.0.. 0.0.6 3.06.0 3.0.0:. .5555
0.02.0060. 0.0.. 0.0.3 62050—266 .0605 300 ..0550m
00.02.00 0386 6.605 8.8.05

mm0_.0o.0 0.3. w0m=6 83. on. 0:3 .008 £90.05
«36. .80—0008. «8005 .0...O

302 .0605 300 .otO

5.03.06 3.0.08 26.50.. 0. 80:00 .5553

36. 60.00.6500 >52... 630.5

«.32 0000.60. .80 630.5

magaafi BZEZEAOE Magma—3m

.58. .3000

.0 .0835 o... 0. 0.05.000 b.0850 0.3. «E. 806 >6:
82.0030 600 «00.6.0055 0808.05.05 .00.. .520. $00.0
.00. 0:8. 5.0.08 3000003380 .8565 00. $008.50.
go... ..0 0.08 a 8. . .V 0.0.. 0.00.. a 00 .00. 00.03.».

wou:00..0...05
€030 $000. «2. 0.0. :0» 6.003 300 405.00 0058.05.05
600 40008.05 8. .83. «E. .0 060.305— .:05 00.00

00:55.... 0.5 «520000.00. 058...... gazeswm

03.8.5

329

on.

3.
mo.

no

nu.
mo.
3.
oo.
3..
cm.

8.
mo.

ww.

Gm :32 Z Gm 53: Z

.83— RES...“

9:
ms;
4 aria
2..—

86
ov. _
cofi
8N
no...

«Em
and

8N

RV

on”.
m:
nmv
e:

on».
wbv
53.
>3.
2%

w:
8v

5:

@332" w "3389.80" m ”.262 u H

.2. :82

mm.“
“SN
mm;
2.—

8. H
on.

056. mm
3.6. mm
mvfi. m9
cwfi. 8H
cod 2:
$6 2:
«fin No“
:82 z

ho—ug

mwc=o8 :2: matoEmcoo 305:3 flonEoE E3. of 33...:

82: £5 .96
2 96:3 1....“ 305.3 «.3808 E3. .0 8:2. 2: 89.20

own—Eu .8 mcozmowwam 38%;—
Son :25 90.. a .8 36.96. m_ .0982: E8. 5 :2: 80m
.33 :30 .0: .o «E 2.8. on 9.23.96 8;. min”;

3.30%.:
.33. 05 8.3:. g a :96 «.3802. Eco. .8 a: "653m

«.050 ..o :58 E «.3808 83. 8&6:th
flue—poi ficofloq 55 «.3808 EB. «Eu:
ow :3 93 :3. 0.0:. 8:950

883.03.: 2 and... an

8.. n8» a
88. Eon—=08 :88 05 .0 0:0 cock 5238.9? 836.me

«.3808 E3. 05 .8 E028 230.com 809

.0;
.0 82 2:3 8&8? «.3808 Ewe. 5:3 E 03» o. gamed

“.2234 88% 2E.

ngmmmma‘. filmy—mam..—

uéwofibm con—«come: VS. 38.38 @2260

835 820.3 5:83....“

v0.33. ma $32.08 88. 32. .00—om

3:08:38... 83. ..oamam o. «.382: Eflxéoc zo=om

858:»...3 US.
8:533 ER: go coco—9:8 on. .8 £528“. 9:8: 832

amazon»... Ewe. Escozum
2.25.38.
.0 228:8 2.0 5.5 0.0:. 88.? 35 9.038.. 832

5.53:0

330

Gm :82 z
8.; 850...”:

2..

3.
mm.

on.
ma.

ow.

mu.
m...

mp.
we. a
mo.
2...
am.

8..
3.
2..
or:
3..

8.
Gm

m3 8..
2... 3..
.8 E.
a... E.
a... E.
a... we.
.8 8..
an E.
8.. ME.
mm... 2..
.3. E.
.8 m0.
.3 8..
2... E.
0.... 0:.
8.. E.
80 0:.
5.. E.
a: 2..
.2 E.
a. 0:.
N2 8..
:82 Z
8:802

80.. 0:0
.0 0: 0.80: 88 30: 00.08.. 0. 0.2808 88. .8 0.83

..8..0 .383 0. 0.2808 88. ”03.03.30? 88.0000...“
0:8. 0: 230:... :008 30: 80:0 20,—.

3.03 .000 8305.0

8:38 «.83. 2.. .2... 22.80

82 30: .0: .0 0.: .20 8:...

0:03 08...0>0 8mm.=00cm

08:0 9:0» 0.80: 0.0.88
82.008. :0 0.2808 88. .0:.0 0:. ..0 .8689. 0:. «.00

008.000
0.8 0.00808 88. a: 0.008 0.0 .0:. 82.838 0.?—

00:08.08 £8 0: :8 0:.. 59.0.... v.—

.0: .0

a... 5.3 9.82 8:3 3.3 a .8. £388. 58. 2.. 8%:
885.0 9.2. 2 05...; m.

86.2.8. 58.5.. a :2. .28. .8 90.8% a 8.0.5
13.00 .0: .0 0.: 8 8:808 88. 0:. $8....

8.0. 30: .8000 0. >5? 8

80_.0a ..0:. 8 0.2808 88. 0:. 0: 8.03

80:88
88. 0:. «:08 0.0.08 5.: ..0 0083.008. 0:. 883m

.8... 0.0:808 88. 8:8 0:. was—=88 50:..3 m.0<
88.00 .0: .0 0.: 80—08 0. 8080M

0.8.88 0 8:08 0:3 8:808 88. 0:. :82?

88:. 3:80 ..0 083 888:0 0.88M

8.05:0 .0:w_: 8

08:. :23 8:80... 000» 8 0.3808 88. 0:. 000.. 0. 8E.

:82 Z 838:0

hO—vg

331

0>..00..m. 2.0.0080 88.00am. 0.0.0.0002uv 82.08.05".

Gm :85. Z
.85. 08.0.me

.N..
m...
8..
8..
..NA
8..
5..

m0.
am.
00.
m0.
.0.
co. .
0....

0.0.
8.

co.
m0.

0...
a...
Om

8.0.
mm...
36
am...
8.0.
36
as...

00.0 8..
8.. .S
9... E.
.8 0:.
8.. 0:.
a... 00..
a...“ 8..
8.0. E.
8.0 N...
s... 08
0.... NS
.00 a...
8.0 E.
:82 7.
8:80.).

2...
w:
m:
2...
cm».
3.».
a:

Gm :85. Z

hvfig

8:08:98...
88. .0 00.0.0 0:. 88808 0:0 8.00. 0.0...28 8.880.000

88. 0:. w:..:0...:00 00.00.30 :83 w8.8n.

3:000:00 .0 38:88 0 .3 :0..08800 88. 08.8.6.0
00.8.8. 00 80.008. w:..:00m

:0..00..0 0:0 0:08 88. 8.80.802

.0....:00 .0808. 880:0:

8080208. 8:88 080.808 08.308

.08:0>..08.m 0880003

0808:8000
88. 0.0.0000 .00 0:80.008. 03 .0:. 00:80:00 0.

8.03 ..0 880.... 0:. 8800:08m.

88.080 ..0 w:..008 0:. 8800:08m.

0:00

0: =0:0 .. 30: 0:0 0:00 0: =0:0 .0:? ..0.00 8 000.000
88.0 .88.» .8 82.808 88. 8:000

0000.» .0 88. 800088080 .0:.0 ..0 08:0 w:.0: 808.8

.8:
0. ..:.:. 8:. >03 0:. .003 80:. 00 82.808 88. 0:. 0.0..

80:808
88. 0:. 0: 0008 0:80.000 80:0 008.88 080: :0 0.0.0:.

08030.0 0. 8:08.000 30: 800.0

8.8080
3:08.880 80:. 0. 0.. 0.53 80:808 88. .0:. .. 0. 80m

..800 0.0.6 8 0808:8000 0.00.
.00 8.00 ..0 0003 0.00830 30:8 80:88 88. .0:. 0.0.0:.

88. 0...:0 0:.
.0 000» 0:. .8 80:88 88. 0:. 80... 80.0.88 .8 0:04

8.00. .0...0...00 0. 80:808 88. 0:803.

8.08:0

332

Gm :82 z
..33— 353%..—

Nm. H avgv n:
ma." NOW v:

Gm :32 Z
non—=92

am :82 Z
.5584

£9 «.8982: :03 wcmccou 8%.; wcrccflo

v.13» E8. 098% wcfimznflmm
568:0

"‘TIE@ifl'flnflfliflllfiflmtmwflfiflfflI?“