...n.v Liz-‘01....
.L _ :l I

w ‘ .. ..
it. ”Irina?!

; .

, r .

. 3.3. a. a

55.??? .
z .s. ,1 .

.03

Is:

.5...
I.
a.

.21.
it,

$3.3.

 

 

. . ‘ .
{Vin-Lb. .. f w ‘ _ A: A . .
.anPrzl. 0" 3.6L. V. V Nahuatl-hiss. l. N

THESIS

élcgm

This is to certify that the
dissertation entitled

IMPACT OF INVENTION-TO-IMPLEMENTATION DISTANCE
ON MANUFACTURING PROCESS IMPLEMENTATION
PERFORMANCE: DOES THE IMPLEMENTATION TEAM’S

‘ ABILITY TO LEARN MATTER?
.93
(U
E 3,3 .3
([1 presented by
<1: C a.
‘13 “3,2
Q E c
_.| Q :3
S TEMYOS PANDEJPONG

 

has been accepted towards fulﬁllment
of the requirements for the

DOCTORAL degree in MARKETING AND SUPPLY
CHAIN MANAGEMENT

 

 

{Z .- .
//MJM

/Major Professor’s Signature

é/e/J 7

Date

MSU is an Afﬁrmative Action/Equal Opportunity Institution

 

I -.--.unu-.-.-‘—.—---.--.-.--.---—.---—-.----

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DAIEDUE

DATEDUE

DAIEDUE

 

rm 1 4 2:3

 

ﬁii‘iltii’i

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:lC|RC/DateDue.indd-p.t

 

IMPACT OF lNVENTION-TO-lMPLEMENTATION DISTANCE ON
MANUFACTURING PROCESS IMPLEMENTATION PERFORMANCE: DOES
THE IMPLEMENTATION TEAM’S ABILITY TO LEARN MATTER?

By

Temyos Pandejpong

A DISSERTATION
Submitted to
Michigan State University

In partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Marketing and Supply Chain Management

2007

ABSTRACT
IMPACT OF lNVENTION-TO-IMPLEMENTATION DISTANCE ON
MANUFACTURING PROCESS IMPLEMENTATION PERFORMANCE: DOES
THE IMPLEMENTATION TEAM’S ABILITY TO LEARN MATTER?
By

Temyos Pandejpong

This study investigates factors that affect a manufacturing process
innovation (MPI) implementation project performance. It empirically examines
the effect Of the invention-tO-implementation distance (”0) on MPI
implementation performance. Three types of IID included in this study are
physical, temporal and organizational distance. This study develops a conceptual
framework of the MPI implementation process as a project management task
involving learning and knowledge-creating that requires absorptive capacity. The
primary hypotheses are that distance has a negative effect on implementation
performance, and that the effect is moderated by the level Of technical
competence, functional diversity and psychological safety of the implementation
team. A cross-sectional survey methodology is employed to test the hypotheses
of the conceptual framework. The results indicate that organizational ND is the
only IID associated with lower MPI implementation performance. Technical
competence, functional diversity and psychological safety each has a direct
positive effect on a different type Of implementation performance. Some
interaction effects between ”D and absorptive capacity are signiﬁcantly positive.

The results are discussed in light of their theoretical contributions.

Copyright by
Temyos Pandejpong

2007

Dedicated to my parents, Am and Sawanee Pandejpong.

iv

ACKNOWLEDGEMENTS

First and foremost, I would like to thank Dr. Morgan Swink. As the Chair Of
my dissertation and my mentor for the last 4 years, Dr. Swink has been
encouraging throughout the whole process. I leave Michigan State University
with the highest respect for Dr. Swink's achievements and dedication to the ﬁeld
of Operations Management.

I also thank my other committee members: Dr. Roger Calantone, Dr. Ram
Narasimhan, and Dr. Gary Ragatz. Dr. Calantone and Dr. Ragatz’s laid-back
style and perceptive comments and questions were appreciated throughout the
dissertation process. Dr. Narasimhan has had multiple positive inﬂuences on me
throughout the program both in classrooms and outside.

This research would have been impossible without the cooperation of the
respondents and the funding. I truly appreciate the demands placed upon
managers and executives in taking the time to complete the questionnaires. I
also thank Dr. Robert Nason and Dr. Swink for sponsoring the project.

I would be remiss ifl did not also thank the rest of the faculty in the
department for their encouragement and support since my ﬁrst day of the
doctoral program. I also thank the departmental staff for their warm personalities
and willingness to always assist me.

Finally, in closing, I thank my friends and family for making me the person

I am. My brother, Toryos, and sister, Denla, have always set good examples for

me to follow. My parents, Am and Sawanee, are owed a huge debt of gratitude

for their support in both academic and non-academic efforts throughout my life.

vi

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................ ix
LIST OF FIGURES ................................................................................ Xi
Chapter 1 Introduction ..................................................................................................... 1
1.1 Motivations ............................................................................................................... l
1.2 Deﬁnitions ................................................................................................................. 8
1.3 Research questions .................................................................................................. 12
1.4 Overview of the dissertation ................................................................................... 12
Chapter 2 Literature Review ......................................................................................... 13
2.1 Key Characteristics of the literature ....................................................................... 13
2.2 Past ﬁndings on implementation success factors .................................................... 15
2.2.1 Project personnel .............................................................................................. 15
2.2.2 Implementation policies and support ............................................................... 24
2.2.3 Organizational characteristics .......................................................................... 28
2.2.4 Invention Characteristics ................................................................................. 31
Chapter 3 Theory development ................................................................................... 33
3.1 Model overview ...................................................................................................... 33
3.2 Hypotheses development ........................................................................................ 33
Chapter 4 Methodology ................................................................................................. 45
4.1 Data Collection and Sample .................................................................................... 45
4.1.1 Survey method ................................................................................................. 45
4.1.2 Survey administration ...................................................................................... 45
4.1.3 Population, unit of analysis and sample selection ........................................... 47
4.1.4 Selection and non-response bias ...................................................................... 50

4.2 Measurement instrument ......................................................................................... 55
4.2.1 Construction ..................................................................................................... 55
4.2.2 Measurement items .......................................................................................... 59

4.3 Scale reliability and validity ................................................................................... 75
4.3.1 Single response bias ......................................................................................... 75
4.3.2 Construct validity ............................................................................................. 76
Chapter 5 Analyses and Results ................................................................................. 81
5.1 Analyses .................................................................................................................. 81
5.2 Results ..................................................................................................................... 82
5.2.1 Budget performance ......................................................................................... 86
5.2.2 Schedule performance ...................................................................................... 90
5.2.3 Technical performance ..................................................................................... 94
5.2.4 Overall performance ........................................................................................ 98

vii

Chapter 6 Discussion and conclusion ....................................................................... 103

6.1 Findings and theoretical contributions .................................................................. 103
6.2 Managerial implications ........................................................................................ 111
6.3 Directions for future research ............................................................................... 112
Appendix 1 Final round Q-sort ................................................................................... 116
Appendix 2 Survey Questionnaire ............................................................................. 119
BIBLIOGRAPHY ........................................................................................................... 136

viii

LIST OF TABLES

Table 4.1 Distribution of observations across industries by SIC code ................ 48
Table 4.2 Distribution Of Observations ................................................................. 49

Table 4.3 Chi-squared test of equal distribution between called individuals and
the contact list with respect to plant’s employment ............................................. 52

Table 4.4 Chi-squared test Of equal distribution between called individuals and
the contact list with respect tO SIC code. ............................................................ 53

Table 4.5 Chi-squared test of equal distribution between participants and non-

participants with respect tO SIC code. ................................................................. 53
Table 4.6 Chi-squared test Of equal distribution between participants and non-
participants with respect to plant size. ................................................................ 54
Table 4.7 CFA results for independent variables ................................................ 77
Table 4.8 CFA results for dependent variables ................................................... 78
Table 4.9 CFA results for control variables ......................................................... 80
Table 5.1 Descriptive statistics Of variables for regressions with simple avera9983
Table 5.2 Descriptive statistics of variables for regressions with weighted
average ............................................................................................................... 83
Table 5.3 Correlations Of variables for regressions with simple average ............ 84
Table 5.4 Correlations of variables for regressions with weighted average ........ 85

Table 5.5 Summary of hypotheses testing results for budget performance based
on a one-tail test at 0.05 level Of signiﬁcance ..................................................... 87

Table 5.6 Regression results for budget performance with simple average IID .88

Table 5.7 Regression results for budget performance with weighted average IID
............................................................................................................................ 89

Table 5.8 Summary of hypotheses testing results for schedule performance
based on a one-tail test at 0.05 level of signiﬁcance ........................................... 91

ix

Table 5.9 Regression results for schedule performance with simple average IID
............................................................................................................................ 92

Table 5.10 Regression results for schedule performance with weighted average
IID ....................................................................................................................... 93

Table 5.11 Summary of hypotheses testing results for technical performance
based on a one-tail test at 0.05 level of signiﬁcance ........................................... 95

Table 5.12 Regression results for technical performance with simple average IID
............................................................................................................................ 96

Table 5.13 Regression results for technical performance with weighted average
IID ....................................................................................................................... 97

Table 5.14 Summary of importance weight assigned to each category Of
performance ........................................................................................................ 98

Table 5.15 Summary Of hypotheses testing results for overall performance based
on a one-tail test at 0.05 level Of signiﬁcance ................................................... 100

Table 5.16 Regression results for overall performance with simple average IID

.......................................................................................................................... 101
Table 5.17 Regression results for overall performance with weighted average IID
.......................................................................................................................... 102
Table A1 Instructions for Q-sort ............................................................. 116
Table A2 Final Round Q-sort Correct Placement Ratio ............................... 117

LIST OF FIGURES

Figure 3.1 Theoretical Model .............................................................................. 33

xi

Chapter1 Introduction

This Chapter presents the motivations for conducting this dissertation. It
ends with deﬁnitions of terms used in this study, the research questions and

overview of the dissertation.

1.1 Motivations

The resource-based view (RBV) Of the ﬁrm has recognized that
competitive advantages stem from proprietary resources and capabilities (Barney
1991; Prahalad and Hamel 1995). Arguably however, no asset can yield
competitive advantage forever. Market changes, imitations, and substitutions
constantly reduce the value Of a company’s asset. Companies need to engage in
a continual effort to change and differentiate themselves. Recently, scholars
turned their attention to this ability to change existing resources and capabilities
and referred tO them collectively as dynamic capabilities (Teece et al. 1997;
Eisenhardt and Martin 2000; ZOllo and Winter 2002). One very important
dynamic capability is the ability to innovate (Eisenhardt and Martin 2000).

While innovation has always been an important factor in business
competition, management researchers have only given signiﬁcant attention to
innovation in the last few decades. Social science researchers in a number Of
ﬁelds, on the other hand, have studied innovations from various perspectives for
the last two centuries. Sociology and economics, for examples, have long
recognized the importance of innovations as agents Of social changes and

economic growth (Sundbo 1999). Their focus has been to understand how

innovations create Changes in the society and the economy, as well as how
socio-economic conditions inﬂuence the rate and quality Of innovations, e.g.,
(Sedgley and Elmslie 2001; Albino et al. 2006; Choudhury et al. 2006; Sinha
2006)

Different management disciplines have recognized the value of
innovations, and have focused on different related areas.
- Early strategic management and marketing research focus on how and
what type Of innovation can be used to competitively position the company
in the market. Innovation is viewed merely as another corporate factor of
production (Sundbo 1999).
- Organization behavior and Psychology attempt to identify interactions Of
personnel at the micro level and how factors such as culture affect
innovativeness within ﬁrms (Anderson et al. 2004).
- Accounting tries to understand effects Of different type Of accounting
systems (e.g., costing systems) on rate of innovation, efﬁciency and
proﬁtability (Drake et al. 1999; Drake et al. 2001 ).
While signiﬁcance of innovations to companies as well as a broader
social/economic system are clearly recognized in these ﬁelds, the focus of these
ﬁelds is primarily on the impact and beneﬁts Of innovation. The innovation
process, therefore, has mostly been taken as a black box.

Operations Management (OM) research, on the other hand, concerns
itself with the task Of improving efﬁciency and effectiveness Of the innovation

process. In order to improve the process, OM researchers, therefore, necessarily

need to investigate at the resource level. Here resources refer broadly to both
physical and non-physical assets that an organization employs for the innovation
process. The CM question can then be summarized as: what factors are
important in successfully managing the innovation process?

Early OM research on innovation, however, has focused on new product
development. This was because product innovations were observable to
everyone and perceived to be more advantageous than process innovation
(Damanpour and Gopalakrishnan 2001). Research in operational process
innovation, on the other hand, has been less proliﬁc. However, there is a growing
recognition among both practitioners and academics that operational (e.g.,
manufacturing) processes can be leveraged for competitive advantage (Skinner
1969; Hayes and Wheelwright 1984). Zahra and Das (1993) argue that “although
US companies have excelled in developing new products, successful Japanese
and German manufacturers have stressed process innovations in their bids for
world leadership.” This has generated greater attention to manufacturing process
innovation (MPI). In this study the term MPI will refer to the set of activities aimed
at both developing and putting new manufacturing process ideas into effect
(Schroeder et al. 1989; Bamberger 1991 ).

Reasons for heightened awareness of the importance of MPls include
increasing complexity of products requiring more sophisticated production
systems (Jayanthi and Sinha 1998); and maturity of many products shifting
competitive arena to process improvement (Abernathy and Utterback 1978;

Pisano 1997). At the industry level Abernathy and Utterback (1978) observes

more emphasis on product innovation in the ﬁrst ‘ﬂuid phase”, an emphasis on
process innovation in the ‘transitional phase’, and a balanced level of incremental
product and process improvements in the ﬁnal ‘speciﬁc phase’. As more products
mature in the last decade into ‘speciﬁc phase’, attention shifts back to process
innovaﬁon.

l n addition to easily seen beneﬁts of enhancing operational competences
(cost, quality, delivery, and ﬂexibility), other beneﬁts of MPls include strong
proprietary position through ﬁrst mover advantage (Pisano 1997) and market
reputation through an image of a company committed to excellence (Pisano
1997; Dutta et al. 1999).

There are two main categories of research on MPI. First is research that
focuses on Japanese-derived practices. The success of Japanese companies in
the last three decades, particularly in the automotive industry, has spawned
interests in their management approaches (e.g., Spear and Bowen (1999)). The
inﬂuence reﬂects in what we know today as Just-ln-Time (JlT), Total Quality
Management (TQM), Six-sigma, and continuous improvement programs, for
examples. Examples of MPIs that arise from these programs include projects to
reduce machine setup time and employing of production signaling technique
such as kanban. These MPls can be characterized as incremental improvements
that are internally originated.

A few researchers have investigated success factors of overall JlT and
TQM programs (e.g., Flynn et al. 1995; McLaChIin 2004; Salahedin 2005) , but

not at the individual project level. Thus the success factors investigated are at the

company or plant level and not at the project level. Moreover, while the success
of a major program is necessarily an accumulation of success at small individual
project levels, the magnitude of these programs make it difﬁcult to pinpoint which
factors are more inﬂuential at the individual project level.

The second stream Of research focuses on implementation of a particular
Class of innovations such as advanced manufacturing technology (AMT) (Small
and Yasin 1997; Burcher et al. 1999; McDermott and Stock 1999). While this set
of studies focus at the individual project, the focus on a single technology makes
generalizability of their results to other types of process innovations limited.

This study focuses implementations, a latter stage in MPls, where an
invention is incorporated into an existing operational system. This line of
research is important for two reasons. First, Tyre and Orlikowski (1993; 1994)
ﬁnd that operational process implementation activities are critical in determining
an organization’s ability to realize an innovation’s potential. The implementation
period is when modiﬁcations of the technology and the organization itself are
easiest and provide greatest value to the organization. The reasons include that
patterns of innovation use become rigid and difﬁcult to Change, expectations of
employees are adjusted with their experience resulting in little motivation to make
subsequent Changes, and team membership and enthusiasm erode over time
(Tyre and Orlikowski 1994).

Second, process implementations are difﬁcult and are main causes for
companies not being able to achieve intended beneﬁts of innovation adoptions

(Klein and Sorra 1996; Waterson et al. 1999; Edmondson et al. 2001). Waterson

et al. (1999) report that less than 50% of the companies in their sample thought
that the implementation of new manufacturing practice has more than moderately
met their objectives. A reason is that a successful implementation of process
innovations depends on widespread changes in organizational structure and
administrative systems (Ettlie and Reza 1992; Damanpour 1996). Changes
required in process innovations, i.e., implementing a manufacturing practice new
to the organization, include upgrading of individual skills, redesigning of individual
tasks, restructuring of control, rewards and information ﬂows (Shani et al. 1992)
and even workforce composition (Frohlich and Dixon 1999). In addition, Frohlich
and Dixon (1999) ﬁnd a considerable gap between managerial beliefs and
empirical ﬁndings about what is important in implementing advanced
manufacturing technology (AMT), suggesting that practitioners do not have a
clear idea of how to effectively implement MPls.

According to Kazanjian and Drazin (1986) one of the biggest challenges in
process innovation is “how to best implement these technologies in ongoing work
settings”. Design engineers can never fully anticipate the effects of small
changes in materials, surroundings, and procedures (Khurana 1999). Macher
and Mowery (2003) also note:

“a common problem associated with new process development is the

hand-Off of a new process technology from development to the production

facility. Manufacturability issues often appear only after a new process has
been transferred to the manufacturing facility from a development site
because of differences between the development and production facilities

in equipment, production volumes, worker skills, and numerous other
factors.”

Accordingly, this study will focus on the implementation of MPI.
Speciﬁcally, this dissertation focuses on the impact of the distance/separation
between MPI sources and the adopting organization on the performance of the
implementation process. In this study, this separation will be referred to as
invention-to-implementation distance (IID). Recently, outsourcing and offshoring
of production technology R&D1 (Baines et al. 1999; Kimzey and Kurokawa 2002;
Lewin and Peeters 2006) and adoption of observed best practices (Pilkington
1998; Davies and KOChhar 2000) are becoming more prevalent. While there are
beneﬁts associated with these practices (e.g., better design capability, focus on
core competency) (Baines et al. 1999; Vanhaverbeke et al. 2002), they also
separate innovation implementing units from invention sources. This separation
can potentially have negative consequences on how successful an
implementation of an MPI will be. This is due to the lack of consideration of the
recipient during the invention process (Sen and Rubenstein 1989) and resistance
of the adopting organization referred to as not-invented-here syndrome (Katz and
Allen 1988; Damanpour 1996; Edmondson 2003).

While the potential negative effect is intuitive, opposing ﬁndings also exist.
Menon and Pfeffer (2003) ﬁnd that organizations tend to value external
knowledge more highly than internal knowledge. This is due to fear of internal
competition and lack of access to scrutinize external knowledge. Thus,

employees can be less motivated or even resist an effort to implement MPls that

 

l A recent research by 8002 Allen Hamilton and INSEAD surveying 186 of the world’s biggest

corporations shows that 77% of R&D facilities of major companies that will go up in the next three years
will be located in either China or India (Engardio, P. (2006). R&D offshoring: Is it working?, Business

Week Online.).

are invented internally. This opposing evident warrants direct testing of the

relationship between IID and MPI implementation performance.

1.2 Definitions
It is useful to ﬁrst deﬁne the scope of this study by providing precise

deﬁnitions of the terms that will be used in this study. The terms implementation,
implementation team, implementation performance/success, and (IID) will be
deﬁned to provide a shared understanding. The rest of the terms will be deﬁned
as they are introduced.

The term innovation has been used to refer to many things. Bamberger
(1991) identiﬁed different processes that have been referred to as innovation.
These include invention, adoption/implementation, and diffusion. Invention is the
process where technical speciﬁcations (e.g., processing speed and quality) and
requirements (e.g., technology, particular skills and materials) are determined.
The focus in this study is the implementation process. Following Bamberger
(1991), implementation refers to the process that causes an invention to become
part of the system. This process includes “all events and actions pertaining to
modiﬁcation of both the [invention] and the organization, the initial utilization, and
the continued use of the innovation until it becomes a routine feature of the
organization” (Damanpour 1996). Thus, modiﬁcations of the practice/technology
after overall technical speciﬁcations and requirements have been determined are
part of the implementation effort and not invention. Deﬁning the terms this way

limits invention activities only as activities prior to the implementation stage.

In this study a new manufacturing practice/technology that is implemented
refers to a wide range of things. It can be as small as a procedural change, an
equipment Change, or as large as an automation of a process. A new
manufacturing practice/technology can also be an invention of the implementing
unit, corporate support functions, or completely external sources (e.g., adoption
of new procedures, and purchase of new equipment).

This study takes Leonard-Barton’s (1988) perspective that misalignments
between new technology and the organization necessarily occur in technology
implementation because the invention process can never anticipate all possible
problems, and because an aim of many innovation implementations is to
stimulate Changes. Innovation implementation is, thus, a mutual adaptation
process, where both the technology and organization Characteristics need to be
adjusted to correct for misalignments. Sen and Rubenstein (1989) point out some
speciﬁc events and actions in innovation implementation. These include planning
and carrying out of modiﬁcations to the adopted technology and existing
practices and infrastructures, garnering of skilled personnel, training of intended
users, and making available infrastructure and resources. The implementation
process ends when an innovation-adopting organization works with its own
manpower, skills and resources (Sen and Rubenstein 1989).

Implementation team in this study refers to everyone who is involved in
reinventing/modifying the new technology/practice, deciding which existing
routines (procedures regularly followed) and practices need to be Changed and

how to Change them (including conducting pilot tests), and providing support

(e.g., technical, infrastructure, ﬁnance and training) in institutionalizing new
routines. Modiﬁcation of technology/practice requires the team to understand the
mechanics and logic behind the new process (Edmondson 1999; Baer and Frese
2003; Edmondson 2003). In some cases the team may have to investigate
existing practices and relationships in their own organization to understand how
the new process will link with the rest of the system (Beatty 1992; Pisano 1997).
As problems and conﬂicts can arise at any point, another important task of this
team is solving problems (Edmondson 1999; Baer and Frese 2003; Edmondson
2003). In summary, relevant activities in this team include learning, making
Changes and solving problems as they arise.

Implementation performance or implementation success has been deﬁned
in the organization literature as “the incorporation or routine use of a technology
on an ongoing basis in an organization” (Edmondson et al. 2001). As an
operations management study, efﬁciency and effectiveness of the
implementation process are key concerns in this study. Thus, an implementation
is judged to be successful if it achieves technical/operational improvements
potential of the invention (e.g., reduced production cost, and enhanced
manufacturing quality, reliability, speed of delivery, safety, and environmental
friendliness), and the implementation process is executed quickly and
economically relative to an organization’s expectation. These three broad areas
of success have also been the staples of success measures in the project

management literature (Shenhar et al. 2002).

10

lnvention-to—implementation distance (IID) refers to the degree of
separation between invention sources and the adopting organizational unit. An
invention source is where the implementing unit acquires new
practice/technology. An invention could be an idea for procedural change, or a
new equipment. Studies of teams have associated physical distance (both in
meters/feet and perceptual measures) with separation that prevents
communication and collaboration (Hoegl and Proserpio 2004). This study
proposes two additional dimensions of separation: temporal distance and
organizational distance. Temporal distance refers to elapsed time since the
invention occurred. Organizational distance refers to separation created by
organizational boundaries.

At one extreme, an invention might be designed at the plant, at the time of
the implementation, and solely by members of the implementing unit. This results
in no physical, temporal or organizational distance. At the other extreme, an
invention might come from a source that is located far away (e.g., an oversea
OEM), had been designed for many years, and was purchased off-the-shelf. This
results in high physical, temporal and organizational distance.

An important condition of this deﬁnition is that an adopted innovation can
have multiple sources. The implication on measurement of the construct is that
questionnaire items will need to reﬂect the need to collect information for IID from

more than one invention source.

11

1.3 Research questions
This study builds on earlier technology implementation research from both

operations management and organization literature. Findings and theories from
project management, organizational learning and absorptive capacity will also be
used to explain the hypotheses.

Two speciﬁc questions in this study are:

- Does greater llD result in lower implementation performance in terms of
achievement of technical objectives, implementation budget and
implementation schedule?

- Do project team members’ technical competence, team functional diversity
and psychological safety, which contribute to learning potential of the
implementing team, moderate the effect of MD on implementation

performance?

1.4 Overview of the dissertation
The remainder of the dissertation is structured as follows. Chapter II

discusses previous ﬁndings on variables of interest to this study from technology
implementation, project management and absorptive capacity literature. Chapter
III discusses the theoretical framework and hypotheses. Chapter IV proposes the
methodology to be used in this study. Finally, Chapter V concludes the
dissertation with discussion of theoretical contributions, managerial implications

and potential future extensions.

12

Chapter 2 Literature Review

This Chapter reviews three relevant sets of literature: technology
implementation literature, project management and absorptive capacity and
organizational learning. Section 2.1 points out key Characteristics of the three
streams of literature and their relevance to this study, and Section 2.2

categorizes and discusses past ﬁndings.

2.1 Key Characteristics of the literature
First, there are two sets of literature speciﬁcally on innovation

implementation: those that originate in operations management and industrial
engineering and those that originate in organization theories. Repenning and
Sterman (2002) offers an insight on the distinction between studies from the two
ﬁelds: “Whereas theories originating in operations management and industrial
engineering largely ignore the beliefs and behaviors of those working with the
production technology, organizational theories generally do not account for the
physical structure of the organization and its processes.” Integration of the two
streams of research should foster greater understanding of innovation
implementation process. It is an intention of this study to also provide a step
forward in bringing considerations of the two ﬁelds together.

The second group of literature relevant to this study is Project
Management. The literature suggests that manufacturing process innovation
implementations are often managed as projects which are one-time events
managed by temporary teams (Ettlie and Reifeis 1987; Klein and Sorra 1996).

From its start in the 1950’s as mathematical research on computer application

13

and expert systems for project planning, control, and risk analysis, project
management’s focus has recently shifted to identifying factors, such as
characteristics of the team, that inﬂuence project success or failure (Jugdev
2004). This stream of literature suggests that project success is associated with
the existence of several critical implementation factors which are often contingent
on speciﬁc project characteristics (Pinto and Mantel 1990; Shenhar 2001).

Lastly, the literature on absorptive capacity and organizational learning is
reviewed. Process implementation requires an organization to learn new
processes and solve problems that arise. Learning in implementation
encompasses both technical knowledge (know how) as well as social knowledge
(who knows what) (Edmondson et al. 2001). In addition to learning, the
implementation team also needs to address discrepancies between existing and
required practices by ways of making adjustments to existing practices (e.g.,
manufacturing routines/procedures and production and service tools and
equipment) (Edmondson et al. 2001). Since organizational practices are
manifestations of knowledge, modiﬁcation to them is a form of knowledge
creation, the act of combining information and knowledge into new knowledge
(Smith et al. 2005).

Citing previous studies (Bradshaw et al. 1983; Simon 1985) Cohen and
Levinthal (1990) state that learning and knowledge creation are very similar in
their modes of development and necessary preconditions. They group the two
concepts under the umbrella of absorptive capacity. Absorptive capacity is

deﬁned as “a set of organizational routines and processes by which ﬁrms

14

acquire, assimilate, transform, and exploit knowledge to produce a dynamic
organizational capability” (Zahra and George 2002). This set of literature
examines and points out conditions that facilitate organizations’ ability to learn
and create new knowledge. These conditions include greater overall knowledge
of team members, diversity of the team, and strong networks among team

members.

2.2 Past findings on implementation success factors
Four categories of factors that inﬂuence implementation success emerge

from the review of the aforementioned literatures: 1.) project personnel, 2.)
implementation policies and organization, 3.) adopting organization’s
Characteristics and 4.) innovation characteristics. Table 2.1 summarizes variables
that have been studied by past studies. The following subsections summarize
ﬁndings and arguments on each group of characteristics. It is important to
recognize that variables in different categories are related and interact and may
not be as mutually exclusive as presented here.
2.2.1 Project personnel

Previous studies identify many roles and characteristics of project
personnel as success factors of implementations. Project personnel issues
include issues involving project leader, team members, and team characteristics.

Project leaders have roles within the project team as well as roles as an
interface to the rest of the organization. Roles within the project team include
getting the team to work by way of motivation, coordinate and assign tasks to

member, facilitate discussions, resolve conﬂicts and provide guiding visions

15

(Fedor et al. 2003). From the organizational learning perspective, project leaders
are also in the position to foster the learning orientation among team members
(Edmondson 1999; Edmondson 2003). These include being inspirational,
soliciting inputs and communicating the appreciation of members' skills. Roles
with the rest of the organization include garnering of resources and protecting the
project from outside inﬂuences (Co et al. 1998).

In order to carry out the roles and responsibilities mentioned above,
project leaders should possess certain personal characteristics and skills. As
beneﬁts do not come instantaneously, and there often are short-term drops in
performance (Repenning 2002; Repenning and Sterman 2002; Klein and Knight
2005), project leaders need to be patient and persistent (Beatty 1992). Kerzner
(1987) and Hewitt (1995) identified that managers who are results-oriented and
pragmatism are more successful in managing projects.

In order to garner resources project leaders need to be politically well
established and have a good understanding of the organization (Kerzner 1987;
Co et al. 1998). Co et al. (1998) ﬁnd that projects with politically well established
leaders in the company are more likely to be successful. Some important skills
include technical and analytical skills (Beatty 1992; Belassi and Tukel 1996), and
strong interpersonal skills such as communication, persuasion and motivation
(Kerzner 1987; Beatty 1992; Hewitt 1995; Belassi and Tukel 1996). Beatty
(1992), however, ﬁnds that technically skilled subordinates can make up for the

project leaders lack of technical skills.

16

Team members play an important role in bringing knowledge from external
sources and creating solutions to problems the team faces. Absorptive capacity
of an organization depends on its members’ individual absorptive capacity and
the composition of the organization. Individual absorptive capacity in turn
depends on their depth and breadth of knowledge, their network outside of the
team, and their ability to transfer knowledge to the rest of the team (Leonard-
Barton 1988; Cohen and Levinthal 1990; Smith et al. 2005).

Findings from psychology indicate that accumulation of prior knowledge of
an individual increases absorptive capacity. The depth of knowledge in one ﬁeld
increases ability of an individual to understand and utilize knowledge from a
source, while the breadth and diversity of knowledge (including ‘who knows what’
or ‘who can help’) increases the prospect that incoming information will relate to
what is already known (Cohen and Levinthal 1990).

Smith et al. (2005) ﬁnd that years of formal education is a measure of
knowledge stocks that is positively associated with the ability of an organization
to create new knowledge (knowledge creation capability, KCC). Years of
experience, however, was found to have a non-signiﬁcant negative effect on
KCC. This may suggest a tradeoff between expertise and unwillingness to

Change.

17

Table 2.1: Literature review summary

 

 

 

 

 

 

Theoretical Bases
Innovation lmplementatlon Absorptlve
Variables Project Capacity and
O eratlons
p Organlzatlon Management Knowledge
Management Management
Project personnel
Leaders
strong technical and
analytical skills (or _
supplemented by skilled 393W (1992) BeIaSSI and Tukel (1996)
subordinates)
Fedor at al (2003)

 

communication, motivation
and persuasron skills

Beatty (1992), Hewitt (1995);
Sann and McDermott (2003)

Repenning (2002);
Edmondson (2003)

 

politically well established

Co at al. (1998)

Repenning (2002);

 

Repenning and Sterman

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

patient for results Beatty (1992)
(2002)
practical/result-oriented Kerznerélsgm 983359 and
Team members
overall technical competence Hewitt (1995) COMES: :fzilntgalégfwx
°'°Ss'"a‘"°d Ettlie and Reifeis (1937) Cohen and Levinthal (1990)
personnel/generalists
”my ‘° °°mm“'"'°a‘° 8"" Shani et al. (1992) Leonard-Barton (1988 )
Teach
trouble shooting ability (ability
to handle unexpected crises Pinto and Mantel (1990)
and deviations from plans)
network Smith et al (2005)
Team characteristics
Ettlie and Reifeis (1987);
Beatty (1992); Hottenstein
and Dean Jr. (1992); Ettlie Ancona and Caldwell
functional diversity and Reza (1992); Small and (1992), Pinto et al. (1993); Cohen and Levinthal (1990)
Yasm (1997); Co at al. Cummings (2004)
(1998); Tyre and Hauptman
(1992)
. . Ancona and Caldwell (1992);
Tenure diversrty Cummings (2004)
Demographic diversity (age.
gender, tenure, Industry Cummings (2004)
experience)
Teamwork quality Hoegl anggoizrgiuenden
. . Pinto et al (1993), Hoegl
Physrcal p “)an and Proserpio (2004)
. Edmondson (1999),
psychoggifgugﬁy lculture Edmondson (2003); Beer Smith et al. (2005)
g and Freese (2003)
Implementation
policies and
organization

 

Goals and task

 

requirements

challenging but realistic (e.g.,

Beatty (1992), Linderman et

 

 

 

 

 

 

 

 

supplemented by tools and al (In-press)
SUDDOFI)
clear and well communicated Beatty (1992) Pinto and Mantel (1990)
accepted throughout Beatty (1992); Hewrtt (1995) Pinto et al (1993)

 

18

Table 2.1 (continued)

 

Theoretical Bases

 

 

 

Innovation Implementation . Ab5°fPtIV°
Variables ‘ Project Capacity and
Operations Organization Management Knowledge
Management Management
Suppon
Communication and Hewitt (1995 ): Meyers et al.

 

Information infrastructures

(1999); Frohlich and Dixon
(1999)

 

Training and preparation
(quality and quantity)

Co at al. (1998 ): Small and
Yasin (1997); Meyers et al.
(1999)

Klein and Serra (1996); Klein
et al. (2001)

Fedor et al. (2004)

 

Organizational adaptation

Leonard-Barton (1988 );
Frohlich and Dixon (1999);
Tyre and Hauptman (1992)

 

Resource Slack I availability

Meyers et al. (1999).
Hottenstein and Dean Jr.
(1992)

Klein et al. (2001)

Fedor et at. (2004)

 

Incentives for change

Klein and Sorra (1996); Klein

 

 

 

 

 

Stock and Tatikonda (2004)

at al. (2001)
TOP management Nippon Km": d°;é§g:1£0::‘;m' Pinto and Mantel (1990)
Coordination
external to the plant Tyre and Hauptman (1992)?

 

 

 

Organization
Characteristics

 

culture (Quinn and
Roughbaugh , 1981i)

Zammuto and 0' Connor
(1992 ); Mc Bennett and
Stock (1999)

 

innovation-value fit

Klein and Sorra (1996);
Detert and Schroeder
(2000); Holahan et al. (2004)

 

Klein and Sorra (1996); Klein

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

implementation climate et al. (2001 ); Holahan et al.
(2004)

project manager-line Turner (2004)

manager relationship

past experience with Zahra and George (2002):

innovation Smith et al. (2005)

cross-trained employees Shani et al. (1992) Cohen and Levinthal (1990)

communication skills Shani et al. (1992)

Innovation
Characteristics
incremental/radical Hottenstein and Dean Jr.
Improvement (1992)
Tyre and Hauptman (1992):
Hottenstein and Dean
prevent new a untried (1992) Stock and Tatikonda Edmondson (2003)
(2004)
extent mulred change Tyre and Hauptman (1999 Edmondson (2003)

source of knowtedge Menon and Pfeffer (2003 );

(Intemallexteman Sen and Rubenstein (1989) Edmondson (2003) Fedor et al. (2003)

Institutional distance from Jensen and Szulanskl (2004)

 

19

Past empirical studies on MPI implementation also suggest that these
characteristics have positive association with higher implementation
performance. These include level of technical competence (Hewitt 1995), extent
of cross-training and cross-functioning (Ettlie and Reifeis 1987; Shani et al. 1992)
and communication and teaching skills (Leonard-Barton 1988; Shani et al. 1992).
Pinto and Mantel (1990) also ﬁnd ability to handle unexpected crises and
deviations from plan as an important desirable characteristic of project team
members.

Beside individual characteristics, the characteristics of the team are also
important. These include diversity, teamwork quality, physical proximity of team
members, and psychological safety. Diversity is the most frequently studied team
characteristics. While diversity has been generally recommended, empirical
ﬁndings have been mixed. Diversity is advocated in innovation teams as it brings
in different perspectives and because members have access to a more diverse
set of knowledge (Cohen and Levinthal 1990). Several types of diversity have
been suggested. These include functional diversity, tenure diversity, and
demographic diversity.

The absorptive capacity literature has a good explanation of the effect of
diversity on absorptive capacity. An organization’s absorptive capacity depends
not only on the absorptive capacities of its individual members as mentioned
above. It also depends on linkages and networks the individuals have with
external environment to be able to absorb information and knowledge (Smith et

al. 2005), as well as the ability to internally share knowledge and information and

20

transfer them to where they can be utilized . Thus, while diversity of background
is useful for receiving knowledge from outside, it also makes internal
communication difﬁcult. Cohen and Levinthal (1990) suggest that there is “a
trade-off in the efﬁciency of internal communication against the ability of the
subunit to assimilate and exploit information originating from other subunits of the
environment”.

Smith et al. (Smith et al. 2005) ﬁnd that functional diversity of top
management and knowledge workers, a measure of knowledge stocks, in
addition to years of formal education, has a positive association with the ability of
an organization to create new knowledge. Evidence in support of beneﬁts of
functional diversity on innovation implementation and project management
comes both from case studies (Ettlie and Reifeis 1987; Beatty 1992; Hottenstein
and Dean Jr. 1992) and from large-scale empirical tests (Ettlie and Reza 1992;
Tyre and Hauptman 1992; Pinto et al. 1993; Small and Yasin 1997; Co et al.
1998)

Ettlie and Reza (1992), for example, ﬁnd in correlation analyses that
integration of functions in an organization has signiﬁcant correlation with
improved performance due to an implementation of a purchased process
technology. Small and Yasin (1997) ﬁnd a positive impact of team-based project
management practice (a construct which includes use of multi-disciplinary teams)
on operational improvement for some AMT types. Co at al. (1998) also ﬁnd a
positive relationship with the use of multi-functional team and AMT

implementation performance. From the project management literature, Pinto et

21

al. (1993) ﬁnd cross-functional cooperation to have positive relationship with
perceived project's operational task outcomes. On a related variable to functional
diversity, Tyre and Hauptman (1992) found that functional overlap has a negative
effect on startup speed but a positive effect on operational improvement.

However, Ancona and Caldwell (1992) ﬁnd a direct negative effect of
functional diversity on technical performance and overall performance of product
innovation projects. The authors also ﬁnd an indirect positive effect through
greater degree of external communication, which has a signiﬁcant positive effect
on technical performance. However, the positive indirect effect is not large
enough to overcome the direct negative effect. There is also no signiﬁcant direct
effect on budget or schedule achievement.

Cummings (2004) also ﬁnd no direct effect of functional diversity on work
group performance. He found only a positive moderating effect of functional
diversity on the external knowledge sharing-performance relationship and no
signiﬁcant moderating effect on internal knowledge sharing-performance
relationship. Ancona and Caldwell (1992) and Cummings’s (2004) ﬁndings
support Cohen and Levinthal’s (1990) and Smith et al.’s (2005) contention that
functional diversity only is valuable in enhancing the communication to external
sources.

Tenure diversity and demographic diversity have been less well studied
and supported. Ancona and Caldwell (1992) ﬁnd that tenure diversity (measured
as coefﬁcient of variation of years employed by the organization) has a positive

relationship with group internal process (behaviors aimed at organizing members

22

to get work done, e.g., goals deﬁning, work plan development and work
prioritization) but no effect on external communication. However, the direct effect
of tenure diversity is non-signiﬁcant on innovation quality and team’s satisfaction
with the outcome of the project, and negative on budget and schedule
achievement. Tenure diversity is also part of the measure of demographic
diversity of teams in Cummings (2004), which ﬁnds no signiﬁcant relationship of

demographic diversity and project performance.

Hoegl and Gemuenden (2001) investigate the relationship between
teamwork quality, the quality of interaction among team members, and
performance of 145 software development projects. Their study speciﬁes
teamwork quality as a higher order construct encompassing communication,
coordination, balance of member contributions, mutual support, effort and
cohesion. Structural equation models show that teamwork quality is positively
associated with project performance (effectiveness and efﬁciency) as rated by
team members, team leaders and managers and personal success of team

members measured by work satisfaction and learning.

Pinto et al. (1993) ﬁnd that physical proximity among team members
positively affects cross-functional cooperation. Hoegl and Proserpio (2004) also
ﬁnd that team member proximity, measured perceptually, is positively related to
ﬁve of the six dimensions (except balance of member contribution) of teamwork

quality.

23

Edmondson (1999) deﬁnes psychological safety as a shared belief that
the team members are safe for personal risk taking. Psychological safety results
in increase conﬁdence that members have in taking initiatives due to reduced
concerns about others’ reactions that pose threats (embarrassment or punitive
actions) to many learning behaviors. Edmondson (1999) ﬁnd that psychological
safety positively affects learning behavior which is associated with team
performance and Baer and Freese (2003) ﬁnd positive moderating effect of
psychological safety on the effect of organizational process innovativeness
(measured as change on 12 dimensions) on company goals and ﬁnancial
performance. Sarin and McDermott (Sarin and McDermott 2003) also ﬁnd a
positive relationship between participation-oriented style management (where
team leader invites member’s active involvement in the decision making process)

and learning.

2.2.2 Implementation policies and support
Implementation policies and support include quality and quantity of

training, provision of technical assistance to users, availability of rewards,
communication regarding reasons for change, provision Of time to experiment,
and user-friendliness Of the technology. Klein et al. (2001) ﬁnd support for direct
positive effect of the composite measure of the six implementation policies and
practices on performance. Klein and Knight (2005) note that these policies and
practices seem to be cumulative and compensatory. The following discusses

ﬁnding on more speciﬁc implementation policies and activities.

24

Goals and task requirements

Through analyses of cases, both Beatty (1992) and Hewitt (1995) suggest
that project goals need to be realistic, clear, well communicated, and accepted
throughout the organization. One particularly effective way to communicate goals
is by providing incentives and rewards for speciﬁc behaviors and/or overall
project success (Klein and Sorra 1996; Klein et al. 2001). However, Pinto and
Mantel ( 1990) ﬁnd Clear project mission not signiﬁcantly related to project
success, but detailed speciﬁcation for individual action steps related to project
success. Pinto et al. (1993) ﬁnd that the existence of goals that drive the need for
changes unify functions within organizations to collaborate.

In addition, challenging goals can help motivate organizational members,
but only when the members are equipped with resources to carry out actions.
Linderman et al. (2006) drawing on goal-setting literature ﬁnd that Challenging
goals are positively associated with six-sigma team performance only when

project team members adhere to six sigma tools and mechanisms.

Support (Top management and resource, communication and information
infrastructure, Modiﬁcation to MPI and existing organization)

Top management support is highly important to MPI implementation
success because it is a precondition for resource support, but its beneﬁt may be
contingent on other variables. Top management support provides legitimacy to

everyone in the organization as well as reinforces employees’ collective belief

25

that the use of the innovation will be rewarded. It leads to commitment from line
managers to get the job done by providing resources such as time, equipment
and technical skills (Kerzner 1987). Moreover, since many projects require
collaborations between line managers and project managers, executive
commitment to the project is valuable in resolving conﬂicts between line and
project managers (Kerzner 1987; Belassi and Tukel 1996).

Pinto and Mantel (1990), however, ﬁnd no signiﬁcant relationship between
top management commitment and project performance, and Klein et al. (2001)
ﬁnd that it is availability of ﬁnancial resources, not top management support, that
is signiﬁcantly related to quality of policies and practices. Sharma and Yetton
(2004) show in a meta-analysis that the effect of management support on
information systems implementation success is contingent on task
interdependence (the degree to which implementation tasks need to be
performed together). Only at high level of interdependence between functions

does top management support have signiﬁcant effect on success.

Given that an MPI implementation often requires collaboration between
functions and departments, the interface between these functions and
departments need to be integrated. Collaborative work between different
functions requires compatibility in modes of communication. These include, for
example, compatible CAD system for design and tooling, and common reporting

and command center (Hewitt 1995; Meyers et al. 1999).

26

Beatty (1992) notes that one of the biggest obstacles to implementing
AMT is the integration of the new production process with design and adjacent
manufacturing processes. There needs to be a plan of how the new process will
connect with existing activities. In some cases, the cost of creating the software
that bridges among manufacturing systems and design that are not compatible
offset the saving in acquisition costs. Frohlich and Dixon (1999) study the effects
of adaptations of human resources, information systems, equipment, materials
and operational structure on improvement in operational performance after AMT
implementation. The only adaptation that was signiﬁcantly associated with
improved performance is information system adaptation.

The lack of ﬁt between a MPI and the organization necessitates
modiﬁcation to the MPI and the organization. Tyre and Hauptman (1992) ﬁnd that
projects where more modiﬁcations to the MPI occur take less time to implement
and better achieve the technical goals of the project. On the other hand, one of
the most signiﬁcant modiﬁcations to the existing organization is training users of
the process. Not only does training provide operators with proper skills, it has
implications on positive motivation, attitudes and commitment/ownership
(Hottenstein and Dean Jr. 1992; Meyers et al. 1999). Co at al. (1998) ﬁnd that
AMT projects in which education and training are stressed by top management
and are provided to all level of team members are associated with higher
success. Small and Yasin (1997) ﬁnd that workers’ preparation is positively

associated with better operational performance. Meyers et al. (1999) also identify

27

training of staff on the new technology as critical to industrial process

implementation.

Lastly, some studies ﬁnd the power relationship between the people
involved in the project to be a signiﬁcant factor of success. Kerzner (1987) ﬁnd
that a clear reporting structure between workforce to the project/line manager
and manager to top executives as well as the parity of power between project
managers and line managers improve project performance. Turner (2004) ﬁnds
collaboration between project manager and project owner (e.g., functional areas)

to lead to greater project performance.

2.2.3 Organizational characteristics

The third set of antecedents for MPI project success involves
characteristics of the organization. Studies have investigated implementation
climate, culture, the ﬁt between an innovation and the organization, and resource
availability. Not surprisingly, many of the policies and practices mentioned in
previous sections are aimed at creating desirable organizational characteristics

and/or eliminating undesirable ones.

Recent studies attempt to capture necessary organizational characteristics
in one single construct, implementation climate. Implementation Climate is
deﬁned as targeted users’ shared summary perceptions of the extent to which
their use of a speciﬁc innovation is rewarded, supported and expected within

their organization (Klein and Sorra 1996). Klein and Sorra (1996) argue that,

28

“A strong implementation climate fosters innovation use by (a) ensuring
employee skill in innovation use, (b) providing incentives for innovation
use and disincentives for innovation avoidance, and (C) removing
obstacles to innovation use.”
Klein and Sorra (1996) offer an example from their case study of an
implementation failure where targeted users have achieved satisfactory level of
skill to Operate new technology after training and are excited with the prospect of
improvement. The failure is attributed to limited on-the-job assistance and the
lack of incentives to change.

This contention is supported by Klein et al. (2001). Holahan et al. (2004)
also ﬁnd support in a study on implementation of a computer technology in
science education. In addition, they ﬁnd that implementation Climate mediate the
positive effect of organizational receptivity toward change on implementation
effectiveness. Baer and Freese (2003) ﬁnd a positive effect of climate for

initiative on implementation goal achievement, and that the effect is greater with

the extent of Change.

Another collective aspect of an organization is its culture. Quinn and
Rohrbaugh’s (1981) competing values model suggests that organizations can be
distinguished along two axes reﬂecting different value orientations: ﬂexibility-
Control dimension, and internal-external focus. Scholars have suggested that
these value orientations have implications on the implementations of new
practice and technology. Klein and Knight (2005) suggested that “inclination
toward stability” could hinder implementations. In addition, a perspective from

project management emphasizes that corporate culture should not view project

29

management as a threat to established authority or a cause for unwanted change
(Kerzner 1987). However, McDermott and Stock (1999) ﬁnd no association
between any cultural orientation and operational improvements ensuing AMT
implementation.

Another important organizational Characteristic is the ﬁt between an
innovation and the organization. Klein and Sorra (1996) suggest that commitment
to the use of an innovation depends on the perceived ﬁt of the innovation to
employee’s values, the generalized enduring beliefs about desirability of actions
and their consequences. According to this, employees are more willing to adopt
an innovation if the innovation fosters/builds on existing values. A company that
values production performance more than learning is less likely to sacriﬁce
production time to learning. Similar arguments are made in TQM implementation
culture (Detert and Schroeder 2000). Holahan et al. (2004), however, do not ﬁnd
a signiﬁcant effect on innovation-values ﬁt on implementation effectiveness. The
authors, however, note that this could be due to low variance in the value
measure.

Organizational resource availability is also as an important precondition for
success. The most important resource is personnel. Shani et al. (1992) found in
a case study that the lack of cross-trained employees and the lack of
communication skills (including language skills) necessary for group-based work
were barriers to a successful implementation of AMT.

Other Characteristics such as organizational size, geographical region,

type of institution (university vs. clinic, suggesting past innovation experience), do

30

not distinguish successful implementing efforts and unsuccessful ones

(Edmondson et al. 2001).

2.2.4 Invention Characteristics

The literature suggests several characteristics of an innovation that affects
implementation performance. These Characteristics include: newness of
technology, amount of change required, tacitness of knowledge embedded in the
new technology and source of invention.

Several studies indicate that implementations of a newer technology are
less successful than implementations of an older/mature technology (Hottenstein
and Dean Jr. 1992; Tyre and Hauptman 1992; Edmondson et al. 2003; Stock and
Tatikonda 2004). The main reason is that technical problems that have not yet
been resolved result in unreliability of technology (Klein and Knight 2005).

Amount of change required by the technology is also shown to worsen
implementation performance (Tyre and Hauptman 1992; Edmondson et al.
2003). This is especially true when “an innovation Challenges existing patterns of
interdependence among individuals or groups” and when existing routines are
well deﬁned and supported (Edmondson 2003). Required changes in existing
hierarchy and how works are conducted tend to hinder the implementation
process (Klein and Knight 2005).

Tacitness of knowledge embedded in the technology is generally
recognized to create difﬁculty in the learning process (Nonaka 1991). Edmonson

et al. (2003) found that when an innovation requires tacit knowledge, it takes

31

longer for the team to learn to excel at the task and team stability is greatly
required to smooth out the process.

Another important characteristic of innovation is the source of an invention
(Argote et al. 2003). Previous studies view this variable as a dichotomy of
internal versus external invention source. Some authors believe that external
innovation is generally more difﬁcult to implement (Sen and Rubenstein 1989;
Edmondson et al. 2003; Fedor et al. 2003). Other authors (Menon and Pfeffer
2003; Menon et al. 2006) provide arguments for why a company may tend to
perceive external innovation to be of greater value, and, thus, can be more

motivated to adopt and implement the invention.

32

Chapter 3 Theory development

3.1 Model overview

This study focuses on the effect of “D on manufacturing process
innovation (MPI) implementation performance and the moderating effects of
three variables: functional diversity of the team, team members’ technical
competence, and psychological safety. All three variables pertain to
characteristics of the implementation team. Figure 1 shows the research model

for this study.

 

- Team members’ technical
competence

- Team functional diversity

- Psychological Safety

(. \

 

 

 

 
  
 

     

Invention to
implementation distance
IID

Implementation
performance

         
   

Figure 3.1 Theoretical Model

3.2 Hypotheses development

Direct effect of MD on implementation performance
There is currently no conclusive evidence for either positive or negative
inﬂuence of IID. Studies that exist can only provide anecdotal evidence for its

effect. Explicit large-scale empirical investigation on the relationship is lacking.

33

Studies of Menon and Pfeffer (2003) and Menon et al. (2006) provide evidence
for a positive effect. Menon and coauthors argue that fear of internal rivalry will
lead organization members to value internal knowledge lower than external
knowledge. However, in support of a negative effect, Fedor et al.’s (2003) results
also show that while internal knowledge generation has positive effect on product
and process innovation project team’s evaluation of success, external knowledge
generation has no signiﬁcant direct effect and a negative moderating effect on
the relationship between organizational support and performance.

Two reasons suggest that, ceteris paribas, IID should result in lower MPI
implementation success. First, externally acquired technology may be physically
and culturally unsuitable to the receiving organization. A potential cause for these
mismatches is that systems are Often developed with little or no consideration of
existing conditions at the innovation-adopting organization (Bostrom and Heunen
1977). Physically, this results in a mismatch between existing technological
equipment (e.g., tools, machine and raw materials), organization and structure
(e.g., assembly processes and shop ﬂoor layout), and human resource (e.g., skill
level) and those required by the new technology (Sen and Rubenstein 1989).
The more that invention is done “externally”, the less likely the consideration of ﬁt
between the invention specs and the organizational characteristic will have been
taken into account in the design process. As this logic readily applies to physical
and organizational distance, it applies to temporal distance as well. Even an
invention that is designed speciﬁcally for the implementing organization can

become unﬁt to the organization if time passes before it is implemented.

34

Changes in the working environment such as in the level of workers’ skills, a loss
of key personnel, or deterioration of equipment over time can render an invention
unﬁt for an organization.

This logic can be illustrated in adoptions of best practices and AMT.
Adoption of a best practice is an implementation of a set of routines that may
have been invented without any consideration of adopting ﬁrm’s readiness of
physical assets and equipment, culture, complimentary procedures, and staff of
the ﬁrm. Similarly, many AMT systems (e.g., robotics and computer—aided
design) are purchased off-the-shelf from a technology supplier that does not
consider how the new machine will interface with the rest of the system,
potentially causing extensive process integration requirements (Beatty 1992).
These could include modiﬁcations of bill of material database structure to create
consistency to facilitate data transfer.

Similarly to physical mismatches, cultural mismatches also reduce an
organization’s success in managing the implementation process. Klein and Sorra
(1996) and Detert and Schroeder (2000) posit that commitment to the use of an
innovation depends on the perceived ﬁt of the innovation to employee’s values,
the generalized enduring beliefs about desirability of actions and their
consequences. According to this, employees are more willing to adopt a MPI if
the MPI fosters/builds on existing values. Since cultural beliefs and values are
embedded in MPls, an MPI that is invented closer to the implementation site in
terms of physical, temporal and organizational distance is more likely to

capitalize on existing values. This should be an opposing force to the positive

35

effect due to less fear of internal competition (Menon and Pfeffer 2003). Jensen
and Szulanski (2004) ﬁnd some support that the source-recipient difference in
culture/value system reduces legitimacy of a practice in the eye of the recipient,
thus reducing recipient’s motivation in implementing the practice.

For example, a plant manager whose performance is assessed by plant
utilization may oppose practices such as preventive maintenance and kaizen as
these activities appear to detract away production time and their beneﬁts do not
arise until later. Another example is that an organization that highly values its
people will be more resistant to process innovations that require laying off
workers. As IID increases, the degree of ﬁt between an invention and employees’
values are likely to decline.

The lack of physical and cultural ﬁt as a result of greater IID may increase
amount of Change required. This often prompts resistance as they Challenge
existing structure of political inﬂuence, are more intimidating to organizational
members, and involve greater uncertainty Of requirements (Damanpour 1996).
Edmondson (2003) note that technology implementation is “especially difﬁcult
when an innovation Challenges existing patterns of interdependence among
individuals or groups”. Tyre and Hauptman (1992) also ﬁnd negative effect of
amount of required change on speed of implementation (measured in absolute
term).

The second primary reason for llD’s negative effects on implementation

performance is that IID limits direct communications between an implementing

unit and the innovation source. In the context of innovation implementations,

36

these include: communication to seek Clariﬁcations or advice and explanations
on the logic behind a practice, soliciting assistance in training, or merely
additional direct observations.

Studies in project team setting indicates that physical proximity among
team members inﬂuence type and frequency of interaction, exchanges and
communication (Pinto et al. 1993). Hoegl and Proserpio (2004) ﬁnd team
members’ proximity to each other leads to better communication, coordination,
mutual support, effort and cohesion among team members. Hoegl and
Gemuenden (2001) showed that these teamwork quality variables are positively
associated with team performance.

While these ﬁndings pertain to physical distance among team members,
the logic also applies in the relationship between an invention source and
recipient. Temporal and organizational distances also likely have similar effects
in limiting the amount and effectiveness of communication, interaction and
observation conducted between innovation implementing unit and the innovation
source. An example of communication limitations due to temporal distance is
leaving or retiring of key designers of the process within the organization over
time.

The following hypothesis is advanced based on the above arguments:

H1: IID is negatively associated with MPI implementation
performance.

Direct and moderating effects of other variables

37

This study focuses on three important variables from a learning
perspective: 1.) implementation team members’ technical competence (Ettlie and
Reifeis 1987; Hewitt 1995; Cohen et al. 1996; Smith et al. 2005), 2.) functional
diversity of team members (Ettlie and Reifeis 1987; Sen and Rubenstein 1989;
Small and Yasin 1997; Co et al. 1998) and 3.) team psychological safety
(Edmondson 1999; Edmondson et al. 2001; Baer and Frese 2003; Edmondson
2003). The essential argument of this portion of the study is that these three
variables as manifestations of learning potential, positively moderate the effect of
IID on implementation performance.

Since the arguments on the interaction effects focus on learning, it is
appropriate to note how greater IID may require greater amount of learning. First,
building an earlier argument, greater IID often requires systemic organizational
and cultural shifts as well as adjustment of the invention. More required
adaptation to the invention or to the organization requires greater amount of
learning and problem solving for the implementation team.

Second, the ND increases the difﬁculty and the costs of , and in some
cases prevent, direct communications with the source of the invention. These
include further observation, discussion of the logics behind the practice with
personnel from the inventing unit, and seeking assistance (e.g., training) directly
from the source. When it is required, the implementing unit will need to be more
involved in independent (of the source) information seeking, acquiring insights

into the practice and ﬁguring out requirements. Thus, in order to implement

38

externally invented technology successfully, a greater amount of learning will be
required in general.
The following is the discussion of the direct and moderating effects of the

three variables that contribute to an organization’s potential to learn.

Implementation team members’ technical competence

Team members’ technical competence is deﬁned as the level of technical
expertise represented by the project team. This team aspect is emphasized
because the innovation implementing process often requires extensive learning
and problem solving. Since solution to a problem is a form of knowledge,
problem solving is a knowledge creation activity. The ability of a team to learn
and create knowledge (i.e., problem solving in this case) depends largely on its
individual members’ absorptive capacity, because individual members are agents
that acquire, assimilate, transform and exploit knowledge.

Past ﬁndings on the effect of technical competence on performance have
been mixed. Smith et al. (2005) ﬁnd a positive association between years of
formal education and an organization’s ability to create knowledge. Pinto and
Mantel (1990) ﬁnd that team member’s problem solving ability is signiﬁcantly
associated with budget, schedule and technical performance, however, technical
expertise does not have a signiﬁcant effect on project’s success. Subramaniam
and Youndt (2005) also ﬁnd that the presence Of technical competent individuals
in the organization has a positive effect on the organization’s ability to innovate
when there is strong social interaction among the individual organizational

members.

39

Prior knowledge of a person, reﬂected by the depth and breadth of his
knowledge, is said to limit the individual’s ability to understand and create new
knowledge. This is because learning is a process of establishing linkages
between new information and existing memory and knowledge generation is a
process of drawing on existing knowledge and applying it in a different context
(Cohen and Levinthal 1990). Thus, the more prior technical knowledge a team
member possesses, the more linkages he can make, and the easier it will be for
him to learn. Thus, it can be argued that team members’ technical competence,
their prior technical knowledge, is important to innovation implementation
success (Leonard-Barton 1988; Cohen and Levinthal 1990; Smith et al. 2005).

Implementations of innovation often require speciﬁc skills. For example,
AMT implementations often need process integration that requires programming
skills. New procedure implementations require analytical ability of the team in
order to understand logics behind a new process and knowledge of the
environmental context at the adopting organization to properly apply new
processes.

H2 a: Implementation team members’ technical competence is
positively associated with MPI implementation performance.

As argued above, team members’ technical competence contributes to
team’s absorptive capacity. Through the ability to learn the team will better
resolve ﬁt issues resulting from greater IID. Thus, IID should less negatively
affect implementation performance when the team is composed of technically

competent members.

40

H2 b: Implementation team members’ technical competence
positively moderates the effect of IID on MPI implementation

performance

Functional diversity of implementation team

Many functions will be affected by organizational adaptations required in
adopting new manufacturing innovations (Ettlie and Reifeis 1987; Hottenstein
and Dean Jr. 1992). These functions typically include design engineering,
production control, maintenance, production/manufacturing, quality control,
purchasing, and management information system. Functional diversity is deﬁned
as the extent to which multiple functions are represented on the implementation
team

Including representative members of these functions in implementation
team has several beneﬁts. These include: breadth of knowledge of the
implementation team (more access to more variety of knowledge, thus greater
absorptive capacity), wider range of perspectives, greater opportunity of
detecting potential problems associated with each function and addressing them
beforehand, and increasing employees’ buy-in and support as their concerns are
heard (Ettlie and Reifeis 1987; Sen and Rubenstein 1989; Davenport 1993;
Small and Yasin 1997).

The use of a cross-functional implementation team has been found in
some studies to be a success factor in innovation implementation and project
management (Ettlie and Reifeis 1987; Sen and Rubenstein 1989; Tyre and

Hauptman 1992; Davenport 1993; Pinto et al. 1993; Small and Yasin 1997). For

41

example, involvement of R&D personnel assist in providing solutions to
modiﬁcation Of equipment and providing training to the implementing unit’s
employees as R&D personnel tend to be the most current and knowledgeable
about new technologies (Sen and Rubenstein 1989; Sen and Rubenstein 1990).
Inclusion of manufacturing personnel on the team provides information on
workforce skill level and current practices, informing the team what needs to be
Changed.
Based on the above arguments, the following hypothesis is advanced:

H3 a: Functional diversity of the implementation team is positively

associated with MPI implementation performance.

As argued above, functionally diverse teams exhibit greater team’s
absorptive capacity due to its breadth of knowledge embodied in the personnel.
Breath of knowledge affords the team access to more sources of knowledge and
a greater variety of perspectives that results in better solutions to problems
(Cohen and Levinthal 1990; Zahra and George 2002). Since functionally diverse
implementation teams are better at learning and solving problems (a form of
knowledge creation), they are better able to address learning and problem
solving requirement resulting from mismatches due to IID and lack of direct
communication. This argument leads to the following hypothesis.

H3 b: Functional diversity of the implementation team positively

moderates the effect of IID on MPI implementation performance.

Team’s psychological safety

42

Psychological safety is a shared belief that team members are safe for
personal risk taking (Edmondson 1999). Psychological safety is reﬂected in “a
sense of conﬁdence that the team will not embarrass, reject, or punish someone
for speaking up” (Edmondson 1999). It has been found that psychological safety
is positively associated with learning behaviors (Edmondson 1999), which are
important to successfully implementing a new MPI.

Learning can also be thought of as a process of detecting and correcting
errors provides another perspective on (Argyris and Schon 1978; Edmondson
1999). Thus, the ability of a group to learn depends on team members’ ability to
detect and correct errors in their understanding and beliefs. Behaviors that have
been shown to be effective in learning include feedback seeking from colleagues,
experimentation with new ways of doing things, voicing opposing opinions and
discussion of errors (Edmondson 1999). Since these behaviors are potential
sources of embarrassment and threat to career advancement, team members in
low psychological safety environment tend to conform to the norms and suppress
unusual information or observations.

Edmondson (2003) studies implementations of new cardiac surgery
technology which requires radical changes in operational procedures. She ﬁnds
that in surgical teams where the implementation of this new technology is framed
as a learning opportunity, members freely voice their Opinions and suggestions
without fear of any social/career punishment. This so called psychological safety
encourages learning behaviors which result in more successful implementation of

the technology (apply it to more variety of cases, reducing operation time and

43

continued usage). This line of argument has also been supported by ﬁndings of
Baer and Freese (2003) on organizational beneﬁts from adopting new
manufacturing innovations and Sarin and McDermott (2003).

H4 a: Psychological safety is positively associated with MPI

implementation performance

As argued above, psychological safety results in exercising of effective

learning behaviors. Thus, implementation teams with high psychological safety
are better able to resolve the mismatches issues due to IID. This argument leads
to the following hypothesis.

H4 b: Psychological safety positively moderates the effect of IID on

MPI implementation performance

44

Chapter 4 Methodology

This chapter covers methodological issues in this study. These include

data collection and sample, measured variables and scale reliability and validity.

4.1 Data Collection and Sample
4.1.1 Survey method

The intent of this study is to test a set of generalizable hypotheses in
manufacturing process innovation (MPI) implementation projects. Thus, the data
to be used should represent companies from different sizes, geographical
locations and industries. Three reasons make a survey an appropriate mode of
data collection for this study.

First, it allows gathering of a large amount of data economically in a
relatively short period of time. Second, a survey can cover a geographically
dispersed group of population. The population of this study is MPI
implementation projects in manufacturing plants in North America. The vast
region makes traveling to sites cost prohibitive. Lastly, survey provides privacy in
responding and anonymity to respondents. It requires no face-to-face encounter
making respondents more comfortable of sharing information without disclosing
their full identities, as well as insulating respondents from expectation of an

interviewer which may affect ﬁndings of the study (Mangione 1998).

4.1.2 Survey administration

This study collected data online between October 2006 and January 2007.

Key advantages of online data collection are reduced response time, lower

45

administration cost (printing and postage savings should offset programming
costs) and reduced data entry error (Granello and Wheaton 2004). Four
invitations in total were sent to each people on the contact list. An invitation letter
(regular mail), an e-mail invitation, a postcard reminder and a ﬁnal e-mail
invitation were sent. Follow up phone calls were also made to a list of random
non—respondents to assess non-response bias.

Generally Cited limitations of online survey include lower
representativeness of the sample and lower response rate (Granello and
Wheaton 2004). Web-based survey may exclude people who are not comfortable
using or do not have a ready access to the internet. However, given the target
respondents were production and engineering managers, nearly all should have
had access to the internet. Misrepresentation of the population, therefore, should
be limited. In fact, from the follow-up phone conversation with 102 non-
respondents, none of the eligible people to participate in this study mentioned not
having access to the internet as a reason for not participating.

While responses have been shown to be lower in online surveys, to
enhance the total number of responses, surveys were sent to as many people as
possible; This was possible due to lower administration cost (Granello and
Wheaton 2004). Several strategies to increase response rate were also used.
These include multiple reminders (in more condensed intervals than done in
paper-based surveys), use of Closed-ended questions when possible to lower

abandonment or dropout rate, and respondent-friendly designs.

46

A few principles of respondent-friendly designs suggested by Granello and
Wheaton (2004) were followed. These included: a motivational welcome screen;
formats that are similar to paper-and-pencil formats; limited line length to reduce
the need for Ieft-to-right scrolling; and computer operation instructions for
different question formats at the location in the survey where the instructions will

be implemented, rather than at the beginning of the instrument.

4.1.3 Population, unit of analysis and sample selection

The population in this study is completed MPI implementation projects
conducted in the United States. Since the focus of this study is the effects of
characteristics of the invention source and the implementation teams on
implementation performance, required information is at the project level from the
participants. Target respondents were mid- to high-level managers and
engineers from companies in Standard Industry Codes (SIC) 20-39
(manufacturing) who had recently been involved in a completed MPI
implementation project. Personnel at these levels have direct involvement with
MPI implementation projects (of various sizes) while at the same time are
knowledgeable and experienced enough to assess and provide answers
pertaining variables in this study. Contacted respondents were also requested to
provide additional project members that can be contacted for a second opinion.
For projects that have multiple usable responses, averages of the responses
were used.

This study used a mailing list Obtained from Society of Manufacturing

Engineers (SME), a leading national organization of manufacturing and

47

production engineers. SME list contains 700,000 individuals who are associated
with SME as subscribers of SME magazines and as members. To ensure
representation of the sample in terms of plant size and industry, a stratiﬁed

sample is drawn from SME’S member list.

Table 4.1 Distribution of observations across industries by SIC code

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SIC Industry description %
20 Food and kindred products 1.78
21 Tobacco products 0.00
22 Textile mill products 0.59
Apparel and other ﬁnished products made from
23 fabrics 0.59
24 Lumber and wood products, except furniture 1.78
25 Furniture and fixtures 6.51
26 Paper and allied products 0.59
27 Printing, publishing, and allied industries 1.78
28 Chemicals and allied products 1.78
29 Petroleum reﬁning and related industries 0.59
30 Rubber and miscellaneous plastics products 3.55
31 Leather and leather products 0.59
32 Stone, clay, glass and concrete products 5.33
33 Primary metal industries 2.37
34 Fabricated metal products 26.63
Industrial and commercial machinery and computer
35 equipment 15.38
Electronic and other electrical equipment and
36 components 8.28
37 Transportation equipment 5.33
38 [Measuring analyzing and controlling instruments 5.92
39 TMiscellaneous manufacturing industries 10.65
Total 100.00

 

 

 

 

 

A set of keywords in the job title including management, senior, director,
leader, engineering, project, process, systems, production, operations, plant, and
quality was used to identify managers that were likely to have been involved in
implementing a new MPI. The identiﬁed list included managers from plants with

employment of at least 100 only. A total of 3,492 contacts were randomly

48

selected to closely match the distribution of companies across manufacturing
industries in the US, by SIC codes according to Almanac of Business and
Industrial Financial ratios (Troy 2006). Tables 4.1 and 4.2 summarize the

distribution of the 169 observations in the ﬁnal sample for the analyses.

Table 4.2 Distribution of observations

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Respondent's position %
First-level Management 7.48
Middle Management 51.40
Upper Management 34.58
Others (e.g., researcher, engineer, consultant) 6.54
Company's employment globally %
Between 501-1000 people 41.46
Between 1001-5000 people 10.57
More Than 5000 people 47.97
Scope of impact %
A single work center 11.90
A production line 37.50
A department 27.98
The entire plant 14.88
Multiple plants 7.74
Plant's employment %
100-249 37.72
250-499 21.56
500-999 17.37
1000-2499 15.57
More than 2500 7.78
Type of technology %
Decision Support 2.96
Communication 4.14
Equipment replacement 75.15
Procedure change 11.83
Process Automation 5.92
Primary reasons for implementation"z %
A new product launch 20.71
A product modiﬁcation 9.47
Kaizen/Continuous improvement 15.98
Parent corporation requirement 1.18
Customer's demand 46.15
Vendor's suggestion 7.10
Others (e.g., cost saving, outdated/broken equipment,
Erocess improvement, safety issue, capacity 27.22
xpansion)

 

 

2 .
Some respondents Cited more than one reason.

49

 

4.1.4 Selection and non-response bias

Two types of biases tend to persist in survey research: selection bias and
non—response bias. Selection bias occurs when the sample that is obtained
systematically excludes part of the population. Since target respondents of this
study may have been active in multiple process improvement projects, they were
asked to respond with regards to the most recent (within 9 months) completed
MPI implementation projects in which they were directly involved. This approach
was adopted to reduce tendencies to select certain types of improvement over
the other (e.g., toward successful, or large-scale projects).

Non-response bias occurs when respondents are systematically different
from non-respondents. To reduce this bias this study used two types of
incentives to appeal to people of different natures. Participants were promised an
executive summary of ﬁndings of this study, and offered a donation of $5 to their
choice of Habitat for Humanity, National Childhood Cancer Foundation, Red
Cross and UNICEF upon completion of the survey.

3,492 letters and e-mails were sent to managers asking for participation.
622 letters and/or postcards were undeliverable and returned, reducing the total
pool to 2,870. 202 managers responded or referred someone else that
responded on their behalf. However, of these 202 managers who responded, 41
did not complete the survey due to various reasons. 15 had never been involved
in implementing a MPI, 22 had not been involved in a project that was completed
within the last 9 months, two had insufﬁcient time, and two did not provide a

reason. This resulted in a total of 161 partial and complete surveys. An additional

50

10 people participated in the survey after follow-up phone calls. These people
either had never received the invitations or had not had time to do it. Later, two
responses were dropped due to excessive missing values, yielding a total
sample size of 169.

The test of difference between respondents and non-respondents
indicates that non-response bias should not be signiﬁcant in this study. The
majority of respondents (37 out of 39) who did not participate in the survey were
actually not part of the target population for this study. To further assess the
presence of non-response bias, phone calls were made to non-respondents.
Information on 359 randomly selected non-respondents became available. 257
were either no longer employed by the company, or had relocated. 36 reported
that they had never been involved or had not recently been involved in
implementing MPls. Only 66 of the non-respondents (18.38%) were actually
eligible to complete the survey.

Chi-squared tests for difference in the distribution of industry and plant
size between phone contacts and the whole contact list were conducted to check
whether the results from the phone contacts were legitimate to use to check non-
respondent bias. The chi-squared test results support that the 359 phone
contacts are representative of the original contact list. (Chi-squared = 4.64, p =
0.33 for 5 plant size categories; and Chi-squared =17.23, p = 0.574 for 19
industry groups). Table 4.3 and Table 4.4 report the distribution of called
individuals and the Chi-squared statistic for the test of difference between the

distribution of the called individuals and the rest of the contact list.

51

Table 4.3 Chi-squared test of equal distribution between called individuals and
the contact list with respect to plant’s employment

 

 

 

 

 

 

 

 

 

 

 

1000-
Plant’s employment 100-249 249-499 500-999 2499 > 2500 Total
Called people (Oi) 162 92 58 34 13 359
Expected (Ei) 150.51 100.24 51.61 37.52 19.12 359
(Oi-Ei)"2/EI 0.88 0.68 0.79 0.33 1.96 4.64 (p = 0.33)

 

With the numbers from the phone contacts, the estimate number of

qualiﬁed managers from the 2,309 (2,870-202-359) managers whom we had

never heard from is 424 (66/359*2309). With 165 who reported online and 66

who reported on the phone that they were qualiﬁed to participate, the total

estimate of qualiﬁed managers in the contact list came to 655, and the effective

response rate among qualiﬁed managers was 25.80% (169/655). The effective

response rate is comparable or better than recent similar studies (eg. Sila

(2007)). A recent survey by Tu et al. (2006) using SME mailing list reported a

10.7% response rate, further ensuring that the response rate for this study is

typical or even better than typical.

52

 

 

 

 

 

 

 

8F. "3 F0: 000 00.0 00.0 00.0 2.0 00.0 00.0 00.0 00.0 3.0 00.0 F00 00.0 00.0 00.0 SF 20 00.0 0.0 0.93-00
00F 00.0 00.0 00.F F0.0F00.0000$ F0.0 3.0 F00 00.: 00F F00 00F 0F.F F00 00F 00F 00F 00F 00090000....00002
00F 0F 2 0 0F 00 3 0 2 F 0 F 0 0 F FF 0 F F F. 090.000.2000
.98 00 00 F0 00 00 v0 00 00 F0 00 00 00 F0 00 00 F0 00 00 00 0008 0.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.0000 9m 9 000000.. 53> 000090300000: 0:0 00:092th c0058 5:39:06 .0300 F0 02 0000300420 m6 050.0

 

 

 

 

£0.10 0.: 00.0 00F 0F.0 00.0 F00 F00 00F 00.0 0.0.0 00.0 00.0 00.0 00F 00.0 F00 00.0 00.0 F0.F 00.0 002060
000 0032.00 00.0F :00 00.00 F000 00.0 FvFF 00.0 00.00 90 00.0F 00.0 0F.0 00.00 00.0 00.0 8.0 00.0 00: 00.0030
000 3 0F 0F 00 00 00 0 0 0 00 F FF F 0 0 F 0 v 0 090.000.000.00
.98 00 00 F0 00 00 e0 00 00 F0 00 00 00 F0 00 00 .00 00 00 00 0008 0.0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.008 9m 9 000009 530 00: 69:00 05 000 2032205 00:00 c0053 5:39:06 6:00 Fo F000 000030-20 30 0500.

53

Next non-response bias was further tested by comparing respondents and non-

respondents. Chi-squared tests for distributions of survey participants’

companies across plant size and industry were conducted. Chi-squared test

rejected the null hypothesis of no difference between the distributions of

respondents (N=169) and non-respondents (N=3331) across plant Size at alpha

= 0.05 (Chi-squared statistic = 12.72, p=0.013). Table 4.6 and Table 4.5 report

the distribution of the sample across the 5 plant size categories and 19 SIC

codes and the Chi-squared statistic for the test of difference between the

distribution Of the sample and the distribution of non-participants.

The sample overrepresents larger plants (employment 500 and above)

and underrepresents plants in smaller-size categories based on distribution of all

companies in the US. However, this is not a serious threat to the validity of

hypothesis tests. All plant size categories are fairly well represented, and it was

expected that larger plants had a greater probability of having MPI

implementation projects due to more resources to invest. The sample for this

study should be representative of the population of interest for this study.

Table 4.6 Chi-squared test of equal distribution between participants and non-
participants with respect to plant size.

 

 

 

 

 

 

 

 

 

 

 

1000-
Plant's employment 100-249 249-499 500-999 2499 > 2500 Total
Participants (Oi) 62 37 30 27 13 169
Non-participants (E) 71.30 47.70 24.00 17.19 8.80 169
(OI-Ei)"2/Ei 1.21 2.40 1.50 5.60 2.01 12.72 (p=0.013)

 

The chi-squared test, however, cannot reject the null hypothesis of no

difference between the distribution of respondents and non-respondents across

54

 

19 SIC codes (chi-squared statistics = 14.91 p=0.728, Table 4.5). Test of
difference in the mean plant size and performance (budget, schedule, technical
and overall) between respondents in the ﬁrst two waves and the latter two waves
also Show no signiﬁcant difference between the two groups (p = 0.29, 0.27, 0.71,
0.27, 0.26). These tests give evidence that the projects that are reported are
representative of projects that are carried out in manufacturing companies in the

US

4.2 Measurement instrument

4.2.1 Construction

This section discusses the construction of measurement instruments to
ensure validity for variables in this study. It discusses procedures taken to ensure
the validity of construct measures in the survey development stage, reasons for
the use of perceptual measures, and measures used to prevent common method
bias.

Validity refers to correspondence between a variable/construct and the
operational procedure to measure it. One disadvantage of collecting information
via mailed surveys is that respondents do not have the opportunity to clarify the
meaning of each question. It is thus very important that questions are written so
that variation in interpretation is limited. This requires clearly written and easy to
understand language free of unclear jargon/technical terms. To ensure this, as

much as possible, measures developed and tested in previous studies were

55

modiﬁed and used. In addition, deﬁnitions of key terms were provided in the
online survey to help Clarify the scopes and meanings of terms used in the
quesﬁons.

To enhance validity of the questions, ﬁve managers with experiences in
implementing MPI read the survey questions and commented on readability and
appropriateness of questions. Adjustments were made to these questions
according to the comments from these managers.

Next, four rounds of Q-sorting for questionnaire items with Likert-type
scales were conducted following procedures recommended by Stratman and
Roth (2002). Q-sorting is an iterative process that helps assess and improve
validity and reliability of constructs (Nahm et al. 2002). In each round participants
were provided with deﬁnitions of each construct and a set of questionnaire items
in random order. Participants were requested to assign each questionnaire item
to a construct he/she believed the item was intended to measure. At the end of
the round participants input were used to help reﬁne questionnaire items to
improve readability and discriminant validity. Revised questionnaire items and
construct deﬁnitions were then given to a new set of participants.

Participants in the ﬁrst two rounds consisted of nine doctoral students at
Michigan State University. The ﬁrst round had ﬁve participants and the second
round had four participants. Participants in the third round were ﬁve manager-
level personnel in manufacturing companies. Participants in the fourth round
were 37 MBA students at Michigan State University. The average item placement

ratio (percentage correctly sorted) for the 41 questionnaire items in the ﬁnal

56

round is 75.87% with a standard deviation of 20.79%. Item placement ratio of
70% or greater is generally considered acceptable (Moore and Benbasat 1991).
A few items do not pass this criterion possibly due to the high interrelation among
the constructs and the large total number of items that had to be sorted. Some
adjustments were made to these items after the ﬁnal round as appropriate. The
placement rate for the ﬁnal round of Q-sort is included as an Appendix 1.

This study used perceptual questions to measure most constructs. While
objective questions are not affected by subjective interpretation (Starbuck and
Mezias 1996), they are not always available or suitable. For example, a latent
variable such as psychological safety is not objectively measurable. In addition,
when objective measures are available, in different industries and project size
and type the same value of a variable may imply signiﬁcant difference. For
example, timeliness may be measured by lateness as a percentage of original
schedule, but for a project that was expected to last a day versus a project that is
expected to last a year, 100% late means one day late versus one year late.
Lastly, participants may be reluctant to share some objective information,
especially cost ﬁgures. This can result in lower response rate and higher missing
value rate. For these reasons, perceptual measures were used to capture the
majority of the variables of interest in this study. Procedures recommended by
Starbuck and Mezias (1996) were followed to improve accuracy of perceptual
measures to reduce perceptual bias.

Lastly, common method bias refers to bias in ﬁndings that is attributable

to the measurement method rather than to the construct of interest (Bagozzi and

57

Yi 1991). When one method is used to capture all constructs, It can result in
observed dependent relationships between measures of different constructs that
in fact do not exist (Podsakoff et al. 2003). However, Malhotra et al.’s (2006)
ﬁndings conﬁrm Cote and Buckley’s (1987) ﬁndings that common method biases
are usually not serious in single-method studies and that the biases are lower in
studies for variables in marketing and information system than in psychology,
sociology and education where variables are more abstract. Common method
bias should be low in this study since the variables in this study are relatively
more concrete, dealing with actual goals achievement, system and technology
and Characteristics of people on the team.

In order to reduce common method bias recommendations from Podsakoff
et al. (2003) were followed in constructing the measures and the survey. First,
the measures for independent and dependent variables were psychologically
separated by having different response formats for the main independent
variables (distance variables) and dependent variables (ﬁll in the blank vs. Likert-
type scales). Second, question order was counter-balanced in the questionnaire
with dependent variables positioned before independent variables to reduce
participant-inferred cause-effect relationship that were not actually present in the
project. Lastly, the questionnaire was carefully constructed to reduce any
ambiguity in interpretation, and anonymity was ensured to participants to reduce
evaluation apprehension.

Harman’s single-factor test was conducted conﬁrmatory factor analysis

(CFA). The CFA procedure compared chi-square difference from two CFA

58

models, one with one method factor, and the other with hypothesized constructs.
The Chi-square difference of 455.627 with 36 degrees of freedom is highly
signiﬁcant (p= 4.20E-74). Therefore, the CFA indicates that common method bias

does not signiﬁcantly bias the results of this study (Podsakoff et al. 2003).

4.2.2 Measurement items

This section discusses the speciﬁc items that were used in the ﬁnal
survey. Items that empirical analyses of construct validity suggested low validity

were dropped. The full survey is included as Appendix 2.

Performance variables

This study focuses on three types of MPI implementation project
performance: implementation cost, timeliness and technical improvement. In this
study, these three performances are measured as goals achievement to allow
comparability across observations. These measures were modiﬁed from earlier
studies on product innovation projects performance (Primo and Amundson 2002;
Swink 2002). Since relative importance of each performance criteria is different in

each project, this study also sought to measure overall performance.

Cost performance

Cost performance is deﬁned as the degree to which expenses for the
implementation project adhered to the initial planned budget of the project. The
survey asked respondents to report their perception on different facets of cost

performance compared to original budget with a 5-point Likert-type scale.

59

- How much man power was actually used in the planning, testing
and installation of the new practice/technology? (signiﬁcantly
more than originally planned to signiﬁcantly less than originally
planned)

- How much capital expenditure was actually used in the
implementation? (signiﬁcantly more than originally planned to
signiﬁcantly less than originally planned)

- To what extent was the original budget/projected cost Of the
implementation project met? (signiﬁcantly more than originally

planned to signiﬁcantly less than originally planned)

Timeliness

Timeliness is deﬁned as how fast the implementation project is executed.
Three anchors are used to judge timeliness: group’s expectation, other similar
projects undertaken in the company, and top management’s expectation.
Respondents were asked to state on a 5-point Likert-type scale format the
degree to which they perceive how long the implementation took compared to the
three anchors.

- How much time did the implementation actually take?
(signiﬁcantly more than originally planned to signiﬁcantly less than
originally planned)-

- Relative to other similar implementation projects undertaken in

this company, how long did this project take? (signiﬁcantly more

time to signiﬁcantly less time)

60

- How satisﬁed were top management with how fast the
implementation project took place? (very dissatisﬁed to very

saﬁsﬁed)

Technical performance

Technical performance is deﬁned as the degree to which the performance
of the new practice/technology matches the original operational improvement
targets of the project (e.g., improving production quality, time or cost). Five items
were used to measure overall technical achievement. Respondents were asked
to indicate on a 5-point Likert scale the degree to which they agree/disagree with
the following statements regarding the different facets of technical achievement
of the project.

- Using the new practice/technology, we achieved greater
productivities than we originally planned. (strongly disagree to
strongly agree)

- The new practice/technology improved the quality of our process
more than we had originally expected it would. (strongly disagree
to strongly agree)

- The new practice/technology improved our responsiveness more
than we had hoped. (strongly disagree to strongly agree)

- The new practice/technology achieved signiﬁcantly better overall
technical improvements than we had originally expected it would.

(strongly disagree to strongly agree)

61

- The new practice/technology performed signiﬁcantly better than
we had originally expected it would. (strongly disagree to strongly

agree)

Excluded item:
- The new practice/technology was not as effective or efﬁcient as
we had originally hoped it would be. (strongly disagree to strongly

agree)

Overall performance

In addition to individual category performances, this study also measured
overall performance. Each dimension of performance may be affected by
conscious managerial decisions and intentional compromise. Thus, while an
individual measure gives a more precise indicator of an outcome, it may not be
as meaningful a measure of success or excellence as overall performance is.
The survey asked participants for the weight assigned for each category of
performance based on perceived importance of the success for the project.

- In evaluating the performance of this implementation project, what

percentage weight (or importance) was given to each criterion of
success? (Answers must add up to 100%) (Technical goals

achievement; Budget attainment; Schedule adherence)

62

Distance from invention source

This study makes the distinction between the two processes in MPI:
invention and implementation (see Chapter 3). This study measures three
different types of distance between the implementation team and the invention
source. The deﬁnition of the distance construct recognizes that there may be
more than one source for an innovation. To capture this, we used the simple
average and weighted average of the distance from each source. For weighted
average, the weight is the proportion of the innovation attributed to the source.
Degree of separation of a source of technology and the adopting unit was
measured along three dimensions: physical separation, organizational separation

and temporal separation.

Physical distance was measured by the distance between the inventing
unit and the implementing unit.
- How many miles away was each design/invention source from the

implementation site?

Organizational distance was measured as the strength of the relationship
between implementation team members and the invention source. When the
source was within the same plant, organizational separation was set to zero.
Strength of relationship was measured with a 5-point Likert-type scale. The
following question was used to measure organizational distance from contributing

corporate support functions and the three most signiﬁcant external sources.

63

- Please indicate the item that best describes the working
relationship that personnel at the implementation site had with the

source. (not Close at all to extremely Close)

Temporal distance was measured by asking respondents to estimate how
long prior to the implementation process began had each source made the
contribution to the design of the MPI.

- Approximately how many years before implementation did each

source make their contribution to the design/invention?

Source contribution

Participants were asked to provide the contribution on the design of the invention
each of the most important 3 sources of the invention had.
- What percentage of the design/invention of the new
manufacturing practice/technology you described was done by:
(Answers must add up to 100%)
- Personnel (managers, engineers and workers) at the
implementation site? (%)
- Corporate support functions (engineers and others not
located at the site)? (%)
- External sources (vendors, suppliers, university labs,
etc)? (%)
- What percentage of the total design/invention did External source

[1, 2 and 3] contribute? (%)

64

The following is the formula for calculating the weighted average of each
of the three distances for each observation.

2 WrikDiik
Distance (k = f

where Distance is the calculated distance, W is the weight for each source
by contribution, and D is the measured distance. The index i represents the
observation, j represents the distance type (physical, temporal and
organizational) and k represents the distance source (internal, corporate support

functions, or external sources)

Team members’ technical competence
Team member's technical competence is deﬁned as the combined quality and
depth of knowledge that was represented by people on the implementation team.
Three 5-point Likert scale questions modiﬁed from Subramaniam and Youndt’s
(2005) measures for human capital were used.
- Implementation team members were highly skilled. (strongly
disagree to strongly agree)

- Implementation team members were considered among the best
people in the organization. (strongly disagree to strongly agree)
- Implementation team members were experts in their particular

jobs and functions. (strongly disagree to strongly agree)

65

Functional diversity

This study used the entropy-based index developed by Pfeffer and
O’Reilly (1987) to measure heterogeneity in a system, and adopted by Ancona
and Caldwell (1992) to measure diversity. This index takes into account the
proportion of team members from each function, as well as the number of
functions present. Functional diversity is measured as, H: -Z(pi(ln pi)), when i
denotes a represented functional home on the project team, and pi denotes the
fraction of function i on the team.

The following question asked the participant to identify the proportion of
the MPI implementation team from 8 functions that have been identiﬁed in
previous studies as functions that are typically involved with MPI implementation.

- What percentage of the personnel on the implementation team

was from the following functional area? (Answers must add up to
100%) (Manufacturing engineering; Product design engineering;
Quality control; Production control; Maintenance;

Production/Manufacturing; Purchasing; Management information

system; Others)

Psychological safety

Measures of psychological safety are drawn from Edmondson (1999).

These measures tap the perception of the respondent on how the team members

66

contribute through their reaction to opposing viewpoints and new ideas and their
eagerness to help. The measures were 5-point Likert scale questions.

- If an implementation team member made a mistake on this team,
it was held against him/her (R). (strongly disagree to strongly
agree)

- Team members feared that bringing up problem/issues would
have negative effects on their career prospects (R). (strongly
disagree to strongly agree)

- No one on this team would deliberately act in a way that
undermined my efforts. (strongly disagree to strongly agree)

- Each member of the implementation team felt they were valuable
to the team. (strongly disagree to strongly agree)

Excluded items:

- People on this team sometimes rejected other members.

- It was safe to take a risk on this team.

Control variables

Several variables potentially interact, are correlated with the variables in
the model, or are expected to explain signiﬁcant amount of variances in the
performance variables. These variables need to be controlled for to reduce
biases in the results and enhance the precision of the ﬁndings. Variables that are
likely to be correlated with the focal variables include provenness of technology,

aggressiveness of goals, communication effectiveness within the implementation

67

team and team size. Other variables that were expected to have signiﬁcant
explanation power include project selection rigor and implementation radicalness
and other demographic variables including: type of process (job shop, batch,
assembly line and continuous), type of technology (processing vs. information
and decision support technology), industry (discrete vs. continuous) and plant

size.

Provenness of technology

Provenness of technology refers to the degree to which an adopted
technology is free of technical problems. Provenness is potentially related to
temporal distance. The longer an invention has been around (temporal distance),
the more likely it will have been ridden of bugs and irregularities. An invention is
usually easier to implement if they have been proved and implemented since
their effects on the organization are more predictable (Leonard-Barton 1988;
Edmondson et al. 2003). While closely related, provenness should have an
opposite effect to temporal distance on implementation performance. Thus, this
study needs to control for its effect to not bias the estimation of the effect of
temporal distance on performance.

Provenness was measured by asking respondents to rate on a 5-point
Likert scale the degree to which a MPI invention, that they were involved in the
implementation process, has been successfully adopted by other organizations.

- Before we implemented this practice/technology, its effectiveness

had been demonstrated by successful adoptions in many other

organizations. (strongly disagree to strongly agree)

68

This practice/technology had been implemented successfully in

other organizations before we adopted it. (strongly disagree to
strongly agree)

Excluded items:

- Before we started the implementation process, there were known
unresolved problems with the practice/technology itself. (R)

- Irregularities of this practice/technology had been resolved before
we adopted it. (5-point Likert scale ranging from strongly disagree

to strongly agree)

Goals Aggressiveness

This study measured performance as goals achievement. However, goals
achievement is also affected by how optimistic and/or aggressive are the initial
goals. While it has not been Clearly argued in the past, Menon and Pfeffer’s
(2003) ﬁnd that organizations can value external knowledge more highly than
internal knowledge, partially suggesting that invention externality may be related
to optimism of initial predictions (and hence the aggressiveness of the goals) of
the effect it will have on the organization. Conversely, the not-invented-here
syndrome (Katz and Allen 1988) can cause implementing organization to view
external invention more negatively. Because of the possibility that distance to
invention source may be associated with how an organization may set a goal of a
project, this study explicitly measured and controlled for goal aggressiveness. In

this way, its impact does not bias the effect of invention separation on

69

implementation performance. At the same time it gives greater clarity for the rest

of the ﬁndings as its effect on performance is factored out.

Aggressiveness is measured as how likely the respondents felt at the

beginning of the project about the ability of the organization to achieve the goals.

This study used one 5-point Likert scale question for each performance variable.

The question asked the extent to which the participant agree or disagree with the

following statement.

At the start of the project, we felt the original project schedule was
overly optimistic. (strongly disagree to strongly agree)

At the start of the project, we felt the original project budget was
overly optimistic. (strongly disagree to strongly agree)

At the start of the project, we felt the original project technical

goals were overly optimistic. (strongly disagree to strongly agree)

Excluded items:

At the start of the project, we were conﬁdent that we could
achieve the project budget.

At the start of the project, we were conﬁdent that we could
achieve the project schedule.

At the start of the project, we were conﬁdent that we could

achieve the project's technical goals.

lntra-team communication effectiveness

70

Diversity affects breadth of knowledge and access to external knowledge,
but also increases communication difﬁculty among team members (Cohen et al.
1996; Zahra and George 2002). By controlling its negative effects through
difﬁculty in communicating, the effect of increased breadth of knowledge due to
diversity can be more Clearly studied. Additionally communication effectiveness
may be Closely related to psychological safety since the level of psychological
safety should inﬂuence team members’ perception of whether communication
among team members is open and encouraged.

Infra-team communication effectiveness was measured as the
respondent’s perception of whether communication among team members was
useful, and satisfactory. The following items were modiﬁed from Earley and
Mosakowski (2000) and Moenaert and Souder (1996).

- Implementation team members talked openly and freely. (strongly

disagree to strongly agree)

- Implementation team members did not seem to understand what
one another was saying during their discussions. (R) (strongly
disagree to strongly agree)

- When implementation team members shared information it was
likely to be used. (strongly disagree to strongly agree)

- Team members often shared valuable information. (strongly
disagree to strongly agree)

Excluded item:

- Team members shared information in useful formats.

71

Implementation team size

Several reasons suggest that team size should be controlled for. Past
research on group performance suggests that performance is negatively
associated with team size (Gooding and Wagner III 1985). On the other hand,
team size is a natural limit for functional diversity; the smaller a team is, the less
functionally diverse it can be. At the same time the collective knowledge of the
team which inﬂuences team’s absorptive capacity undoubtedly increases with
team size. Further, psychological safety may be negatively associated with team
size. As the team size grows, a person may be more intimidated to voice
opinions, discuss his mistake or request feedbacks. Implementation team size

will be measured as how many people were in the team (Swink 2000).

- At the point in the implementation project when the number of
staff was at its peak, approximately how many persons were:
0 Assigned full-time to the implementation project?
0 Assigned on a part-time basis only to the implementation

project?

Project selection rigor

Project selection rigor is deﬁned as the degree to which past experiences
and existing working environment are used in selecting the project. Projects that
are rigorously chosen should have goals that are more achievable and

technologies that are better ﬁt to use. The following items were modiﬁed from

lansiti (2000).

72

- Before choosing this practice/technology, we compared the
resources that were required (e.g., tools, equipment, operators'
skills) to what we had. (strongly disagree to strongly agree)

- In choosing this practice/ technology, we used detailed
information on the existing user environment, product and
production system. (strongly disagree to strongly agree)

- In choosing this practice/technology, we used information from
past experiences in implementing new manufacturing processes.
(strongly disagree to strongly agree)

Excluded items:

- Before Choosing this practice/technology, we did a thorough
feasibility study.

- Before choosing this practice/technology, we followed rigorous

technology selection guidelines.

Implementation radicalness

Implementation radicalness is deﬁned as the required amount of change
that need to take place. Two questions capture changes required for operators of
the new MPI and one for changes in tools and equipment. The measures were
modiﬁed from Govindarajan and Kopalle (2006).

- In order to use the new practice/technology, the operators/users

had to learn new skills and procedures. (strongly disagree to

strongly agree)

73

- Much training was required for operators/users in order to use the
new practice/technology. (strongly disagree to strongly agree)

Excluded item:

- The implementation of this practice/technology required

signiﬁcant Changes in our assets such as tools and equipment.

Demographic variables

This study also included demographic variables that may affect the
outcome of an MPI implementation project. These include type of industry
(discrete vs. continuous), type of technology (communication and decision
support vs. processing), environment in which the MPI was introduced to (job
shop, batch, line and continuous process), scope of impact and plant size.

Type of industry was determined by the company’s SIC code, and type of
technology was determined by project description provided by respondents. The
following were the items used for other demographic variables.

- Which of the following types of operations best describes the
process that was affected by the implementation of the new
practice/technology you referred to?

0 Job shop (Production of small volumes of a large number of
different products)

0 Batch (Production of large volumes of similar items on a repeat
basis)

0 Assembly line (Production of discrete parts moving from

workstation to workstation at a controlled rate)

74

0 Continuous ﬂow (Processing of undifferentiated materials such
as petroleum or chemicals)

- What was the scope of operations affected by this
implementation? (A single work center, A production line, A
department, The entire plant, Multiple plants)

- How many people are employed at the manufacturing plant where

the practice/technology was implemented?

4.3 Scale reliability and validity

This section discusses the empirical testing of reliability and validity of the
measures. These include testing of single response bias, unidimensionality,
reliability and convergent and discriminant validity (O'Leary-Kelly and Vokurka

1998; Boyer and Verma 2000).

4.3.1 Single response bias

Single response bias or single rater bias refers to the possibility that a
given response may contain “subjective bias due to an individual’s unique
perspective and limited access to information” (Boyer and Verma 2000). For a
total of 12 projects, multiple respondents were available for eight items. Single
response bias was examined by inter-rater reliability (James et al. 1984).
Average inter-rater reliability is 0.694 for the eight items that are available with
multiple respondents in 12 projects. The average inter-rater reliability for each

item ranges from 0.47 to 0.81. This suggests that Single response bias is not

75

signiﬁcant. Thus, this study used the means of the items across respondents in

the analyses.

4.3.2 Construct validity

Unidimensionality and convergent and discriminant validity of the Likert
scale items were tested with conﬁrmatory factor analyses (CFA). Three separate
CFA models were run for independent variables, dependent variables and control
variables because of sample size limitation. Tables 4.2-4.4 report the CFA results
as well as the Cronbach’s alpha. Since the data exhibit multivariate non-normality
(indicated by high Mardia’s coefﬁcient), generalized least squares method (GLS)
in EQS was used to estimate the measurement models instead of maximum
likelihood. Fit indexes in models produced by GLS estimations have been shown
to be less affected by the combination of non-normal data and small sample size
(West et al. 1995).

In all three models ﬁt indices of the CFA models suggested
unidimensionality of the measures. The comparative ﬁt index (CF I) and non-
normed fit index (NNFI) are both greater than 0.95, RMSEA is below 0.05 cutoff
point (Hu and Bentler 1998; Hu and Bentler 1999), and GFI is above 0.90 (Hu
and Bentler 1995). While NFI is below traditional cutoff point in all three models,
given the small sample size and non-normality NF I may not be appropriate to use
to judge model ﬁts. NFI has been shown to underestimate its asymptotic values

on non-normal data with small sample sizes (Hu and Bentler 1995).

76

 

Table 4.7 CFA results for independent variables

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

g
.E .9 2 ._
<0 '7) 9. 00;.
9 0 8. a < 8 g
'o 3 E E UJ .‘0 “' 4:
Construct and Measurement Items 3 T; 8 ﬂ <>( 5 ‘3»: g
E .-. a 9 é a s
‘U o I U
.6 E 0 v
0)
Independent variables 0.155
Physical distance 1.000
Temporal distance 1.000
Organizational distance 1.000
Team's technical competence 0.704 0.442 0.692
Team members were highly skilled. 0.660 7.434
Team members were considered the
best in the organization. 0.694 7.853
Team members were experts in their
functions. 0.640 7.349
Team's functional diversity 0.846 0.647 0.823
The team included people from diverse
functional areas. 0.752 10.137
Many functional areas were
represented. 0.796 10.989
The team included people from
different functional groups. 0.861 12.142
Team's psychological safety 0.761 0.445 0.705
Mistakes were held against team
members. ( R) 0.658 7.782
Team members feared bringing up
problem/issues. 0.630 7.306
No one on this team would deliberately
undermine my efforts. 0.625 7.281
Each member felt they were valuable
to the team. 0.747 9.129

 

 

 

 

 

 

 

NFI = .760 NNFI =.991, CFI =.994, GFI =.954, RMSEA =.009
Chi-Square =50.743 based on 50 degrees of

freedom (p=.444)

 

 

77

Table 4.8 CFA results for dependent variables

 

 

8)
.E .9 .‘0" L
a s 3. 9 9‘».-
g 8 35 LIJ .w ““3 2
Construct and Measurement Items .5 E 8 33 2 5 g g
E t e 2 é a s
U _.I o I 0
.6 E 5 v
a)
Dependent variables 0.260
Budget performance 0.758 0.534 0.714
How much man power was actually used? 0.418 4.977
How much capital expenditure was actually
used? 0.720 8.190
Original budget of the implementation
project was met. 095410.186
Schedule performance 0.640 038010.607
How much time did the implementation
actually take? 0.759 7.995
Relative to similar projects, how long did
this project take? 0.571 6.247
How satisﬁed were top management with
the speed? 0.487 5.401
Technical goals achievement 0.820 0.477 0.827
We achieved greater productivities than we
originally planned 0.698 8.896
Process quality improved more than we had
expected 0.688 8.655
Responsiveness improved more than we
had hoped 0.686 8.636
Overall improvements were better than we
had expected. 0.657 8.211
The new technology performed better than
we expected. 0.724 9.509

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NFI = .801, NNFI =.989, CFI =.993, GFI =.961 , RMSEA =.012
Chi-Square =35.848 based on 35 degrees of

freedom (p=.43)

 

 

 

All measured variables load signiﬁcantly (p<0.01) in the expected direction

with the intended latent constructs. This suggests convergent validity (Bagozzi et

78

al. 1991; O'Leary-Kelly and Vokurka 1998). The square of the correlation
between any two factors is lower than average variance extracted for each of the
construct in all three models. This shows that the constructs have discriminant
validity (Fornell and Larcker 1981 ). The LM test is also not signiﬁcant in the
model for independent variable and the model for control variable, suggesting
discriminant validity at the item level. The LM test was signiﬁcant for two
variables in the dependent model. However, since the performance measures
were not included in any same regression model, and covariations among
performance were expected, this was not a serious threat to discriminant validity.
Cronbach’s alpha and Werts et al.’s (Werts et al. 1974) composite reliability are
all greater than 0.60 suggesting adequate reliability for the constructs (Nunnally
1967)

Construct scores were created as the weighted average of measured

variables. The weights are factor loadings from the CFA models.

79

Table 4.9 CFA results for control variables

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(D (U
-e-0 .C h
8 "17> .9- 2“».
.IS (I) a) 8. é‘ < 3 CC)
'9 CD :3 E = UJ 5” ’3: 4:
Construct and Measurement Items :0 E E o 5 > 5 w E
'o m > O = <( (O (I) a)
C .9 01- (n a) .0 .C t
59 'ﬁ '“ r: 2’ 0
m g 9 =5 °
0
Control Variables 0.399

Provenness 0.826 0.713 0.797

The effectiveness had been demonstrated. 1.000 17.485

This technology had been implemented

successfully before. 0.652 8.916

Selection rigor 0.707 0.452 0.692

Resource requirement were compared to what we

had. 0.510 5.863

Information on the existing environment was used. 0.746 8.825

Past experiences were used. 0.734 8.829

Implementation radicalness 0.781 0.657 0.725

Operators/users had to learn new skills and

procedures 0.561 7.349

Much training was required for operators/users. 1.000 17.370

Goal aggressiveness 0.687 0.429 0.617

Original project budget was overly optimistic. 0.766 7.766

Original project schedule was overly optimistic. 0.658 6.847

Original project technical goals were overly

optimistic. 0.517 5.675

lntrateam communication effectiveness 0.801 0.505 0.759

Team members talked openly and freely. 0.738 9.663

Team members did not understand one another. (R

) 0.568 6.887

Shared information was likely to be used. 0.755 10.235

Team members often shared valuable information. 0.764 9.808

 

 

=.015

 

(p=.391)

Chi-Square = 67.537 based on 65 degrees of freedom

NFI = .748 NNFI =.980, CFI =.986, GFI =.943, AGFI =.907, RMR =.053, RMSEA

 

 

80

Chapter 5 Analyses and Results

This chapter summarizes results of hypotheses testing via regression
analyses. The variables in the regressions are factor scores based on CFA
discussed at the end of Chapter 4. First, the analyses and the rationale behind
each model are discussed. Next, results from the regression models are

summarized.

5.1 Analyses

Ordinary least squares (OLS) hierarchical moderated regression models
were run based on the hypotheses for four types of performance: budget,
schedule, technical goals achievement and overall performance. As discussed in
Chapter 4, two operationalizations of IID (invention-to-implementation distance)
were employed. These were simple average IID and weighted average IID by
percentage contribution.

Cook’s distances were calculated to identify observations that may have
undue inﬂuence on the estimates in the regression. As a result four Observations
with the highest inﬂuence were omitted from the analyses (same observations for
both simple average IID, and weighted average IID analyses).

The Shapiro-Wilk test for normality of error terms from the regression
models showed that the distribution of the residuals are not signiﬁcantly different
from the normal distribution. Breusch-Pagan test also showed no violation of the

homoskedasticity assumption.

81

Lastly, due to the concern for multicollinearity arising from correlations
among product terms, focal variables in this study were mean-centered.
Tolerance and Variance inﬂation factors (VIF) were used to detect
multicollinearity. Tolerance is a measure of how much an independent variable
can be explained by all other independent variables in the model. It is calculated
as 1-R2 of the regression of the independent variable on the rest of the
independent variables. VIF, the reciprocal of tolerance, over 10 is an indication of
excessive inﬂuence on standard error estimates (Neter et al. 1996). The highest
VIF is 2.61 for models with simple average IID, and 3.97 for models with

weighted average IID. This suggests no strong multicollinearity effect.

5.2 Results

This section summarizes ﬁndings from the regression models. Results with a
priori hypotheses are reported as one-tail tests and results without a priori
hypotheses as two-tail tests. All results are reported at the 0.05 level of
significance. Coefﬁcients for IID variables are expected to be negative, while
coefﬁcients for absorptive capacity-related variables and the all the interaction
terms are expected to be positive (Sharma et al. 1981; Baron and Kenny 1986).
Table 5.1 - Table 5.4 below provide the descriptive statistics of the variables in

the regression analyses and the correlations among them.

82

Table 5.1 Descriptive statistics Of variables for regressions with simple average

 

 

 

 

 

 

 

 

Std.

Variables Obs Mean Dev. Min Max
Budget performance 165 2.68 0.67 1.00 4.66
Schedule performance 165 2.87 0.74 1.27 5.00
Technical performance 165 3.69 0.66 1.59 5.00
Overall performance 165 3.31 0.61 1.60 4.93
Project selection rigor 165 3.92 0.60 1.00 5.00
Team size 165 2.07 0.91 0.00 5.09
Provenness 165 3.33 1.02 1.00 5.00

 

Communication effectiveness 165 4.01 0.49 2.26 5.00
Project goals aggressiveness 165 2.78 0.69 1.00 4.73

 

 

 

 

 

 

 

 

 

 

 

 

 

Physical distance (miles) 165 559.92 1225.44 0.00 8000.00
Temporal distance (years) 165 1.23 1.56 0.00 11.67
Organizational distance 165 1.93 1.27 0.00 5.00
Team's technical competence 165 3.92 0.60 2.00 5.00
Functional diversity 165 0.52 0.18 0.00 0.89
Psychological safety 165 4.01 0.67 1.52 5.00

 

Table 5.2 Descriptive statistics of variables for regressions with weighted
average

 

 

 

 

 

 

 

 

Std.

Variables Obs Mean Dev. Min Max
Budget performance 165 2.68 0.67 1.00 4.66
Schedule performance 165 2.87 0.74 1.27 5.00
Technical performance 165 3.69 0.66 1.59 5.00
Overall performance 165 3.31 0.61 1.60 4.93
Project selection rigor 165 3.92 0.60 1.00 5.00
Team size 165 2.07 0.91 0.00 5.09
Provenness 165 3.33 1.02 1.00 5.00

 

Communication effectiveness 165 4.01 0.49 2.26 5.00
Project goals aggressiveness 165 2.78 0.69 1.00 4.73

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Physical distance (miles) 165 330.36 980.53 0.00 8531.05
Temporal distance (years) 165 1.28 1.72 0.00 14.00
Organizational distance 165 0.80 0.84 0.00 4.00
Team's technical competence 165 3.92 0.60 2.00 5.00
Functional diversity 165 0.52 0.18 0.00 0.89
Psychological safety 165 4.01 0.67 1 .52 5.00

 

83

 

00.0 v 0 ..

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0F.0v0 .
00.0-2000 F00- 00.0 2.0000000 tvwd No.0 :90 30me 620205930 .mF
0F.0 F00- 00.0-tF0010F0 00.0- 00.0 00.0 imFd- >F_0.0>_0_0co_Focsn_.vF
00.0 $00.0 00.0 000 1.5.0 000 V0.0 00.0- :0 00.00088
5000. 0.E00._. .mF
03.0 0F.0- F00 F000 0F.0 .mFd- 00.0- FFN.0-....NN.0-0000F0_0000000.005 .NF
F00- 0F.0 0F.0 00.0- 00.0 00.0 00.0 00.0- F00- 0.000
00:00.06 _0..OQE0._. . F F
00.0 00.0- 00.0 00.0 00.0 00.0- 00.0- 00.0. 00.0- 00:60
000906 _00_0>cn_ .9
F 00.0- 00.0- 00.0 00.0 00.0- 00.0 .0F.0-....00.0. 000002002000
0.060 F0085 .0
F 00.0- 00.0 :vvgtmmdtwwd 00.0 03.0 000526000
00:00E3EE00 .0
F 00.0- 00.0 00.0- 00.0- 0F.0- 0F.0- 0005065....
F 000190000. 000. F00- 0N_0 800.00
F temdtmmd FF.0 :FF.0 5000000200 80.8.0.0
F :FF.0:Fm.0 tFed 00:0ant00=000>0.v
F $00.0 .mF.0 00:0E.ot00_00_:;00._..m
F $010 0000Ecot00 0300;0m .N
F 00:0E00t00 0.00.0.5 .F

0 0 F 0 0 v m N F

 

 

0000020 9900 FEB 0020000000 00F 0050.005 Fo 00.000.00.00 0.0 0300.

84

00.0 v 0 t

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0F.0v0 ..
F FF.0-..00.000.0-0F.0 00.0- 00.0-...00. F00- 00.0 :00. 2.000.000 00.0 .00 .0900 00.00.0500 .2
F 0F.0 0F.0 S0 .30 S0 F0.0-00.0-...F0.0....0F. 00.0- 00.0 00.0 .90. 00.020.000.800“. .3
F .2080. 00.0 00.0....000000 00.0 :00 00.0 00 00.0- :0 0000.038 605E852
F 00.0-...F00 F0.0..0F.0-00.0 00.0 ...FF.0....0F.0.00.0- FF.0-....0F.0- 8090.308000000LO.0F
F 00.0- F00- :0 00.0 00. 8.0 00.0 90 00.0- F00- .20008090.20.00E0FFF
F F00 00.0- 00.0 00.0-...0F0 00.0 30. 00.0 2.0 F8..E08§0.0.8_0Ea .9
F 000000-000 00.0 00.0- 00.0 .0F0-1000000002009000 0.80 80.8.0 .0
F 00.0- 00.0 ....$.0....00.0...00.0 00.0 .20 000002.800 :0_F8_SEE00 .0
F 00.0- 00.0 F00- 00.0- 0F.0- 0F.0- 08:00.60 .F
F F00 $80.80. 00.0- 00. 00.0 E85
F 3.000000 :0 FE .00.. 00.80.028.90.
F ...FF.0...F0.0:FF..0 800E508 .0050
F ....F0.0..0F.0 800E508 8.068 .0
F 3.00.0 000000000000 050050.01
F 00:0..Eot00 0005 .F
0F0F0F0FFF0F0 0 F 0 0 v 0 0F

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

000020 000.290; FEB 002000000. .8 0050:? F0 08:30:00 v.0 0500.

85

5.2.1 Budget performance
Table 5.5 summarizes hypotheses testing results on budget performance,

and Table 5.6 shows the regression results with simple average IID. Based on R-
Squared, the simple average IID model for budget performance explains 23.30
percent of the variance in budget performance (12.64% based on adjusted R-
squared). The hypothesis variables and their interactions account for 13.43% of
the variance in budget performance (difference in adjusted R-squared = 5.61%).

Table 5.7 shows results for budget performance with weighted average
IID. Based on R-Squared, the weighted average IID model for budget
performance explains 27.21 percent of the variance in budget performance
(17.10 percent based on adjusted R-squared). The hypothesis variables and their
interactions account for 17.34% of the variance in budget performance
(difference in adjusted R-squared = 10.07%).

Two project Characteristic variables are found to be signiﬁcant. The
coefﬁcient for project selection rigor is positive, and the coefﬁcient for project goal
aggressiveness is negative in both models. Both of these coefﬁcients have the
sign in the expected direction.

Hypothesis 1 is supported only for the effect of organizational distance on
budget performance with both operationalizations of IID. Hypothesis 2a is also
supported in the model with weighted average IID. There was no support for
hypotheses 3a and 4a.

There was evidence in support of Hypotheses 2b and 3b but none for
Hypothesis 4b. The coefﬁcient of the interaction term temporal distance *

technical competence is positive and signiﬁcant in models with both simple

86

average and weighted average IID, supporting Hypothesis 2b. The interaction
term temporal distance * functional diversity has a positive and signiﬁcant

coefﬁcient in both models, supporting Hypothesis 3b.

Table 5.5 Summary of hypotheses testing results for budget performance based
on a one-tail test at 0.05 level of signiﬁcance

 

Budget performance

 

Simple
average IID
Weighted
average IID

Hypotheses

 

H1 : Direct positive Physical distance

effect of invention-to-
implementation
distance (IID) Organizational
distance
H2a: Direct positive effect of technical
competence
H3a: Direct positive effect of functional
diversity
H4a: Direct positive effect of psychological
safety

 

Temporal distance

 

Supponed Supponed

 

 

Supponed

 

 

 

(4%; Positive Physical distance

 

moderating effectof Temporal distance Supported Supported

 

 

technical
competence Organizational
distance
Physical distance
H3b: Positive

 

moderating effect of Temporal distance Supported Supported
functional diversity

 

Organizational
distance

 

 

Physical distance
H4b: Positive

moderating effect of Temporal distance
psychological safety

 

 

Organizational
distance

 

 

 

 

 

87

5.0 v a 0200.05 .2. .mod v Q 09020:. ..

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

~00 00.0 F00 F00 V0.0 mms. 600.
2.0 FF.0 00.0 00.0 F00 8.00000. .2
00.0 FF.0 0F.0 0F.0 0F.0 00.0:00-m
00.0 00.0 00.0 00.0 F00 n. A 095
0F.0 00.0 00.0 9.0 00.0 2-..-.. .0...
00F 00F 00F 00F 00F 000 Fo .0952
Fwd 80 00v 004. F.F.w 0:001
00.0 00.0 3800 50.00 .. 00:90.0 0.0
00.0 00.0 30.00 :000 .. 00:90.0 0E0.
00.0 v0.0 30me £000 . 00:90.0 02.0
00.0- 00.0 3.0.020 - 00:90.0 0.0
. F0.F 0F.0 3.0.020 . 00:90.0 0E0.
F00 000 3.0.0020 .. 00:90.0 0.30
00.0- 00.0- 00:0F00Eo0 :09 .. 00:90.0 0.0
. FF 0 F0 00:0F00Eoo :02 - 00:90.0 0E0.
F00 00.0 00:0.00Eoo :02 .. 00:90.0 0.9.:
on. F F F0 mm. F 0F.0 no. F 9.0 .30me 60.00.0500.
«0. F- 00.0- . 0F. F- 0F.0- . 00.0- 0 F0- 3.0.020 0:000:00
V0.0 000 0 F. F 00.0 00.0 00.0 00:0000E8 _00.::00. 0.Em0-_-
. FF.0- F F.0- : 00.0- F00- : 00.0- 0F.0- 00:90.0 .0:000~.:00.O
F00- 00.0 00.0- F00- 00.0- 00.0- 0.000 8090.0 0.8%:
00.0 F00 00.0- 00.0. 00.0- 00.0- F00_.Ev 00:90.0 00.05.“.
t 00.0- 00.0- . 0F.0- 0F.0- .. FF.0. F00- : 00.0- 00.0- : 00.0- 00.0- 08002009000 0.000 800.0
00.0- 00.0- 00.0- 000- F00- 00.0- F00 v0.0 000 00.0 0030200000 :ozmoEsEEoo
00F- 00.0- 00F- FF.0- 00 F- 0F.0- 00F- 3.0. 04-3. 0F.0- 0005035
F F0 F00 F00 00.0 00.0 00.0 FF.0- F00- 00.0- 00.0- 00.0 E00F
.. 00.0 FF.0 .. F00 0F.0 00F 0F.0 . 00.0 0F.0 00F 30 .00.. 800200 60.8.0
0 0.00 F 900 F 90m. F 0000 F 000m. 00:08.0..00 F0005

 

 

0: 000.020 29:0 5.3 00:0E.ot00 00000 .8 0:000. 5.000.001 0.0 050.0

88

5.0 v 0 8.8.0... :
00.0 v 0 09020:. *

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F00 00.0 3.0 00.0 00.0 wws. F000
:0 9.0 00.0 2.0 8.0 8.00.0-0 .3.
F00 0F.0 0F.0 0F.0 90 00.00000
00.0 00.0 00.0 00.0 8.0 n. A 00.0
00.0 00.0 «0.0 00.0 00.0 FF-0.-c .00“.
00F 00F 00F 00F 00F 000 F0 .00802
00.0 0F.0 00.0 0F.0 F.F.v 0:00I
00.0- 5.0. .0800 00.00 . 00:00.0 0.0
00F 0F.0 30.00 00.00 . 00:0F0.0 0E0F
00.0 00.0 3000 00.00 .. 00:90.0 0.9.0
00.0 00.0 3.0.020 . 0890.0 0.0
2. 00.0 00.0 3.0.020 . 00:00.0 0E0F
0F.0- 00.0- 3.0.020 . 00:0F0_0 0.9.0
010 00.0 00:93:50 :09 . 00:90.0 0.0
. 00.0 0F.0 00:98:50 :09 . 00:90.0 9%:
VNF FF.0 08200800 :02 .. 00:90.0 0.30
00. F 0 F0 0 F. F F F0 00. F 0 F0 3000 .00.00_000>00
00.0- 00.0- .. 0 FN- F F.o- . mod- 0F.0- 5.0.020 .0:0_F0:0u.
. «0.0 $0 :1 0F.0 00.0 00.0 8:00.050 00.202 0.E00._-
2 end. Fwd- t. NON- mwd- s. F.F.N- NNd- 00:0F0_0 .0:0.F0N.:00.O
8.0 8.0 8.0- 5.0- 00.0- 8.0- .0000 8:900 0.853
00. F 0 F0 * F00 0F.0 . NF. F 0 F0 A00_.EF 000000 00.0.0.0
.. 00.0- F00- : 00.0- 8.0- .. FF.0- F00- : 8.0- «No- .. 8.0- 00.0- 088208.000 280 00.0.0
000- F00- F00- 00.0- F00- 00.0- 00.0 00.0 00.0 00.0 000:02Fo0t0 :0_F00.:0:.Eoo
00. F- F F0- 0F. F- 0F.0- 00. F- 0F.0- N0. F- F F.0- N0. F- NFd- 000::0>o.n.
00.0 «0.0 00.0 00.0 00.0 00.0 00.0- F00- 00.0- 00.0. 0N.0 E00-F
. v0.0 0F.0 . F.F.N 0F.0 00F 0F.0 . 00.0 FF.0 00F 0F.0 .00.. 00000.00 60.6.0
F 900 F 900 F 080 F 900 F 000 00:0E.ot0n. F0000m

 

 

 

 

 

 

 

 

 

n... 000.0>0 0900.03 5.2, 00:0E.0F.00 F0002. .oF 0:000. 8.000.000 F0 0.00-F

89

5.2.2 Schedule performance

Table 5.8 summarizes hypotheses testing results on schedule
performance. Table 1.9 shows the regression results with simple average IID.
The simple average IID model explains 15.30 percent of the variance in schedule
performance (adjusted R-squared = 3.54). The hypothesis variables and their
interactions account for 10.65% of the variance in schedule performance
(difference in adjusted R-squared = 1.89%).

Table 5.10 shows the regression results on schedule performance with
weighted average IID. The weighted average IID model explains 16.15 percent of
the variance in schedule performance (adjusted R-squared = 4.50%). The
hypothesis variables and their interactions account for 11.50% of the variance in
schedule performance (difference in adjusted R-squared = 2.85%).

For both operationalizations of “F, the F-test is not signiﬁcant at the 0.05
level of signiﬁcance, but some of the individual coefﬁcients are signiﬁcant with
evidence supporting some of the hypotheses.

One project characteristic variable, Project selection rigor, is positive and
signiﬁcant. There is support for Hypothesis 1, but only for the effect of
organizational IID in the simple average IID model. Hypotheses 2a and 4a are
not supported. Hypothesis 3a is supported with a positive and signiﬁcant for
functional diversity in the model with weighted average IID. There is no evidence
in support of Hypothesis 2b, but there is support for Hypotheses 3b and 4b in the

model with weighted average IID. The interaction terms temporal distance *

9O

functional diversity and temporal distance * psychological safety both have a

positive and signiﬁcant coefﬁcient.

Table 5.8 Summary of hypotheses testing results for schedule performance
based on a one-tail test at 0.05 level of signiﬁcance

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

e 1, .9
Schedule performance 2 I, 93 I,
Q- 0) 5 c»
E e 9’ e
‘0 0; é’ 9
Hypotheses co co
H1: Direct positive Physical distance
eff?“ 0f Invention-to- Temporal distance
Implementation
distance (IID) Organizational Supported
distance
H2a: Direct positive effect of technical
competence
H3a: Direct posatlve effect of functional Supported
IverSIty
H4a: Direct positive effect of psychological
safety
H2b: Positive Physical distance
moderating. effect 0f Temporal distance
technical
competence Organizational
distance
Physical distance
H3b: Positive
moderating effect of Temporal distance Supported
functional diversity Organizational
distance
Physical distance
H4b: Positive
moderating effect of Temporal distance Supported
psychological safety Organizational
distance

 

 

 

 

 

91

00.0 v 0 $08.05 .- .000 v 0 $0865 .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2.0 2.0 3.0 9.0 :0 mm: 63.

v0.0 v0.0 50 V0.0 «0.0 096309”; .51

m F0 0 F .0 00.0 00.0 00.0 020000-”.

00.0 2.0 0N0 00.0 00.0 n. A 09d

00.0 am... Re 5.0 9.: tax-0 .00“.

mm: mm: 000 000 me So ho .50:52
00.0 mmd 0v.m 00.0 m ﬁm 0081
3.0 «0.0 330.0 2030 . 8:920 90
2.0 No.0 328 £960 . 8:830 082
00. T 3.0- 3900 £960 . 8:906 «Ea
3.0 no.0 E2020 . 8:906 08
mm; N00 36-620 - 8:906 0E9
00.0 00.0 36.620 . 8:906 930
$0 00.0 8020088 :02 . 80920 9o
00.0 V0.0 802an8 :92 .. 8:906 902
5.0 00.0 mucouanoo com: . 8:906 mfg
8.0- 00.0- «0.0 00.0 N00- 00.0- boﬁw _mo_0o_oco>mn_
t; 3.0 50 no.0 mmd v0.0 E9020 .mcozocsu
5.0- 00.0- mm. T 5.0. nv. T m 0.0. 809an8 52:58. m.Emo-_.
0 mm. 0 r .0- t vmd- Fwd- t. mmd- _.N.0- oocﬁwﬁ .mcoszEmQO
00.0- 00.0- 00.0- 00.0- $0. v0.0- 3.090 80020 559:0...
00.0- 8.0- 8.0 00.0 3.0 0.0 $200 8:920 62930
R. F- 2.0. 00. F- :0- 00. F- 2.0- 00. w 9.0- 50. T 20. 3000289000 280 08.90
3.0- «0.0. 00.0 00.0 00.0 00.0 5.0- 3.0- 00.0- 00.0- mmocoéooto 02002008800
00.0- 00.0- S. T 00.0- mm. T 3.0. mm. T 0 0.0- 3. T 3.0. 080099.“.
5. T 3.0- 2. F- 00.0- 3. T 00.0- 00. T 00.0- N0. T 00.0- 050 E00...
m0._- 00.0 v0.0 00.0 9.9 00.0 00.0 0.20 00.0 N_-.0 000: 000028 ﬂop—oi

0 Sem 0 Sum . Sam 0 Swm 0 90m. mocmEcoton. oioocom

 

 

0: 00.0520 9950 53> oocmctotoa 2:00:00 .8 9.53. 028203. 0.0 050.0

92

F00 v 0 09862 .-
000 v 0 80865 .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

and 3.0 3.0 and 3.0 wws. 6cm
00.0 F00 5.0 No.0 No.0 umcmscm-m _0<
0F.0 00.0 00.0 no.0 00.0 vocmscmi
3.0 0N0 0N0 0F.0 0F.0 n. A 09.”.
00. F «N. F R. F we. F 00. F FF-0.-c .0“.
00F 00F 00F 00F 00F 000 Fe 50802
00.0 no.0 03” no.0. 0F.0 mcool
no.0 3.0 3900 5an . 8:900 9o
. mnF nF.0 >6me conma - oocﬂwﬁ 0E9
0.0.0- 00.0- 2300 c003 - oocmEu mnca
mmF 3.0 £0520 . oocﬁmﬁ 0.0
. X: n F.0 £9020 l 8:906 0E9
00.0 3.0 38-520 . 0009.06 993
3.0 no.0 oocouanoo :02 . 80.035 0.5
00.0 00.0 802an8 :09 . mocmomﬁ 9:9
N0.F 00.0 802an8 :09 . 8:905 930
00.0. 5.0. FN.o- No.0- No.0- no.0- boﬁm _mo_mo_oco>mn_
.. 00F FN.0 mno 00.0 00.0 v0.0 360020 .mcozoczm
3.0. 00.0. N0. F- NF.0- nv. T 0 F0. 8:9an8 32:59 0.88..-
00. F- 0F.0- mw. T NF.0- n0. F- 3.0- 8:906 602550090
Fm. F- N 0. 0 F. F- 00.0- 00.0- 00.0- €80 8:920 .9863
0nd. 0F.0- no.0 F0.0 0F.0 F00 32:5 8:920 .0295;
F0. F- NF.0- 00. F- 0F? 00. F. 0F.0- 00. F- NF.0- 0. F- 0F.0- 808239000 280 seen.
3.0 No.0 80 No.0 00.0 00.0 N00- v0.0- 0F.0- F00. 300026900 cozmoEzEEoo
00.0. 00.0- mm. F- 0F.0- mm. F- F F0. 3. T 0 F0. Sq. F- FF.0- mmocco>9a
«N. F- 0F.0- FF. F- 00.0- FF. F- 000- ST 00.0- NF. F- 00.0- 0.20 Ewe..-
. 00.N 0N.0 SF :0 0v.F 0F.0 wn.F 0F.0 om.F NF.0 0oo: 00:00.3 000.85
0 Eom F 900 0 99m 0 900 0 Sam mocmELotom cabocom

 

 

 

a: 00085 09:90; 53> oocmEcotoa 2:00:00 .2 £509 00.002001 oF.m 9an

93

5.2.3 Technical performance

Table 5.11 summarizes hypotheses testing results for technical
performance. Table 5.12 shows the results for technical performance with simple
average IID. The simple average IID model explains 20.44 percent of the
variance in technical performance (adjusted R-squared = 9.39%). The hypothesis
variables and their interactions account for 9.02% of the variance in technical
performance (difference in adjusted R-squared = 0.76%).

Table 5.13 shows technical performance results with weighted average
IID. The weighted average IID model explains 23.98 percent of the variance
(adjusted R-squared = 13.43%). The hypothesis variables and their interactions
explain 12.56% of the variance in technical performance (difference in adjusted
R-squared = 4.80%).

Only one project characteristic variable, project selection rigor, is
signiﬁcantly related to technical performance. The relationship is positive in the
simple average model. There is no support for Hypotheses 1, 2a or 3a in
technical performance. Hypothesis 4a is supported in the model with weighted
average IID.

There is some evidence supporting Hypothesis 2b with both
operationalizations of IID. The interaction term temporal distance * technical
competence is positive and signiﬁcant. Hypotheses 3b and 4b, however, are not
supported by either operationalization. in fact there is evidence opposing

Hypothesis 4b. The coefﬁcient for the interaction term organizational distance *

94

psychological safety is negative and signiﬁcant in both models (two-tail p = 0.015,

0.034).

Table 5.11 Summary of hypotheses testing results for technical performance
based on a one-tail test at 0.05 level of signiﬁcance

 

Technical performance

 

Hypotheses

Simple average IID
Weighted average llD

 

H1: Direct positive Physncal distance
effect of invention-to-
implementation

distance (IID) Organizational
distance
H2a: Direct positive effect of technical
competence
H33: Direct positive effect of functional
diversity
H4a. Direct posntlve effect of psychological Supported
safety

 

Temporal distance

 

 

 

 

 

 

sz; Positive Physical distance

moderating effect of .
technical Temporal distance Supported Supported

competence Organizational
distance

 

 

 

Physical distance
H3b: Positive

moderating effect of Temporal distance
functional diversity

 

 

Organizational
distance

 

Physical distance
H4b: Positive

moderating effect of Temporal distance
psychological safety

 

 

 

 

 

 

 

 

95

5.0 v 0 09020:. t.
00.0 v a 09020:. ,.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00.0 00.0 00.0 00.0 00.0 mws. Foom.
00.0 00.0 00.0 00.0 00.0 09800-1 F0<
0N0 3.0 3.0 NF.0 F F.0 0009000-”.
No.0 5.0 00.0 5.0 00.0 n. A sea.
mm: mud 00.0 00.N 0F.0 3-0.: .00“.
00F 00F 00F 00F 00F 08 0o 08:52
00.0 «0.0 00.0 00.0 00.0 0:81
2. nvN- nn.0- 3900 :90: . 00:90.0 90
00.0 00.0 3900 :30: . 00:90.0 9:9
00.0- 5.0. 30.60 :20: .- 8:90.0 0030
00.0- 00.0. 3.9020 . 00:90.0 90
on. F- 3.0- 0.9020 - 00:90.0 0E9
FF. F- 9.0- 3.9020 . 00:90.0 9::
3.0 no.0 0890an0 :09 . 00:90.0 05
.. N03 0F.0 8:90an0 :09 . 00:90.0 0E9
00.0. 00.0- 8:98:50 :09 - 00:90.0 0?:
n F. F 0F.0 Fv. F 3.0 N0. F 3.0 3900 _8_0o.o:o>0n.
00.0 F00 F00 n0.0 nn.0 00.0 3.9020 .8265“.
F0. F- 00.0- 0 F. F- 00.0- mm. F- 0 F .0- 8:900Eoo .mo_::o9 0.Em0...
F00- moo- 0F.0- 5.0. 00.0- 5.0. 8:900 5:269:85
00.0 00.0 00.0 00.0 00.0 00.0 0.000 00:90.0 .900E0n
00.0- 00.0- 00.0- no.0- 00.0- no.0- .0205 8:900 .8020
00.0 00.0 ~00 00.0 F00 00.0 9.0 00.0 :0 00.0 0882009000 0000 08.05
00.0 3.0 F00 00.0 n00 FF.0 0n.F 0F.0 n0.F 0F.0 008028900 :ozmoEaEEoo
FF. F- 00.0- Nno- 00.0- 0n.0- 00.0- 00.0- 00.0- mn.0- 00.0- 000::0>9n.
0N. F- 0F.0- 0N. T 0F.0- 0N. F- 0F.0- 00. F- 00.0- FF. F- 00.0- 0N.0 E00...
. mmN 0N0 2. 0n.m nn.0 r. 00.0. 8.0 r. nod 3.0 1 mod mmo com... 8.80.00 80.6.5
. 90m. . 90m. 0 90m. . 90m. . 90m. 8:0Ecot0d .8_::00..-

 

 

 

 

 

 

 

 

 

 

n... 0o90>0 0.9:.0 5.; 8:090:00 .00.::o9 :9 9.000.. 8.00900”. NF.m 0.03

96

5.0 v 0 0900.05 2.
00.0 v 0 09020:. ..

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

«0.0 00.0 00.0 00.0 00.0 wws. .oom.

0F.0 00.0 00.0 00.0 00.0 020000-: .3.

00.0 3.0 3.0 NF.0 FF.0 00:0:00i

00.0 5.0 00.0 5.0 00.0 n. A 09.".

RN 00.0 00.0 9.0 0F.0 .F-x: .000

00F 00F 00F 00F 00F 0:0 00 00:52
nn.0 FF.0 00.0 00.0 00.0 0:8-
m F.N- FN.0- 2900 :00: . 00:90.0 0:0
50 00.0 3900 :00: . 00:90.0 0E9
00.0 00.0 2900 :0>00 . 00:90.0 0>:0
FN.0 «0.0 3.0020 . 00:90.0 0.0
0N0 No.0 3.0-.020 . 00:90.0 0E9
00.0- 5.0. 3.0020 . 00:90.0 020
00. F- 0 F0. 8:909:00 :09 .. 00:90.0 0:0
NNN 0N0 8:909:00 :09 .. 00:90.0 0E9
0N0- No.0- 8:909:00 :09 . 00:90.0 0.9.0
00. F N00 00. F 3.0 N... F v F .0 E900 .00.00.0:0>0n.
80 No.0 F00 no.0 nn.0 00.0 3.0020 .mcozocsn.
nm. F- 3.0- 00. F- 00.0- mm. F- 0F.0- 8:909:00 .00.::09 0.E00..-
00.0 00.0 FN.0- No.0- 00.0- 00.0. 00:90.0 .m:o..0~.:00:O
00.0 00.0 F00 00.0 00.0 no.0 .9000 8:80 98:5
3.0- 00.0- n00- 00.0- n00- no.0. .0205 8:000 8.00.0
~00 30 00.0 00.0 F00 «0.0 3.0 00.0 FF.0 00.0 08820800.... 280 00.90
m F.F 0F.0 0nd 00.0 n00 F F0 003 3.0 n0.F 0F.0 000:0>.«00t0 :0..8.:0EE00
9. F- 00.0- nn.0- 00.0- 0n.0- 00.0- F00- 00.0- 0n.0- 000- 000::0>0.n.
00. F- F F.0- 0N. T 0F.0- 0N. F- 0F.0- NF. F- 00.0- FF. F- 00.0- 0N.0 E00..-
~03 0F.0 : 0nd 3.0 .2 00d 80 : nn.0 0N0 : mmd Nmo :00: 00:00.00 .0905

0 90m. 0 90m. 0 90m. 0 90m 0 90m 8:090:00 .00.::00.F

 

 

 

 

 

 

 

n... 0090>0 09:0.03 5.3 8:999:00 .mo.::09 :00 9.000: 5.009001 0F.0 0.00..-

97

5.2.4 Overall performance

Table 5.14 summarizes importance weight the participants assigned to
each type of performance that are used to calculate overall performance. On

average about half of the weight is assigned to technical performance.

Table 5.14 Summary of importance weight assigned to each category of
performance

 

 

 

 

 

 

 

 

 

 

. Std.
Variables Obs Mean Dev. Min IMax
Budget weight (%) 169 22.92 15.35 0.00 80.00
Schedule weight (%) 169 26.92 18.75 0.00 90.00
Technical achievement weight (%) 169 51.46 22.77 10.00 100.00

 

 

Table 5.15 provides the summary of the hypotheses testing results on
overall performance. Table 5.16 provides regression results for overall
performance with simple average IID. The simple average IID model explains
26.60 percent of variance in overall performance (adjusted R-squared = 16.41%).
The main variables and their interactions account for 11.29% of the variance in
overall performance (difference in adjusted R-squared = 3.76%).

The results for overall performance with weighted average IID are
provided in Table 5.17. The variables in the weighted average IID model account
for 28.60% (adjusted R-squared = 18.69%). The main variables and their
interactions account for 13.29% of the variance in overall performance
(difference in adjusted R-squared = 6.04%).

Only one project characteristic variable, team size, is signiﬁcantly related

to overall performance. Its relationship is negative. There is no support for

98

Hypotheses 1, 2a and 3a. Hypothesis 4a for the direct positive effect of
psychological safety is supported in both models of IID.

There is support for Hypothesis 2b in both operationalizations of IID. The
interaction term temporal distance * technical competence has a signiﬁcant
positive coefﬁcient. Hypothesis 3b is supported with the signiﬁcant positive
coefﬁcient of the interaction term temporal distance * functional diversity in the
model with weighted average IID. Lastly, there is also evidence supporting
Hypothesis 4b. The coefﬁcient of the interaction term temporal distance *

psychological safety is positive and signiﬁcant with weighted average IID.

99

Table 5.15 Summary of hypotheses testing results for overall performance based
on a one-tail test at 0.05 level of signiﬁcance

 

Overall performance

 

Simple average IID

Hypotheses

Weighted average IID

 

H1: Direct positive Physical distance

effect of invention-to-
implementation

distance (IID) Organizational
distance

H23: Direct positive effect of technical

competence
H3a: Direct positive effect of functional
diversity

H4a: Direct positive effect of psychological
safety Supported Supported

 

Temporal distance

 

 

 

 

 

 

sz; Positive Physical distance

moderating effect of .
technical Temporal distance Supported Supported

competence Organizational
distance

 

 

 

Physical distance
H3b: Positive

moderating effect of Temporal distance Supported
functional diversity

 

 

Organizational
distance

 

Physical distance
H4b: Positive

moderating effect of Temporal distance Supported
psychological safety

 

 

Organizational
distance

 

 

 

 

 

 

100

5.0 v : 0900.00. 1. .000 v : 0900.00. .-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

000 n00 nn.0 n00 nn.0 wms. .001
0F.0 0F.0 0F.0 0F.0 9.0 020000-: .04.
nN.0 0F.0 to to 0F.0 00.0:00-0
00.0 00.0 00.0 00.0 00.0 n. A 00::
F0.~ 0~.0 F00 00.0 0n.0- .F-x-: .0“.
00F 00F 00F 00F 00F 0:0 00 .9902
«0.0 0n.m 00.0 N3. 0020 0:00I
00. F- NF.0- 3900 :00: . 00:90.0 0.0
nvF 0F.0 3900 :00: - 00:90.0 :E9
0F.0 F00 3900 :00: - 8090.0 0>::
F00 00.0 3.0020 .. 00090.0 000
3.0 00.0 3.9020 . 00090.0 :E9
00.0- no.0- 3.00020 .. 00090.0 0>::
nF.F 0F.0 00:90:E8 :09 . 00090.0 00
t 00.0 N00 8:90:E00 :09 - 00:90.0 :E9
0n.0- 00.0- 80909000 :09 . 00:90.0 0?:
. Fn.F 0F.0 0.0-F 0F.0 NVF 3.0 3900 50.00.0000:
00.0 «0.0 F00 No.0 0F.0 F00 3.0-.020 000.0000“.
nn.0. no.0- 00.0- 00.0- 00. F- 00.0- 80909000 .00.0:09 0.000.:
mm. F- 0F.0- Fn. F- 0 F0. 00. F- NF.0- 8090.0 000.000.0005
00.0- 5.0. to. F00- 5.0 00.0 9000 8090.0 .90:E00-
FF.0 F00 0F.0- F00- 0F.0- F00- 09:0 8:900 020.3:
0F.0- F00- 0F.0- F00- 5.0. 000- F00- ~00- 00.0- 00.0- 0852009000 280 00.9:
0.3 0F.0 N0.F 0 F0 nn.F 0F.0 1. 0nd 0N0 a. no.0 0N0 00000200000 00.000.00EE00
F00- n0.0- 00.0- no.0- 00. F- 00.0- 00. F- no.0- FF. F- 00.0- 0000005.:
. 000- 0 F0. nvN- 0 F0. . 9mm- 0 F0. . Fwd- 0F.0- . nvN- 0 F0. 0N.0 0000.:
nn.F 0F.0 nF.N 0F.0 00F 0F.0 . nod to 003 0F.0 .00.. 00000.00 .0900:
0 000m 0 90m 0. 90m. . 90m. 0 90m. 800E500: ..90>O

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

n... 0090>0 0.90.0 :03 000000000: .905 .00 9.000. 00.00900m. 0F.0 0.00...

101

5.0 v 0 00.00.00. t
00.0 v 0 00.00.00. -

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00.0 00.0 00.0 00.0 00.0 wws. .00”.

0F.0 0F.0 0F.0 0F.0 0F.0 090000-”. .0<

0.0-.0 0F.0 0F.0 0F.0 0F.0 090000-”.

00.0 00.0 00.0 00.0 00.0 n. A 00.0

00.0 8.0 5... FF... 000 3...: .5“.

00F 00F 00F 00F 00F 0:0 .0 00.002
000 000 00.0 0F.0 MN... 00001
NF. F- FF.0- 30.00 000.00 . 8090.0 0.0
00F 0F.0 30.00 000.00 . 0000.0.0 0E0.
00.0 00.0 30.00 :00... . 0000.0.0 0>:0
00F 0F.0 3.0.0>.0 . 0000.0.0 0.0
00F 0F.0 3.0.020 . 0000.0.0 0E9
no.0 Fo.o 3.0.020 . 00090.0 030
N00. vod- 000909000 :09 . 0000.00 0.0
mFN 0F.0 000900500 :00. . 00090.0 909
00.0- 00.0- 000909000 :00. . 00090.0 030
n F.N 0N0 Fn.F m F.o «v. F 0F.0 30.00 00.00.002.00
000 00.0 00.0 00.0 0F.0 F00 3.0.020 002.000“.
00.0- 00.0- 00.0- no.0- 00. F- 00.0- 0000.00E00 200.0000. 0.E00-_-
0F. F- o F.0- 00. F- m F.o- 00. F- 0 F0. 00090.0 .000..0N.000.O
.00. ~00- B0. .00- 00.0- 00.0- .280 800.00 0.80:;
0F.0 00.0 00.0 00.0 00.0 00.0 .0000. 00090.0 .00.0>:n.
0F.0- 8.0- 0nd- No.0- F0...- 000- 00.0- 00.0- 00.0- 00.0- 000002009000 0.000 60.0.0
mmF 0F.0 NNF 0F.0 nn.F 0F.0 mmN Fwd 2. ~00 mNd 000002.0000 00..00.0:EE00
mm...- nod- mo. F- mod- mo. F- mod- 00. F- mod- F F. F- 00.0. 000000>0.n.
00.0- 0F.0- 3.0- 0F.0- 00.0- 0F.0- 00.0- 0F.0- . tim- 0F.o- 0~.0 E00.-
mm. F 0F.0 NNN owd mm. F 0F.0 NNN m F.o 8. F m F.o .00.. 00:00.00 .00..0.n.

. 0.0m . 0.0m . 0.0m . 0.0m . 0.0m. 0000000000 _.0.0>O

 

 

n... 0090>0 09:902. :..>> 0000000000 .005 .0. 0.300. 00.00900”. n F.m 0.00-P

102

Chapter 6 Discussion and conclusion

This chapter concludes the study with the discussion of the ﬁndings and
theoretical contributions, managerial implications, and ends with limitations and

suggestions for future research.

6.1 Findings and theoretical contributions

Control variables

Five control variables were included in the regression models. There is
evidence that project selection rigor has a positive relationship with budget,
schedule and technical performance, but not with overall performance (the
relationship is positive with two-tail p =0.075 in both models). This means that a
technology/practice that is chosen based on information from past experiences
and existing user environment tends to require less resources and time and
achieve better technical performance.

Team size has a negative effect on overall performance, but it does not
have signiﬁcant effect on any of the individual performance score. The results
conﬁrm ﬁndings in the management literature prior to 1985 (Gooding and
Wagner Ill 1985) that as team size grows, efﬁciency-based group performance,
the focus of this study, tend to be lower. This is explained as the effect of higher
social distraction, greater coordination requirements, more behavioral masking
and diffusion of responsibility (Wagner Ill and Hollenbeck 1992). The results
contradict ﬁndings in a recent meta-analytic review on project team size (Stewart

2006). However, Stewart (2006) does not distinguish between different types of

103

performance measure (output vs. efﬁciency) which may bias the ﬁnding. The
explanation given for the more recent ﬁnding is that larger teams/organizations
are more able to obtain resources such as personnel and money.

Goal aggressiveness is negatively associated with budget performance,
but not associated with other performance. The fact that aggressiveness is not
associated with schedule and technical performance is interesting, because it
suggests that schedule and technical achievement are more difﬁcult to predict,
and that initial goals’ aggressiveness do not explain much of the variance in
performance score.

Provenness of technology and intra-team communication effectiveness
are not related with any of the performance measures. The lack of signiﬁcance
for the relationship between provenness and implementation performance is
surprising, since a proven technology’s effects on an organization are more
predictable (Leonard-Barton 1988; Edmondson et al. 2003). Another interesting
observation is that provenness is positively but not signiﬁcantly associated with
temporal distance. The lack of signiﬁcant relationships between communication
effectiveness and implementation performance may be explained by the high
correlation (r=0.66) between communication effectiveness and psychological

safety.

lnvention-to-implementation distance (IID)

This study extends the literature by explicitly measuring and testing the IID

construct. Previous studies on this topic have used this variable as a dichotomy

104

of internal and external to the organization. This study recognized that more than
one party could contribute to the design of a manufacturing process innovation
(MPI), and that IID is composed of three distance dimensions. This study
accounted for those MPls that were collaboratively invented and measured the
dimensions of MD accordingly. It also tested the hypotheses of direct effect of MD
on MPI implementation project performance.

The results of this study imply that organizational distance is the only
aspect of IID that has a signiﬁcant negative effect on implementation
performance, and this is true for budget and schedule performance only. The
results suggest that physical distance between an MPI source and the
implementation site does not introduce any barrier in implementing an MPI.
Temporal distance does not have a signiﬁcant direct relationship on performance
either, but it has a moderating effect which will be discussed further.

The ﬁndings on organizational distance are contrary to what would be
suggested by prior ﬁndings (Menon and Pfeffer 2003, and Menon et al.2006).
Menon and coauthors ﬁnd that organizations are likely to resist internal
innovations and welcome external innovations for fear of internal competitions.
Szulanski’s (2004) results indicate that although cultural difference between the
invention source and the recipient results in lower recipient’s motivation to
implement the invention, it has 3 overall positive effect on knowledge transfer
process ease. The ﬁndings of this study might indicate that how an organization
values a set of knowledge (which feeds into motivation) and cultural difference

between the source and the recipient organizations may not be as signiﬁcant

105

factors in determining the success of a manufacturing process implementation
project.

The results of the present study on organizational distance conform with
Fedor et al.’s (2003) ﬁnding. Fedor et al. (2003) ﬁnd that internal knowledge
generation has a positive impact on integrated product and process development
project performance, while external knowledge generation has no signiﬁcant
effect. The ﬁndings indicate that the origin of knowledge matters, and knowledge
from a more distant source has lower beneﬁt on project performance. Fedor et
al.’s (2003) ﬁnding implicitly suggests that distance has a negative effect on

project performance.

Team members” technical competence

The study provides an empirical test of the application of the absorptive
capacity framework in the context of manufacturing process implementations.
Technical competence of the team members, measured as perception of team
members’ skill and expertise, is found to be positively associated with budget
performance, but no other type of performance. Technical competence of
implementation project team members is expected to be positively related with
the team’s absorptive capacity which is critical in the implementation activities
(Cohen and Levinthal 1990; Zahra and George 2002).

It is surprising that technical competence is positively related to only one
of the four performance scales. Subramaniam and Youndt’s (2003) ﬁnding

provides an explanation for the lack of support in three of the four performance

106

scales in the present study. The authors ﬁnd that technical competence alone
can negatively affect an organization’s ability to radically innovate, an absorptive
capacity. In addition, organization members’ technical competence needs to be

coupled with social interaction among the members to be beneﬁcial.

Functional diversity

The results indicate a signiﬁcant positive relationship between functional
diversity and schedule performance but no signiﬁcant relationship with other
performances. Functional diversity was expected to have a positive effect on
implementation project performance because more perspectives allows the team
to anticipate, detect and solve more potential problems. This ﬁnding supports the
evidence found in the project management literature (e.g., Sen and Rubenstein
1989 and Pinto et al. 1993).

Similarly to the ﬁnding on technical competence, it is surprising that
functional diversity is positively related to only one performance scale, schedule
performance. A possible explanation is that the effect of functional diversity is not
linear. The absorptive capacity literature (Cohen and Levinthal 1990) suggests
that diversity has a positive relationship with an organization’s absorptive
capacity, but the positive effect diminishes and may even be negative as diversity

makes communication among team members difﬁcult.

Psychological safety

107

The results of this study provide additional support for Edmondson et al.’s
(2001) ﬁndings on technical performance, as well as extends the testing on
project budget and schedule performance. This study uses a relatively large
sample size (N = 51 for Edmondson and co-authors and N = 47 for Baer and
Freese (2003)), thus, provides greater statistical power in testing the theory.

The results of this study show that psychological safety is not signiﬁcantly
related to budget and schedule performance but has a positive relationship with
technical and overall performance. Psychological safety has been shown to
enhance the team’s ability to learn, and thus was expected to be positively
associated with implementation performance.

Team psychological safety is argued to result in greater exercising of
effective team learning behaviors. The effect of these behaviors may be more
direct on technical performance as technical performance is directly driven by the
“quality" of the learning, solutions to problem and newly created procedures to
accompany the implemented practice/technology. Effective learning behaviors
should also be beneﬁcial to lower the implementation cost and duration.
However, budget and schedule performance may be more highly inﬂuenced by

other factors such as the relative importance of each of the goals in the project.

Moderating effect of technical competence
There is a positive interaction effect between technical competence and
temporal distance on budget, technical and overall performance. However, the

main effect of temporal distance is not signiﬁcant in any of the models and the

108

main effect of technical competence is only signiﬁcant in the budget performance
model. Thus, the results should be reinterpreted (Sharma et al. 1981; Baron and
Kenny 1986). The results imply that when an MPI is from a temporally distant
source, team members’ technical competence actually has a greater positive
effect on budget performance. In technical and overall performance the
interrelationships between technical competence and temporal distance are not
clear since neither main effect is signiﬁcant.

The results suggest that absorptive capacity is more valuable at a high
level of IID where there tends to be greater requirements to learn and solve
problems. (Another possible interpretation, which is not as plausible, is that at

high level of MD, technical competence creates greater absorptive capacity.)

Moderating effect of functional diversity

The coefﬁcient for the interaction term between temporal distance and
functional diversity is positive and signiﬁcant in budget, schedule and overall
performance. The interrelationships between functional diversity and temporal
distance on budget and overall performance is not clear, since the main effects of
both variables are not signiﬁcant in either of the models. In the schedule
performance model, the main effect of functional diversity is positive and
signiﬁcant while the main effect of temporal distance is not. The result, thus,
indicates that temporal distance is a positive moderator of the relationship

between functional diversity and schedule performance. The longer an invention

109

 

has been designed, the higher the beneﬁt of functional diversity will be on
ﬁnishing the MPI implementation process on schedule.

The overall message is that functional diversity, and thus absorptive
capacity, has higher value in MPI implementation projects where designs are
older. Since there is greater requirement for learning and problem solving for an
MPI with a more temporally distant source, this means that functional diversity is

more valuable when there is greater learning and problem requirement.

Moderating effect of psychological safety

Only one of interaction term between psychological safety and IID
variable, psychological safety*temporal distance, has a signiﬁcant and positive
coefficient in the overall performance model. The result suggests that
psychological safety is more valuable with higher temporal ”D.

However, the coefﬁcient for the interaction between psychological safety
and organizational distance is negative and signiﬁcant in the technical
performance model. This suggests that psychological safety is less valuable
when organizational IID is high. The result is puzzling and there is no plausible
theoretical argument at this point. Replication of this study in the same and

different context will be worthwhile.

Comparisons between the two operationalizations of II D

In this study two operationalizations of IID were used. One was a simple

average distance, and the other is weighted average distance. The simple

110

average distance has an advantage that it is free of further assumption of how
each distance should be weighted. The weighted average distance assumes that
difﬁculty arising from distance for each source is proportionate to the contribution
of the source.

The results in all four categories of performance show that weighted
average llD models explain more variance in performance than respective simple
average IID models. This is an evidence that difﬁculty of distance that the MPI
implementing unit faces from each source is somewhat proportionate to the
source’s contribution to the project. A recommendation for future studies on the

measurement of this variable is discussed in the last section of the chapter.

6.2 Managerial implications

The results of this study should aid managers in outsourcing decisions as
well as managing the implementation process. First, managers must increase
awareness of the potential negative effect when making outsourcing/off-shoring
decisions along with other effects that have been documented (increasing
dependence, moving away from competency in the technological invention). The
results of this study show that increasing organizational distance between an MPI
source and the implementing organization results in worse budget performance.

Second, stafﬁng and team characteristics have signiﬁcant effects on
whether a project will achieve its goals. Technical competence and functional
diversity of the implementation team members alone may not signiﬁcantly result

in better project performance. The results of this study combined with past

111

research may suggest that their effect is contingent on task uncertainty and level
of communication among team members. Moreover, their beneﬁts are contingent
on the level of temporal IID. At high level of temporal llD, technical competence
and functional diversity are more positively associated with implementation
performance.

Lastly, psychological safety among team members has positive effect on
technical and overall performance. While it may be difﬁcult to encourage, it can
be fostered by selection of leaders and team members that may exhibit certain
characteristics. These include people who encourage others to take risk, and
people that realize that making mistakes is sometimes necessary for learning

during an implementation process.

6.3 Directions for future research

Studies on innovation implementation are still relatively scarce. Large-
scale empirical studies should be conducted in different project contexts to
illustrate the robustness of the ﬁndings here. In addition, there are certain
limitations for the current study that future studies should address. Two speciﬁc
theoretical issues and three methodological issues are discussed.

First, a major limitation for this study is that the models explain relatively
low level of variance in the dependent variables. This suggests that some key
variables and interactions relationships may be missing. This is especially true

with interactions between issues from operations management and organization

112

behaviors as noted by Repenning and Sterman (2002). The following lists
speciﬁc issues that should be included in future studies.

- Resource availability. Many project management studies have
found top management commitment to the project as a success
factor. Top management commitment is likely to be a proxy for
resource availability. Including some measure of resource
availability should enhance the explanation power of the model as
well as explicitly control for variance in performance due to
resource availability.

- Contingency perspective with respect to degree of change
required. The works by Tyre and Hauptman (1992), Shenhar et al.
(2002) and Stock and Tatikonda (Forthcoming in JOM) suggest
that the value of variables such as technical competence and
functional diversity may be contingent on degree of radicalness.
The inclusion of interactions with radicalness should enhance the
understanding of the relationships among the variables under
investigation here.

- Contingency perspective with respect to communication among
members. Subramaniam and Youndt’s (2005) ﬁndings suggest
that the value of organization members’ skills and expertise is
contingent on the level of communication among themselves. This

suggests that the interactions between technical competence,

113

psychological and communication effectiveness may have
significant effect on project performance.

- Interactions among the preconditions of absorptive capacity.
Preconditions for absorptive capacity such as breadth and depth
of knowledge and linkage among organizational members (Cohen
and Levinthal 1990; Zahra and George 2002; Smith et al. 2005)
may have reinforcing effects with one another. Future research
should investigate the interactions among these variables and
their effect on absorptive capacity.

Second, the hypotheses of this study are based on the argument that “D
results in lower ﬁt (culturally and physically) between existing work environment
and the MPI. The lack of physical and cultural ﬁt results in greater required
change that induces greater learning requirement, and possibly resistance
behaviors (caused by lower motivation due to preference for stability). This study
did not include these mediating variables. Explicitly measuring and integrating
these variables into the framework will enhance the understanding of the
dynamics of the implementation process.

The ﬁrst methodological issue is that this study measures performance as
goal achievement and control for goal aggressiveness. Both of these are
perceptual variables that may not accurately reﬂect the true level of
accomplishment of a project. However, this was necessary to make performance

ratings comparable across projects. Future study should explore other

114

operationalizations of performance that can circumvent the incomparability of
performance across projects.

Secondly, the measurement of MD in this study assumes that difﬁculty
arising from distance for each source is proportionate to the contribution of the
source. An alternative way to measure the weight, that removes the necessity of
this assumption, is to directly measure the level of communication to each of the
sources. This should be further considered for future studies.

Finally, this study employed a cross-sectional data that captures the
information after the implementation had been completed. Reported information
on the implementation process may have been distorted by the respondent’s
awareness of the project’s performance. Longitudinal study that captures
information on each project at multiple points in time will eliminate the hindsight

effect.

115

Appendix 1 Final round Q-sort

Table A1 Instructions for Q-sort

First, please read each of the 8 factor deﬁnitions below. 34 statements related to these
factors are listed to the right of this cell. Please place a "1" in the row below each
statement that indicates the factor to which that statement is most closely related.

 

Invention innovativeness: The degree to which the original design of the manufacturing
practice/technology that we adopted was new to the world.

 

Implementation change required The degree to which our systems, skills, procedures,
or equipment had to be changed in order to implement the new practice/technology.

 

Provenness of technology: The degree to which the new manufacturing practice/
technology was well understood and free from malfunctions before we adopted it.

 

Project selection rigor: When we ﬁrst selected the practice/technology, the degree to
which we considered its ﬁtness and feasibility for our situation.

 

Project team technical competence: The combined quality and depth of useful
knowledge that was represented by persons on the implementation team.

 

Project team diversity: The extent to which multiple functions were represented on the
implementation team.

 

Effectiveness of intra-team communication: The usefulness of communications among
implementation team members.

 

Psychological safety: The shared belief among implementation team members that it
was safe to take personal risks.

 

Cost performance: The degree to which expenses for the implementation project
adhered to the initial planned budget of the project

 

Technical goals achievement: The degree to which the performance of the new
practice/technology matched the original operational improvement targets of the project
(e.g., improving production quality, time or cost)

 

On-time performance: The degree to which the timing and completion of the
implementation project adhered to the initial planned project schedule

 

 

 

116

Table A2 Final Round Q-sort Correct Placement Ratio

 

%

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Construct Questionnaire item Correctly
soned
PeOple on the implementation team talked openly and freely. 91.89
Team members shared information in useful formats. 89.19
Effectiveness 0f Implementation team members did not seem to understand 94 59
infra-team what one another was saying during their discussions. (R) '
communication . _
Team members often shared valuable information. 78.38
When implementation team members shared information it 67 5.,
was likely to be used. '
It was safe to take a risk on this team. 91.89
If an implementation team member made a mistake, it was 64 86
held against them. (R) '
People on the team were often afraid to ask other members 24 32
for help. (R) '
. Team members feared that bringing up problems/issues
PSY:2?;?€'C3' would have negative effects on their career prospects. (R) 70°27
Competition among team members sometimes suppressed 37 84
individual contributions to the team. (R) '
People on this team sometimes rejected other members. (R) 43.24
Each of the members of the implementation team felt that 48 65
they were valuable to the team. '
The actual implementation cost was signiﬁcantly more than 100 00
the original budget. (R) '
Cost performance Signiﬁcantly more capital expenditure than originally planned 10000
was used In the Implementation. (R)
Signiﬁcantly more manpower was used in the
. . . . 48.65
Implementation than originally planned. (R)
The new practice/technology performed signiﬁcantly better 94 59
than we originally expected. '
The new practice/technology improved our responsiveness 86 49
more than we had hoped. '
The new practice/technology improved the quality of our ' 97 30
, process more than we had originally expected it would. '
Technical goals
achievement The new practice/technology was not as effective or efficient 89 19
as we had originally hoped it would be. (R) '
Using the new practice/technology, we achieved greater
. . . . . 97.30
productlvmes than we originally planned.
Compared to our original expectations, the new
practice/technology achieved signiﬁcantly better overall 97.30

 

 

 

 

 

technical improvements.

 

 

on-time
performance

 

The implementation actually took signiﬁcantly more time than
originally planned. (R)

 

100.00

 

117

 

Table A2 (continued)

 

°/o

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Construct Questionnaire item Correctly
soned
In order to use the new practice/technology, the 75 68
operators/users had to learn new skills and procedures. '
Implementation Much training was. requnred for operators/users in order to 78.38
. use the new practice/technology.
radicalness
The implementation of this practice/technology required
significant changes in our assets such as tools and 91.89
equipment.
This practice/technology had been implemented successfully 56 76
in other organizations before we adopted it. '
Irregularities and problems with this technology/practice had 78 38
been resolved before we adopted it. '
Provenness Of Before we started the implementation process, there were
technology known unresolved problems with the practice/technology 75.68
itself. (R)
Before we implemented this technology/practice, its
effectiveness had been demonstrated by successful 72.97
adoptions in many other organizations.
Before choosing the practice/technology, we compared the
resources that were required (e.g., tools and equipment and 54.05
operators' skills) and what we had.
Before choosing this practice/technology, we followed 78 38
rigorous technology selection guideline. '
. . In choosing this technology, we used detailed information on
PrOJecrtigsjection the existing user environment, product and production 54.05
system.
In choosing this practice/technology, we used information
from past experiences in implementing new manufacturing 27.03
processes.
Before choosing this practice/technology, we did an initial 75 68
i feasibility study. '
Implementation team members were highly skilled. 86.49
Each individual on the implementation team had useful 67 57
Project team experience. '
technical implementation team members were experts in their 56 76
competence particular jobs and functions. '
implementation team members were considered among the 91 89
best people in the organization. '
Many functional areas were represented on the 86 49
implementation team. '
Project team The implementation team included people from different 94 59
diversity functional groups. '
The implementation team included people from diverse 94 59

 

 

 

 

functional areas.

 

 

 

118

1.1..

 

Appendix 2 Survey Questionnaire

 

MSU Manufacturing Process Implementation Study

At any point in the survey, you can go back to change your previous responses
by clicking "Previous Page". All responses are submitted to us only when you
click "Submit Survey" on the last page of the survey.

If you have to leave this page at any point, as long as you use the same
computer, when you access the survey again you can go back directly to the part
where you have left off.

1) Please enter the pass code contained in the invitation letter or e-mail. If you do
not have a pass code, please enter "Guest".

r

2) Please ﬁll in your name and e-mail address below, so that we can send you an
electronic summary of the best practices uncovered by this research. (We also
need this information to remove you from future invitation mailings.)

 

 

 

Name: [
E-mail: [

 

3) Please select the charity that you would like us to donate $5 to.

Habitat for humanity
National Childhood Cancer Foundation

Red cross
UNICEF

3'33”)

4) Have you been directly involved in a manufacturing process implementation
project (e.g., implementing new routines/procedures and tools, equipment or
machinery) that was completed in the last 3-9 months, and are you willing to
participate in this survey?

Yes --- Please skip to question 7.

No --- Please answer questions 5 and 6, and click "Next Page" to

119

submit this information and close the survey.

5) Could you refer us to someone in your organization who may have been
involved in a manufacturing process implementation project that was completed
in the last 3-9 months?

 

Name: I

 

E-mail: I

 

Name: I

 

E-mail: I

6) May we know the reason you are choosing not to participate?

r‘ l have not been involved in implementing any new manufacturing
process.
I—

l have not been involved in implementing any new manufacturing
process that was completed in the last 3-9 months.
I.

l—

I do not have time to complete the survey.

Other (please specify)

If you selected other, please specify:

1

7) Which of the following titles best describes your position in the organization?
(Please select all that apply.)

 

Upper Management

Middle Management

First Level Management

Engineer or Other Trained Professional
Consultant/Contract Employee

Researcher

7777777

Other (please specify)

If you selected other, please specify:

 

120

8) Which of the following best describes the functional area in which you work?

7

Marketing/Sales
Development/Engineering
MlS/lnformation Technology

F inance/Accounting
Manufacturing/Operations/Maintenance

Personnel/Human resources

”377777

Other (please specify) 1

If you selected other, please specify: r
| v:
9) What is the name of your company?

I

10) Approximately how many people are employed globally in your company?

 

 

Between 100-500 people
Between 501-1,000 people
Between 1,001-5,000 people
More than 5,000 people

77'3”)

11) Think about all of the manufacturing process innovations and improvements
your company has implemented over the past 5 years. What percentage of these
innovations was: (Answers must add up to 100%)

 

designed/invented mostly by parties outside your ﬁrm? (%) I
designed/invented mostly by internal R&D? (%) l
designed/invented mostly as a collaboration between external l
and internal parties? (%)

12) Over the past 5 years, approximately what percentage of manufacturing
process innovation projects undertaken by your company was: (Answers muct
add up to 100%)

incremental changes which build upon existing l

121

practice/technology? (%)

completely new to the ﬁrm and replacing existing l
practice/technology? (%)

 

13) The rest of the questions in this survey will ask you about the most recent
manufacturing process innovation that you were directly involved with.

Please keep the following deﬁnitions in mind. Manufacturing process innovation
refers to all activities required to invent and implement a new manufacturing
process. Our deﬁnition of a new manufacturing process innovation includes
everything from a minor change such as a change in procedures to a major
change such as installation of new equipment.

Design/Invention phase refers to the design process in which requirements (e.g.,
technology, particular skills and materials) and technical speciﬁcations (e.g.,
processing speed, quality) of a new manufacturing practice or technology are
determined. This may occur somewhere else outside your organization. We will
refer to an invention outcome as a new "practice/technology".

Implementation Phase is the process in which the new practice/technology is
modiﬁed and installed. This could include making changes to the user
organization as well as the practice/technology itself.

Please check below to indicate that you have read and understand the above
deﬁnitions and instruction.

I understand the above deﬁnitions and instruction.

14) In 1-2 sentences please brieﬂy describe the last manufacturing
practice/technology implementation project completed between 3—9 months, in
which you were directly involved.

3

15) How long ago was this project completed (weeks)?

{—

16) Which of the following types of operations best describes the process that
was affected by the implementation of the new practice/technology you referred
to?

 

 

Job shop (Production of small volumes of a large number of different
products)

122

r Batch (Production of large volumes of similar items on a repeat
basis)

Assembly line (Production of discrete parts moving from workstation
to workstation at a controlled rate)

Continuous ﬂow (Processing of undifferentiated materials such as
petroleum or chemicals)

17) What were the primary reasons the new manufacturing practice/technology
was implemented? (Check all that apply)

A new product launch

A product modiﬁcation
Kaizen/continuous improvement event
Parent corporation’s requirement
Customer's demand

Vendor's suggestion

7777777

Other (please specify)

If you selected other, please specify:

I

18) How many people are employed at the manufacturing plant where the
practice/technology was implemented?

I_—

19) What was the scope of operations affected by this implementation?

 

A single work center
A production line

A department

The entire plant

7777')

Multiple plants
20) Compared to what was originally planned:
Significantly W

AS Signiﬁcantly

   

123

How much man
power was actually
used in the
planning, testing
and installation of
the new
practice/technology?

How much capital
expenditure was
actually used in the
implementation?

How much time did
the implementation
actually take?

 
 

more than originally lesszthan
originally planned $11333
plaqned _-3.. . . ,,_..5...

E; ,. 5.
.‘ ,
. .

r I ”£1,
ELF ‘
-

r r

r.
F
-

 

21) Please rate the extent to which you agree or disagree with the following
statements regarding achievement of technical targets.

Using the new
practice/technology, we
achieved greater
productivities than we
originally planned.

The new
practice/technology
improved the quality of
our process more than
we had originally
expected it would.

The new
practice/technology
improved our
responsiveness more
than we had hoped.

The new
practice/technology

    
 
 

Strongly
Agree

_-5- -

F I

Strongly
Disagree

_._1-..

f“ F

r. F
(n. F
1’"

 

(.-

124

achieved signiﬁcantly
better overall technical
improvements than we
had originally expected it
would.

The new

practice/technology

performed signiﬁcantly .2
better than we had

originally expected it

would.

The new
practice/technology was
not as effective or
efﬁcient as we had
originally hoped it would
be.

 

22) To what extent was the original budget/projected cost of the implementation

project met?

Actual cost was signiﬁcantly more than the original budget.
Actual cost was slightly more than the original budget.
Actual cost was as originally budgeted.

Actual cost was slightly less than the original budget.

77777

Actual cost was signiﬁcantly less than the original budget.

23) Relative to other similar implementation projects undertaken in this company,

this project took:

Signiﬁcantly less time
Slightly less time
Same amount of time
Slightly more time
Signiﬁcantly more time

777777

We have not undertaken any similar implementation projects.

24) How satisﬁed were top managers with how fast the implementation project

took place?

125

Very satisﬁed
Somewhat satisﬁed
Neither satisﬁed nor dissatisﬁed

Somewhat dissatisﬁed

77777

Very dissatisﬁed

25) In evaluating the performance of this implementation project, what
percentage weight (or importance) was given to each criterion of success?
(Answers must add up to 100%)

Technical goals achievementI

Budget attainment I

Schedule adherence I

26) Please give us your estimate of the percentage return on investment (ROI)
and the payback period of the implementation project?

ROI (%): I——_
Payback (years): I

27) Is the implemented practice/technology still being used?

Yes
No (please specify why in the comments ﬁeld)

Additional comments:

I

28) At its end, to what degree (%) was the implementation project over- or under-
budget? ((Actual cost/ budget cost -1)*100)

 

Percent below initial budget (%): I
Percent above initial budget (%): I

29) How long (weeks) was the implementation project initially intended to take?

[_—

126

30) How long (weeks) did it actually take?

'—

31) Questions on this page will ask you about the design of the adopted
manufacturing practice/technology, how it was selected and where its
design/invention took place.

As a reminder, the design/invention phase refers to the design process in which
requirements (e.g., technology, particular skills and materials) and technical
speciﬁcations (e.g., processing speed, quality) of a new manufacturing practice
or technology are determined. This may occur somewhere else outside your
organization. We will refer to an invention outcome as a new
"practice/technology".

Please rate the extent to which you agree or disagree with the following
statements regarding the design/invention of the practice/technology that was

adopted.
Neutral
_3..

Strongly
Agree

Strongly
Disagree

_1...

The design of this
practice/technology was new to the (7
world.

The new practice/technology was
simply a minor modiﬁcation of 1"
industry best practice/ technology.

  
 

The new practice/technology made
existing industry best f“
practice/technology obsolete.

32) Please rate the extent to which you agree or disagree with the following

statements regarding the provenness of the new practice/technology:

Strongly
Agree
_-_5___

I F
F F

Strongly
Disagree

Neutral

Before we implemented this
practice/technology, its effectiveness

had been demonstrated by f“
successful adoptions in many other
organizations.

(7

Before we started the

implementation process, there were (4
known unresolved problems with the
practice/technology itself.

127

This practice/technology had been
implemented successfully in other 1’"
organizations before we adopted it.

Irregularities of this
practice/technology had been f‘
resolved before we adopted it.

p) .

EH
'5 7
II
‘ ”a

33) Please rate the extent to which you agree or disagree with the following
statements regarding the selection of the adopted practice/technology:

Strongly
Agree
_-5_

Strongly
Disagree

...1-._

Neutral
-3-

Before choosing this
practice/technology, we
compared the resources
that were required (e.g., 1"
tools, equipment,
operators' skills) to what
we had.

.,
I
‘3 .

Before choosing this
practice/technology, we r“
did a thorough feasibility
study.

7

Before choosing this
practice/technology, we
followed rigorous «"‘
technology selection
guidelines.

7

In choosing this practice/
technology, we used
detailed information on
the existing user
environment, product and
production system.

In choosing this
practice/technology, we
used information from
past experiences in f"
implementing new
manufacturing
processes.

p—A
N

’7
OO
3 -.
7 ' " ~ ‘

34) Please rate the extent to which you agree or disagree with the following
statements regarding the changes required in this implementation project:

Strongly
Disagree

---1 --_

The implementation of
this practice/technology
required signiﬁcant F r
changes in the our -
assets such as tools and
equipment.

In order to use the new
practice/technology, the
operators/users had to t“ .
learn new skills and
procedures.

Much training was
required for
operators/users in order ("
to use the new
practice/technology.

 

35) What percentage of the design/invention of the new manufacturing
practice/technology you described was done by: (Answers must add up to 100%)

a.) Personnel (managers, engineers and workers) at the [_—
implementation site? (%)

b.) Corporate support functions (engineers and others not located I—_—
at the site)? (%)

0.) External sources (vendors, suppliers, university labs, etc.)? (%) I—

36) If your answer to 35 c.) is more than zero, please answer this question.
Othewvise, please proceed to the next question.

Think of the most important external sources of the design/invention. For up to 3
of these, please approximate the percentage of the design that each contributed.
If there were less than 3, leave the extra ﬁelds below blank. (The sum of these
contributions cannot exceed your answer for 35 c.)

What percentage of the total design/invention did External source I

1 contribute? (%)

What percentage of the total design/invention did External source I

129

2 contribute? (%)

What percentage of the total design/invention did External source I
3 contribute? (%)

37) If corporate support functions or external sources were involved in the
design/invention, please answer this and the next question. Otherwise, please
skip to question 39.

Please indicate the item that best describes the working relationship that
personnel at the implementation site had with the following sources.

EXt’eme'Y Somewhat 0:12;
close close
___1_..._ ___3___ at all
--5--
Relationship with
contributing corporate r" r p
support functions
Relationship with

external source 1

Relationship with
external source 2

Relationship with -.
external source 3

7
.3
7

38) How many miles away was each design/invention source from the
implementation site? (Enter NA if not applicable.)

Contributing corporate support functions (miles):

External source 1 (miles): I
External source 2 (miles): I
External source 3 (miles): I

39) Approximately how many years before implementation did each source make
their contribution to the design/invention? (Enter NA if not applicable.)

Personnel at the implementation site (years): I
Corporate support functions (years): I
External source1 (years): I

 

External source 2 (years):

External source 3 (years): I

40) Questions on this page will ask you about the implementation team members
of the manufacturing process innovation which you referred to in the previous
page of the survey.

As a reminder, the implementation phase is the process in which the
pratice/technology is modiﬁed and installed. This could include making changes
to the user organization as well as the practice or technology itself.

Which of the following titles best describe(s) your role on the implementation
project? (Select all that apply.)

'— Project manager

Functional manager

I
I '

Project team member

I.
[_

'— Project sponsor/champion
I.

Other (please specify)

If you selected other, please specify:

41) Approximately what percentage of the personnel on the implementation
team: (Answers must add up to 100%)

 

had a high school degree? (%) I
had a 2-year associate degree? (%) I
had an undergraduate degree? (%) I

had a graduate degree? (%) I

 

42) Approximately what percentage of the personnel on the implementation
team: (Answers must add up to 100%)

had less than 5 years of work experience? (%) I
had between 5-15 years of work experience? (%) I '
had more than 15 years of work experience? (%) I

43) Please rate the extent to which you agree or disagree with the following
statements regarding implementation team members.

131

Strongly Strongly
Disagree Nigttal Agree
--_1_.. ...5-._
Implementation team
members were highly r (‘ r
skilled

The implementation
team included people
from diverse functional
areas.

7

Implementation team
members were

considered among the f”
best people in the
organization.

Many functional areas
were represented on the C
implementation team

Implementation team
members were experts
in their particularjobs
and functions.

7

EEEEEEEI
EEEEEEEI
7

Each individual on the
implementation team f“
had useful experience.

The implementation
team included people
from different functional
groups.

44) What percentage of the personnel on the implementation team was from the
following functional area? (Answers must add up to 100%)
Manufacturing engineering (%):
Product design engineering (%):
Quality control (%):
Production control (%):

11111

Maintenance (%):

1

b)
N

Production/Manufacturing (%):
Purchasing (%):

Management information system (%):
Others (%):

1111

45) What percentage of the personnel on the implementation team worked at the
implementation site before the implementation project began? (%)

I——

46) Please rate the extent to which you agree or disagree with the following
statements which describe the implementation team.

Strongly Strongly
Disagree Negtral Agree
_-1—- 7' f v -5—
If an implementation
team member made a
mistake on this team, it P f“ C
was held against
him/her.

Team members feared

that bringing up
problem/issues would C
have negative effects on

their career prospects.

People on this team
sometimes rejected f“
other members.

It was safe to take a risk

_ r r r
on this team.
No one on this team
would deliberately act in (z r“ r“

a way that undermined
my efforts.

Each member of the
implementation team felt
they were valuable to
the team.

 

U)
EEHEEEI
. a
EEIEEEI
7

1

U.)

47) Please rate the extent to which you agree or disagree with the following
statements describing how the implementation team felt about the project goals
at the beginning of the implementation project:

Strongly Strongly
Disagree Negtral Agree
---1-— __ — --5-—-
At the start of the
project, we felt the
original project budget F F r
was overly optimistic.
At the start of the
project, we felt the
original project F t“ r“
schedule was overly
optimistic .
At the start of the
project, we felt the
original project 6‘ - F . F
technical goals were .
overly optimistic. ' -
At the start of the
project, we were
conﬁdent that we could F F r“
achieve the project
budget
At the start of the
project, we were
conﬁdent that we could F t‘ r‘
achieve the project
schedule.
At the start of the
project, we were
conﬁdent that we could f” F F
achieve the project's
technical goals.

48) Please rate the extent to which you agree or disagree with the following
statements about communication among the implementation team members:

Strongly Neutral IR:”I2% Strongly

 

134

Agree
-_.5-..

Disagree

Implementation team
members talked openly f“
and freely.

Implementation team

members did not seem

to understand what one i“
another was saying

during their discussions.

When implementation

team members shared r
information it was likely

to be used.

Team members shared
information in useful f“
formats.

Team members often
shared valuable f“
information.

 

49) At the point in the implementation project when the number of staff was at its
peak, approximately how many persons were:

Assigned full-time to the implementation project? I
Assigned on a part-time basis only to the implementation project?I

50) Can you refer us to other members of the implementation team that may be
able to answer questions in this survey?

 

Name: I

 

E-mail: I

 

Name: I

 

E-mail: I

BIBLIOGRAPHY

Abernathy, W. J. and M. Utterback (1978). "Patterns of technological innovation."
Technology Review 80(June-July): 40-47.

Albino, V., N. Carbonara and l. Giannoccaro (2006). "Innovation in industrial
districts: An agent-based simulation model." International Journal of
Production Economics 104(1): 30.

' Ancona, D. G. and D. F. Caldwell (1992). "Demography and design: predictors of
new product team performance." Organization Science 3(3): 321-341.

Anderson, N., C. K. W. D. Dreu and B. A. Nijstad (2004). "The routinization of
innovation research: a constructively critical review of the state-of-the-
science." Journal of Organizational Behavior 25(2): 147.

Argote, L., B. McEvily and R. Reagans (2003). "Managing knowledge in
organizations: an integrative framework and review of emerging themes."
Management Science 49(4): 571.

Argyris, C. and D. Schon (1978). Organizational learning: a theory of action
perspective. Reading, MA, Addison-Wesley.

Baer, M. and M. Frese (2003). "Innovation is not enough: Climates for initiative
and psychological safety, process innovations, and ﬁrm performance."
Journal of Organizational Behavior 24(1): 45.

Bagozzi, R. P. and Y. Yi (1991). "Multitrait-multimethod matrices in consumer
research." Journal of Consumer Research 17: 426-439.

Bagozzi, R. P., Y. Yi and L. W. Philips (1991). "Assessing construct validity in
organizational research." Administrative Science Quarterly 36(3): 421-458.

Baines, T., D. E. Whitney and C. Fine (1999). "Manufacturing technology
sourcing practices in the USA." International Journal of Production
Research 37(4): 939-956.

Bamberger, P. (1991). Reinventing innovation theory: critical issues in the
conceptualization, measure,ent, and analysis of technological innovation.
Research in the sociology of organizations. S. B. Bacharach. Greenwich,
CT, Jai Press Inc.

Barney, J. (1991). "Firm resources and sustained competitive advantage."
Journal of Management 17(January): 99-120.

136

Baron, R. M. and D. A. Kenny (1986). "The moderator-mediator variable
distinction in social psychological research: conceptual, strategic, and
statistical considerations." Journal of personality and social psychology
51(6): 1173-1182.

Beatty, C. A. (1992). "Implementing advanced manufacturing technologies: rules
of the road." Sloan Management Review 33(4): 49-60.

Belassi, W. and O. I. Tukel (1996). "A new framework for determining critical
success/failure factors In projects. " International Journal of Project
Management14(3): 141- 151.

Boyer, K. K. and R. Verma (2000). "Multiple raters in survey-based operations
management research: a review and tutorial. " Production and Operations
Management 9(2) 128- 140.

Bradshaw, G. F., P. W. Langley and H. A. Simon (1983). "Studying scientiﬁc
discovery by simulation." Science 222: 971-975.

Burcher, P., G. Lee and A. Sohal (1999). "Lessons for implementing AMT Some
case experiences with CNC in Australia, Britain and Canada."
International Journal of Operations 8 Production Management 19(5/6):
515.

Choudhury, M. A., T. Azid and M. A. Klasra (2006). "Knowledge-induced
technological change:A study involving the automobile innovation in
Malaysia with layers of techniques." International Journal of Social
Economics 33(11): 744-765.

Co, H. C., B. E. Patuwo and M. Y. Hu (1998). "The human factor in adanced
manufacturing technology adoption." lntemational Journal of Operations 8
Production Management 18(1): 87-106.

Cohen, M. A., J. Eliashberg and T.-H. Ho (1996). "New product development:
The performance and time-to-market tradeoff." Management Science
42(2): 173.

Cohen, W. M. and D. A. Levinthal (1990). "Absorptive Capacity: A New
Perspective On Learning And Innovation." Administrative Science
Quarterly 35(1): 128.

Cote, J. A. and M. R. Buckley (1987). "Estimating trait, method, and error

variance: Generalizing across 70 construct validation studies." Journal of
marketing research 24(3): 315-318.

137

Cummings, J. C. (2004). "Work groups, structural diversity and knowledge
sharing in a global organization." Management Science 50(3): 352-364.

Damanpour, F. (1996). "Organizational complexity and innovation: developing
and testing multiple contingency models." Management Science 42(5):
693.

Damanpour, F. and S. Gopalakrishnan (2001). "The dynamics of the adoption of
product and process innovations in organizations." Journal of
Management Studies 38(1): 2001.

' Davenport, T. H. (1993). Process innovation: rengineering work through
unformation technology. Boston, MA, Harvard Business School Press.

Davies, A. J. and A. K. Kochhar (2000). "A framework for the selection of best
practices." International Journal of Operations & Production Management
20(10): 1203.

Detert, J. R. and R. G. Schroeder (2000). "A framework for linking culture and
improvement initiatives in organizations." Academy of Management
Review 25(4): 850.

Drake, A., S. F. Haka and S. P. Ravenscroft (2001). "An ABC simulation focusing
on incentives and innovation." Issues in Accounting Education 16(3): 443.

Drake, A. R., S. F. Haka and S. P. Ravenscroft (1999). "Cost system and
incentive structure effects on innovation, efﬁciency and proﬁtability in
teams." The Accounting Review 74(3): 323.

Dutta, 8., O. Narasimhan and S. Rajiv (1999). "Success in High-Technology
Markets: ls Marketing Capability Critical?" Marketing Science 18(4): 547.

Earley, P. C. and E. Mosakowski (2000). "Creating hybrid team cultures: an
empirical test of transnational team functioning." Academy of Management
Journal 43(1): 26-49. .

Edmondson, A. C. (1999). "Psychological safety and learning behavior in work
teams." Administrative Science Quarterly 44: 350-383.

Edmondson, A. C. (2003). "Framing for learning: lessons in successful
technology implementation." California Management Review 45(2): 34.

Edmondson, A. C., R. M. Bohmer and G. P. Pisano (2001). "Disrupted routines:

team learning and new technology implementation in hospitals."
Administrative Science Quarterly 46: 685-716.

138

Edmondson, A. C., A. B. Winslow, R. M. Bohmer and G. P. Pisano (2003).
”Learning how and learning what: effects of tacit and codiﬁed knowledge
on performance improvement following technology adoption." Decision
Sciences 34(2): 197.

Eisenhardt, K. M. and J. Martin (2000). "Dynamic Capabilities: What are they?"
Strategic Management Journal 21: 1105-1121.

Engardio, P. (2006). R&D offshoring: Is it working?, Business Week Online.

. Ettlie, J. E. and S. A. Reifeis (1987). "Integrating Design and ManufaCturing to
Deploy Advanced Manufacturing Technology." Interfaces 17(6): 63.

Ettlie, J. E. and E. M. Reza (1992). "Organizational Integration and Process
Innovation." Academy of Management Journal 35(4): 795.

Fedor, D. B., S. Ghosh, S. D. Cladwell, T. J. Maurer and V. R. Singhal (2003).
"The effects of knowledge management on team members' ratings of
project success and impact." Decision Sciences 34(3): 513.

Flynn, B. B., S. Sakakibara and R. G. Schroeder (1995). "Relationship between
JIT and TQM: Practices and performance." Academy of Management
Journal 38(5): 1325.

Fornell, C. and D. F. Larcker (1981). "Evaluating structural equation models with
unobservable variables and measurement." Journal of marketing research
18: 39-50.

Frohlich, M. T. and J. R. Dixon (1999). "Information systems adaptation and the
successful implementation of advanced manufacturingtechnologies."
Decision Sciences 30(4): 921-957.

Gooding, R. Z. and J. A. Wagner III (1985). "A meta-analytic review of the
relationship between size and performance: the productivity and efﬁciency
of organizations and their subunits." Administrative Science Quarterly 30:
462-481.

Govindarajan, V. and P. K. Kopalle (2006). "Disruptiveness of innovations:
measurement and an assessment of reliability and validity." Strategic
Management Journal 27: 189-199.

Granello, D. H. and J. E. Wheaton (2004). "Online data collection: strategies for
research." Journal of counseling 8 development 82: 387-393.

Ha, A. Y. and E. L. Porteus (1995). "Optimal timing of reviews in concurrent
design for manufacturability." Management Science 41 (9): 1431.

139

Hayes, R. H. and S. C. Wheelwright (1984). Restoring our competitive edge:
competing through manufacturing, John Wiley & Sons, Inc.

Hewitt, F. (1995). "Business process innovation in the mid-19905." Integrated
manufacturing systems 6(2): 18.

Hoegl, M. and Gemuenden (2001). "Teamwork quality and the success of
innovative projects: a theoretical concept and empirical evidence."
Organization Science 12(4): 435.

'Hoegl, M. and L. Proserpio (2004). "Team member proximity and teamwork
innovative projects." Research Policy 33: 1153-1165.

Holahan, P. J., 2. H. Aronson, M. P. Jurkat and F. o. Schoorman (2004).
"Implementing computer technology: a multiorganizational test of Klein
and Sorra's model." Journal of Engineering and Technology Management
21: 31-50.

Hottenstein, M. P. and J. W. Dean Jr. (1992). "Managing risk in advanced
manufacturing technology." California Management Review 34(4): 112.

Hu, L. and P. M. Bentler (1995). Evaluating model ﬁt. Structural equation
modeling: concepts, issues and applications. R. H. Hoyle. Thousand
Oaks, CA, Sage Publications.

Hu, L. and P. M. Bentler (1998). "Fit indices in covariance structure modeling:
sensitivity to underparameterized model misspeciﬁcation." Psychological
Methods 3(4): 424-453.

Hu, L. and P. M. Bentler (1999). "Cutoff criteria for ﬁt indexes in covariance
structure analysis: conventional criteria versus new alternatives."
Psychological Methods 6(1): 1-55.

lansiti, M. (2000). "How the incumbent can win: Managing technological
transitions in the semiconductor industry." Management Science 46(2):
169.

James, L. R., R. G. Demaree and G. Wolf (1984). "Estimating within—group
interrater reliability with and without response bias." Journal of Applied
psychology 69: 85-98.

Jayanthi, S. and K. K. Sinha (1998). "Innovation implementation in high

technology manufacturing: A chaos-theoretic empirical analysis." Journal
of Operations Management 16: 471-494.

140

Jugdev, K. (2004). "Through the looking glass: examining theory development in
project management." Project Management Journal 35(3): 15.

Katz, R. and T. J. Allen (1988). Investigating not invented here (NIH) syndrome:
a look at the performance, tenure, and communication patterns of 50 R&D
project groups. Readings in the management of innovation. W. L. Moore.
New York, NY, Ballinger/Harper & Row: 293-309.

Kazanjian, R. K. and R. Drazin (1986). "Implementing Manufacturing Innovations:
Critical Choices of Structure and Stafﬁng Roles." Human Resource
Management 25(3): 385-403.

Kerzner, H. (1987). "In search of excellence in project management." Journal of
Systems Management 38(2): 30.

Khurana, A. (1999). "Managing complex production processes." Sloan
Management Review Winter.

Kimzey, C. H. and S. Kurokawa (2002). "Technology outsourcing in the U.S. and
Japan." Research Technology Management 45(4): 36.

Klein, K. J., A. b. Conn and J. S. Sorra (2001). "Implementing computerized
technology: an organizational analysis." Journal of Applied psychology
86(5): 811-824.

Klein, K. J. and A. P. Knight (2005). "Innovation implementation: overcoming the
challenge." American Psychology Society 14(5): 243.

Klein, K. J. and J. S. Sorra (1996). "The challenge of innovation implementation."
Academy of Management Review 21: 1055-1080.

Leonard-Barton, D. (1988). "Implementation as mutual adaptation of technology
and organization." Research Policy 17: 251-267.

Lewin, A. Y. and C. Peeters (2006). "The top-line allure of offshoring." Harvard
Business Review 84(3): 22.

Linderman, K., R. G. Schroeder and A. S. Choo (2006). "Six sigma: the role of
goals in improvement teams." Journal of Operations Management 24(6):
779-790.

Linderman, K., R. G. Schroeder and A. S. Choo (In-press). "Six sigma: the role of
goals in improvement teams." Journal of Operations Management.

141

Macher, J. T. and D. C. Mowery (2003). ""Managing" learning by doing: An
empirical study in semiconductor manufacturing." The Journal of Product
Innovation Management 20(5): 391.

Malhotra, N. K., S. S. Kim and A. Patil (2006). "Common method variance in IS
research: a comparison of alternative approaches and a reanalysis of past
research." Management Science 52(12): 1865-1883.

Mangione, T. W. (1998). Mail Surveys. Handbook of Applied Social Research
Methods. D. J. Rog. Thousand Oaks, CA, SAGE Publications Ltd.: 399-
427.

McDermott, C. M. and G. N. Stock (1999) "Organizational culture and advanced
manufacturing technology implementation. " Journal of Operations
Management 17(5): 521.

McLachlin, R. (2004). "Internal plant environment and just-in-time
manufacturing." International Journal of Manufacturing Technology and
Management 6(2): 112-124.

Menon, T. and J. Pfeffer (2003). "Valuing internal vs. external knowledge:
explaining the preference for outsiders." Management Science 49(4): 497.

Menon, T, L. Thompson and H. -S. Choi (2006). "Tainted knowledge v. tempting
knowledge: people avoid knowledge from internal rivals and seek
knowledge from external rivals. " Management Science 52(8): 1129-1144.

Meyers, P. W., K. Sivakumar and C. Nakata (1999). "Implementation of industrial
process innovations: factors, effects and marketing implications." Journal
of Product Innovation Management 16: 295-31 1.

Moenaert, R. K. and W. E. Soqder (1996). "Context and antecedents of
information utility at the R&D/Marketing interface." Management Science
42(11): 1592.

Moore, G. C. and I. Benbasat (1991). "Development of an instrument to measure
the perceptions of adopting an information technology innovation."
Information systems research 2(2): 192-222.

Nahm, A. Y., S. S. Rao, L. Solis-Galvan and T. S. Ragu-Nathan (2002). "The Q-
sort method: assessing reliability and construct validity of questionnaire
items at a pre-testing stage." Journal of modern applied statistical
methods 1(1): 114-125.

Neter, J., M. H. Kutner, C. J. Nachtsheim and W. Wasserman (1996). Applied
linear regression models, McGraw -Hill Irwin.

142

Nonaka, l. (1991). "The Knowledge-Creating Company." Harvard Business
Review 69(6): 96.

Nunnally, J. C. (1967). Psychometric Theory. New York, McGraw-Hill.

O'Leary-Kelly, S. W‘. and R. J. Vokurka (1998). "The empirical assessment of
construct validity." Journal of Operations Management 16: 387-405.

Pfeffer, J. and C. A. O'Reilly (1987). "Hospital demography and turnover among
nurses." Industrial Relations 36: 158-173.

Pilkington, A. (1998). "Manufacturing strategy regained: evidence for the demise
of best-practice." California Management Review 41(1): 31.

Pinto, J. K. and S. J. J. Mantel (1990). "The causes of project failure." IEEE
Transactions on Engineering Management 37(4).

Pinto, M. B., J. K. Pinto and J. E. Prescott (1993). "Antecedents and
consequences of project team cross-functional cooperation." Management
Science 39(10): 1281.

Pisano, G. (1997). The Development Factory: Unlocking The Potential of
Process Innovation. Boston, MA, Harvard Business School Press.

Podsakoff, P. M., S. B. MacKenzie, J.-Y. Lee and N. P. Podsakoff (2003).
"Common method biases in behavioral research: A critical review of the

literature and recommended remedies." Journal of Applied psychology
88(5): 879-903.

Prahalad, C. K. and G. Hamel (1995). "The Core Competence of the
Corporation." Harvard Business Review(July-August): 118-128.

Primo, M. A. M. and S. D. Amundson (2002). "An exploratory study of the effects
of supplier relationships on new product development outcomes." Journal
of Operations Management 20: 33-52.

Quinn, R. E. and J. Roghbaugh (1981). "A competing values approach to
organizational effectiveness." Public productivity review 5: 122-140.

Repenning, N. P. (2002). "A simulation-based approach to understanding the

dynamics of innovation implementation." Organization Science 13(2): 109-
127.

143

Repenning, N. P. and J. D. Sterman (2002). "Capability traps and self-conﬁrming
attribution errors in the dynamics of process improvement." Administrative
Science Quarterly 47: 265-295.

Salaheldin, S. I. (2005). "JIT implementation in Egyptian manufacturing ﬁrms:
some empirical evidence." lntemational Journal of Operations 8
Production Management 25(3): 354-369.

Sarin, S. and C. M. McDermott (2003). "The effect of team leader characteristics
on learning, knowledge application, and performance of cross-functional
new product development teams." Decision Sciences 34(4): 707-739.

Schroeder, R. G., G. D. Scudder and D. R. Elm (1989). "Innovation in
Manufacturing." Journal of Operations Management 8(1): 1.

Sedgley, N. and B. Elmslie (2001). "Agglomeration and congestion in the
economics of ideas and technological change." The American Journal of
Economics and Sociology 60(1): 101-121.

Sen, F. and A. H. Rubenstein (1989). "External technology and in-house R8D's
facilitative role." Journal of Product Innovation Management 6: 123-138.

Sen, F. and A. H. Rubenstein (1990). "An exploration of factors affecting the
integration of in-house R8D with external technology acquisitioni
strategies of a ﬁrm." IEEE Transactions on Engineering Management _
37(4): 246.

Shani, A. B., R. M. Grant, R. Krishnan and E. Thompson (1992). "Advanced
manufacturing systems and organizational choice: sociotechnical system
approach." California Management Review 34(4): 91.

Sharma, R. and Pertton (2004). "The contingent effects of management
support and task interdependence on successful information systems
implementation." MIS Quarterly 27(4): 533-555.

Sharma, 8., R. M. Durand and O. Gur-Arie (1981). "Identiﬁcation and analysis of
moderator variables." Journal of marketing research 18(3): 291-300.

Shenhar, A. J. (2001). "One Size Does Not Fit All Projects: Exploring Classical
Contingency Domains." Management Science 47(3): 394-414.

Shenhar, A. J., A. Tishler, D. L. Dvir and T. Lechler (2002). "Reﬁning the search
for project success factors: a multivariate, typological approach." R&D
Management 32(2): 111-126.

144

Sila, l. (2007). "Examining the effects of contextual factors on TQM and
performance through the lens of organizational theories: an empirical
study." Journal of Operations Management 25: 83-109.

Simon, H. A. (1985). What we know about the creative process. Frontiers in
creative and innovation management. R. L. Kuhn. Cambridge, MA,
Ballinger: 3-20.

Sinha, U. B. (2006). "Patent Enforcement, Innovation and Welfare." Journal of
Economics 88(3): 211-241.

‘ Skinner, W. (1969). "Manufacturing-missing link in corporate strategy." Harvard
Business Review 47(May-June): 136-145.

Small, M. H. and M. M. Yasin (1997). "Advanced manufacturing technology:
Implementation policy and performance." Journal of Operations
Management 15(4): 349.

Smith, K. G., C. J. Collins and K. D. Clark (2005). "Existing knowledge,
knowledge creation capability and the rate of new product introduction in
high-technology ﬁrms." Academy of Management Journal 48(2): 346-387.

Spear, S. and H. K. Bowen (1999). "Decoding the DNA of the Toyota Production
System." Harvard Business Review 77(5): 96.

Starbuck, W. H. and J. M. Mezias (1996). "Opening Pandora's box: Studying the
accuracy of managers’ perceptions." Journal of Organizational Behavior
17(2): 99.

Stewart, G. (2006). "A meta-analytic review of relationships between team design
features and team performance." Journal of Management 32(1): 29-54.

Stock, G. N. and M. V. Tatikonda (2004). "External technology integration in
product and process development." International Journal of Operations 8
Production Management 24(7): 642.

Stratman, J. K. and A. V. Roth (2002). "Enterprise Resource Planning (ERP)
Competence Constructs: Two-Stage Mufti-Item Scale Development and
Validation." Decision Sciences 33(4): 601-628.

Subramaniam, M. and M. A. Youndt (2005). "The inﬂuence of intellectual capital
on the types of innovative capabilities." Academy of Management Journal
48(3): 450-463.

Sundbo, J. (1999). The Theory of Innovation: Entrepreneurs, Technology and
Strategy. Cheltenham, UK, Edward Elgar Publishing Limited.

145

Swink, M. (2000). "Technological innovativeness as a moderator of new product
design integration and top management support." The Journal of Product
Innovation Management 17(3): 208.

Swink, M. (2002). "Product development--faster, on-time." Research Technology
Management 45(4): 50.

Teece, D., G. Pisano and A. Shuen (1997). "Dynamic Capabilities and Strategic
Management." Strategic Management Journal 18(7): 509-533.

I Troy, L. (2006). Almanac of business and industrial ﬁnancial ratios, CCH, Wolters
Kluwer.

Tu, Q., M. A. Vonderembse, T. S. Ragu-Nathan and T. W. Sharkey (2006).
"Absorptive capacity: enhancing the assimilation of time-based
manufacturing practices." Journal of Operations Management 24: 692-
710.

Turner, J. R. (2004). "Five necessary conditions for project success."
International Journal of Project Management 22: 349.

Tyre, M. J. and O. Hauptman (1992). "Effectiveness of organizational responses
to technological change in the production process." Organization Science
3(3): 301-320.

Tyre, M. J. and W. J. Orlikowski (1993). "Exploiting opportunities for
technological improvement in organizations." Sloan Management Review
35(1): 13. '

Tyre, M. J. and W. J. Orlikowski (1994). ’Windows of opportunity: temporal
patterns of technological adaptation in organizations." Organization
Science 5(1): 98.

Vanhaverbeke, W., G. Duysters and N. BNoorderhaven (2002). "External
technology sourcing through alliances or acquisitions: an analysis of the
application-speciﬁc integrated circuits industry‘." Organization Science
13(6): 714.

Wagner III, J. A. and J. R. Hollenbeck (1992). Management of Organizational
Behavior. Englewood Cliffs, NJ, Prentice-Hall.

Waterson, P. E., C. W. Clegg, R. Bolden, K. Pepper, P. B. Warr and T. D. Wall
(1999). "The use and effectiveness of modern manufacturing practices: a
survey of UK industry." International Journal of Production Research 37:
2271-2292.

146

Werts, C. E., R. L. Linn and K. G. Joreskog (1974). "Interclass reliability
estimates: testing structural assumptions." Educational and psychological

measurement 34: 25-33.

West, S. G., J. F. Finch and P. J. Curran (1995). Structural equations models
with nonnormal variables: problems and remedies. Structural equation
modeling: concepts, issues and applications. R. H. Hoyle. Thousand
Oaks, CA, Sage Publications.

,Zahra, S. A. and S. R. Das (1993). "Innovation Strategy and Financial
Performance in Manufacturing Companies: An Empirical Study."
Production and Operations Management 2(1): 15.

Zahra, S. A. and G. George (2002). "Absorptive capacity: A review,
reconceptualization, and extension." Academy of Management. The
Academy of Management Review 27(2): 185.

Zammuto, R. F. and E. J. O'Connor (1992). "Gaining advanced manufacturing
technologies' beneﬁts: the roles of organization design and culture."
Academy of Management Review 17(4): 701-728.

Zollo, M. and S. G. Winter (2002). "Deliberate learning and the evolution of
dynamic capabilities." Organization Science 13(3): 339.

147

 

IIIIIIIIIIIQIIIIIIIIIIIIIIIIIIIIII