THE DEVELOPMENT OF WORK TEAMS
IN A COMPLEX ORGANIZATION

Thesis for H10 Degree of DH. 0.
MICHEGAN STATE UNIVERSITY

John Stephen Heinen
‘3 971

Date

"”9 LXQEAKY'
Michigan State
University K

    
    

Illlll ll! HIM" H! II lllll lllllll

3 1293 10 31 5169

 
 

This is to certify that the
thesis entitled ‘- ~' ~-
The Develogment of Work Teame‘_l

In A Complex Organization;

presented by .. “a,” M
John Stephen Heinen

has been accepted towards fulfillment» - ' i. - . ,
of the requirements for

PhD degree in __Emhology

ELM

v (Major profisor
Euge e Jacobson

 

  

August 13 s 1971

 

0-7639

 

 

——v—__~
u-w !‘-’

 

 

 

 

ABSTRACT
THE DEVELOPMENT OF WORK TEAMS N A COMPLEX ORGANIZATION
By

John Stephen Heinen

The development of a work team approach to management
is recommended by several organizational psychologists
(Likert, 1961, 1967; Blake and Mouton, 196U; Nelson, 19M9).
However, very little research has been done on the industrial
work group, especially the pattern of its development toward
maturity.

This dissertation investigates the development of work
teams within a complex organization. Tuckman (1965) in his
review of the group development literature stressed the point
that most of the studies of group development examined
therapy or sensitivity groups and that there was a lack of
studies involving industrial groups. The only study using
industrial groups was Jacobson's (1956) post hoc analysis of
voluntary organizational groups. Using the Jacobson analysis
as the primary guideline, a model of work team development
was proposed.

Group maturity was defined by the variables of cohesive-
ness, ability to solve problems, and shared norms. Four
group process variables, identification, task development,
communication, and leadership distribution, were hypothesized
to contribute to the group's maturity. Each process was

expected to be critical to the group as it moved through four

John Stephen Heinen

stages of increasing maturity. Identification was hypothe—
sized to be most important to the group at its initial stage
of development; task development most important at stage

two; communication most important at stage three; and leader-
ship distribution most important at stage four.

The model was tested in a medium sized Scanlon Plan
companyo A cross—sectional analysis of twenty work teams
was used in an exploratory examination of the model of work
team development.

None of the hypotheses was confirmed. Sources of error
variance in the study were discussed and some recommenda-
tions for future research were made. With respect to the
model, two points were stressed in the discussion. In test-
ing the theoretical model, the sample of twenty teams was
considered to be representative of industrial work teams on
a maturity dimension. The representativeness of the sample
was questioned since the teams seem to represent a rela-
tively primitive level of development.

The second point emphasized the impreciseness of the
model presented in this dissertation in articulating the
pattern of events occurring within a given stage. This
limitation in the theory seemed to prohibit an accurate test

of the model in this situation.

r\[

.. g j _
Approved: <:LJ ’ CkilfluELM’——-
Date: ( h“ \J
Dissertation Committee: (

Dr. Eugene Jacobson Dr. John Wakeley
Dr. Carl Frost Dr. Joel Aronoff
Dr. Lawrence Messe

 

THE DEVELOPMENT OF WORK TEAMS IN A COMPLEX ORGANIZATION
By

John Stephen Heinen

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Psychology

1971

ACKNOWLEDGEMENTS

Many people provided invaluable assistance in com-
pleting this dissertation. Dr. Eugene Jacobson not only
served as my dissertation chairman, but also has been a
dominant influence throughout my graduate career. With
the patience and skill of a master craftsman, he has
attempted to hone my talents as a social psychologist.

Dr. Carl Frost introduced me to the importance, excitement,
and responsibilities of field research in industry. Dr.
Lawrence Messe contributed his expert knowledge of small
group processes and statistical analysis. Dr. John Wakeley
and Dr. Joel Aronoff were both extremely helpful in clarify-
ing my conceptual thinking of this problem.

I am indebted to all the members of the organization
who participated in this research for their cooperation but
especially to the manager of the organization development and
personnel department for his support throughout this project.
Mr. Steve Iman of the Institute for Social Research at the
University of Michigan graciously extended his help in ob-
taining the cooperation of the Institute in the data collec—
tion.

My inspiration in this project came from the love of my

wife, Jean, and my son, Chris. The birth and early

ii

development of my son during this dissertation often
stimulated my thinking about group development. However,
more than that, his cheery laugh made all my problems seem
small. Jean was always there to help perform some task,

to listen to my troubles or to share my happiness.

iii

LIST OF
LIST OF
Chapter

I.

II.

III.

IV.

VI.

TABLE OF CONTENTS

Page

TABLES .......... ........................... .... Vi
FIGURES .............................. ..' ..... ...viii
Introduction .................... . ..... .... ...... 1
Group Development Literature ................. 6
Therapy Groups .................. .......... 8

Sensitivity Groups ........................ 10
Task Groups ............................... 13
Comparison of the Different Theoretical

Approaches .. ..... ........................ 17

A Theoretical Model of Group Development ........ 2A

Group Maturity ......................... ...... 2“
Stages in Group Development .................. 27

The Work Team Concept at the Research Site ...... 32
Methods and Procedures .............. ............ 37
Data Gathering Procedures and Research Design. 37

subjects 0 0000000000000000 O OIOIOOOOOOOI 39
Definition and Measurement of the Variables .. 39

Group Maturity Measures ....... .. ..... ..... 39

Group Process Measures ..... ............... A3
ReSUltS OOOOOOOOOOOOOOOOOOOOO ....... OOOIOOOOOOIOO “8
Group Maturity Measures ..................... U8
Relation of Group Process to Group Maturity .. SO
DiscuSSion ......... O O O O 0‘ ...... O O I O O 0 O O O O O O O O O O O O 61
Group Maturity Measures ..... . ......... ....... 61

Group Process Measures ...... ............... 67
Relation of Group Process to Group Maturity ., 69

iv

Chapter Page

Comments about the Theory ..... ...... ......... 70

Summary ..... . .................... ............ 75

BIBLIOGRAPHY ........................................... 76
APPENDICES

A. Heinen Work Team Survey ........... . ..... ........ 80

B. Instructions to Managers for Paired Comparison
Work Team Ratings ....... ...... ................. 85

C. Item Analysis ..................... .............. 86

Table

II.
III.

IV.

VI.

VII.

VIII.

IX.

XI.

XII.

XIII.

XIV.

XV.

LIST OF TABLES

Work Teams in Analysis ................... .. ..... A0
The Data Sources of Variables used in the Study . A7
Z Scores of Group Maturity Variables ....... ..... A9

Correlations Among Group Maturity Measures and
the Ratings of the Work Team ................... U9

AOV of Group Maturity Total Score x Group Process 52

ADV of Group Maturity Cohesiveness Score x Group
Process O‘OOOOOOOOOOO0.0.0.0....IOOOOOOOOOOOOOOOO 52

AOV of Group Maturity Problem Solving Score x
Group Process 00...... 00. IOOOOOOOIOOOOOOOOOOOO 53

AOV of Group Maturity Shared Norm Score x Group
Process ..... ...... . ......... . ..... . ..... ....... 53

ADV Mean Summary of Group Maturity Total Score x
Group Process .......... .... ..... ............... 5A

AOV Mean Summary of Group Maturity Cohesiveness
Score x Group Process .... ........ .............. SA

AOV Mean Summary of Group Maturity Problem Solv-
ing Score x Group Process ...... ............... 55

ADV Mean Summary of Group Maturity Shared Norm
Score x Group Process ..... . ............... 55

One Way AOV's of Proportions of Group Process x
Total Group Maturity .... ...... ................. 56

One Way AOV's of Proportions of Group Process x
Cohesiveness ..... ..... .. ..... .................. 56

One Way AOV's of Proportions of Group Process x
Problem Solving . .............. ................. 57

vi

Table
XVI.

XVII.

XVIII.
XIX.

XX.

XXI.

XXII.

XXIII.
XXIV.
XXV.

XXVI.

XXVII.

Page

One Way AOV's of Proportions of Group Process
XShared Norms 0000000000000000000000000000000 57

ADV Mean Summary of Proportions of Group
Process x Group Maturity ..................... 58

Correlations Between Group Maturity Variables . 6O
Correlations Between Group Process Variables .. 6O

Correlations Between Group Maturity and Group
Process Variables 0000000000000000000000000000 6O

ICLS Scales Inter-Item Correlations ........... 89
Median and Range of Inter-Item Correlations,
Item-Cluster, and Item Communalities of
scales in ICLS 000000000000000000000000000000 91
Alphas of ICLS scales 0000000000000000000000000 91
Normative Inter-Item Correlations ............. 93

Leadership Distribution Inter—Item Correlations 95

Median and Range of Inter-Item Correlations and
Alphas for Normative and Leadership Distribu-

tion 0000000000000000000000000000000000000000 97

Variable Means, Standard Deviations, and
Ranges for the Twenty Production Work Teams .. 98

LIST OF FIGURES

Figure Page

1. Proportion of group maturity explained by the
elements of group process at each level of
maturity 0000000000000000000000 00000000000000... 3].

2. Proportion of group maturity explained by the

elements of group process at each level of
maturity 00000000000 0 0000000 0 0000000 00.00.00.000 71

viii

CHAPTER I

INTRODUCTION

The application of social psychological theory and
methods to the study of industrial situations received its
major impetus from two sources during the 1930's. They
were Elton Mayo and his associates' Hawthorne studies and
Kurt Lewin and his associates' studies of group process.

The Hawthorne studies (Roethlisberger and Dickson,
1939) stimulated a change in emphasis in organizational
theory previously dominated by the scientific management
approach of Taylor (1911). The mechanistic approach and
analysis were actually the basis for the experiments at-
the Hawthorne plant of Western Electric. However, the
keen observations of the researchers led to new explana—
tions of work productivity.

The study of the bank wiremen showed that their

behavior at work could not be understood without

considering the informal organization of the

group and the relation of this informal organiza-

tion to the total social organization of the

company (Roethlisberger and Dickson, 1939, p. 551).
From an emphasis on fatigue factors and work conditions,

the authors turned to the analysis of interpersonal

relationships among the workers and between the workers

and management. The results suggested that important
variables to examine in organizational research are the
social organization of the work group, the social rela—
tions between the supervisor and his subordinates, the
informal standards governing the behavior of members of
the work groups, and the attitudes and motives of workers
existing in a group context. The authors concluded:

A great deal of attention has been given to the

economic function of industrial organization.

Scientific controls have been introduced to

further the economic purposes of the concern and

of the individuals within it. Much of this

advance has gone in the name of efficiency or

rationalization. Nothing comparable to this

advance has gone on in the development of skills

and techniques for securing cooperation, that is,

for getting individuals working together effec—

tively and with satisfaction to themselves

(Roethlisberger and Dickson, 1939, p. 552-3).

These studies pointed towards a preoccupation with the
social psychology of the industrial environment and more
specifically with the group dynamics of specific industrial
situations.

Equally important in this change of interest was the
work of Kurt Lewin, the father of group dynamics research.
His innovative development of theory and systematic
methodological techniques have inspired and directed many
researchers. Most notable of his research is the inves-
tigation of group atmosphere and leadership styles (Lewin,
Lippitt, and White, 1939). At different periods in the

study an authoritarian, a democratic, or a laissez-faire

approach was employed by adult leaders of boys' groups.

The results showed that the boys under authoritarian
leaders were dependent on the leader and often responded
aggressively to each other; whereas the democratic groups
were friendly and freer in their suggestions. In the
democratic groups there was no change in productivity when
the leader left the room. In contrast there was a sharp
drop-off in productivity after the authoritarian leader
left the room.

This study motivated many future studies of leader-
ship style and many investigations of how different
leadership styles lead to differential effects upon the
group members.

These two major works gave rise to the human relations
approach in industry. The most attention has been given
to the characteristic pattern of decision making in an
organization. Today, participative management as a pre-
ferred mechanism for decision making is integral to many
organizational theories. There are several different
theories about ways of improving organizational effective-
ness, all of which identify increased participation as a
major component. Involved in this approach for many is
the development of the group or the development of team-
work. The work group is the basic unit or building block
in the theories of Nelson (1949), Likert (1961, 1967) and
Blake and Mouton (196A). The work group also has a

prominent role in Shepard's (1965) theory of organic

structure and Burns and Stalker's (1961) interest in socio-
technical systems. According to all these theorists the
most effective managers tend to develop group process and
involve the group in the management process.

Despite the importance attributed to the work group
by these authors, there has been little empirical research
on the work group within the complex organization. These
organization theorists have generally been prescriptive in
their treatment of small groups. Their emphasis has been
upon the individual and his relationship to the total
organization while the development and operations of the
group have been neglected.

The small group in isolation has been studied exten-
sively by social psychologists and sociologists (Cartwright
and Zander, 1968; Hare, Borgatta, and Bales, 1965; McGrath
and Altman, 1966). Yet, the development of a group,
particularly an industrial group, rarely has been examined.
Most small group research focuses upon a particular element
of a group, such as communication structure, in a laboratory
setting.

Weick (1965) in his discussion of laboratory experi-
mentation pointed out that subgrouping and the interaction
between groups is the essence of an organization. He
further stated that the failure of many laboratory experi-
ments was the inability of the laboratory groups to mirror

natural groups in their history and processes of development.

Even though the importance of the groups in an organi—
zation is noted, Katz and Kahn (1966) in their monograph

on The Social Psychology of Organizations could only talk

 

about three levels of groups on a development scale. The
first group they described is the primitive group in which
all regulatory mechanisms are lacking. An intermediate
level group is the informal group. Their use of this term
is to denote groups which have well developed structures,
but lack the specialized substructures that characterize
formal organizations. The most developed group is the
social institution which is bound together not only by
organizational structure but also in many other ways.

Obviously, these three levels of development offer
limited help in understanding group growth. The Katz
and Kahn description is not concerned with the development
of processes within a group but with the level of develop—
ment of types of groups. Given the importance of groups
to organizational psychology and to participative management
especially, there appears to be a need to examine group
development within an organizational setting.

Therefore, this dissertation is a pilot study (1) to
describe the development of work teams in a complex organ-
ization and (2) to generate hypotheses about the relation-
ships among the elements of group process in the development

of a work team.

Group Development Literature

 

The most comprehensive review of the group development
literature is found in a review article by Tuckman (1965).
Tuckman devised a classification model based upon the
group's setting, the realm of group behavior, and the
position of the group in the developmental sequence. The
four types of group settings distinguished are: (l) the
group therapy setting, (2) the human relations training
group setting, (3) the natural group setting, and (A) the
laboratory-task setting. In the Tuckman review the last
two types were combined because there were such a small
number of studies in that area and because the theoretical
statements in the research reviewed were generalized to
both areas.

In the study reported in this thesis the primary
interest is the industrial work group which would be a Type
3 group. In the discussion of the literature, Type 3 and
Type A groups will be combined under the heading of task
groups. A few representative studies of the develOpment of
Type 1 and Type 2 groups will be discussed in order to
demonstrate the similarities and differences between the
development of industrial work groups and other types of
groups.

Tuckman's model of development describes two realms of
group behavior, interpersonal and task, each of which passes

through four stages of development. The pattern of

interpersonal relationships is referred to as group
structure and the content of interaction is referred to as
task activity.
According to Tuckman the stages are the same for all
types of groups:
Stage 1 is labeled testing and dependence in the group
structure realm and orientation to the task in the
task activity realm. During this period the members
are determining what behaviors are acceptable in the
group, identifying the task boundaries, and defining
the "ground rules" for the group.
Stage 2 is labeled intragroup conflict in the group
structure realm and emotional response to task demands
in the task activity realm. This stage is character-
ized by intense emotional responses particularly
resistance, conflict, and hostility.
Stage 3 is labeled the development of group cohesion
in the group structure realm and the open exchange
of relevant interpretations in the task activity realm.
Harmony is the key to this stage and conflict is
avoided at all costs. Alternative interpretations of
information are discussed openly.
Stage A is labeled functional role relatedness in the
group structure realm and the emergence of solutions
in the task activity realm. The group performs now
in a constructive problem solving manner. The energy

of the group is functionally related to the task.

Before discussing this model, other concepts concerning

group development will be presented.

Therapy Groups

 

The work of Bion (19A8, 19A9, 1950, 1951) with group
psychotherapy emerged from a psychoanalytic framework, and
contributed very importantly to the ideas of many of the
authors interested in group development.

Bion attributed emotional and dynamic prOperties to
the group itself in addition to the properties of the
individual members. The key concept in his theory is the
work group or sophisticated group. Work group functioning
parallels ego functioning within the individual. The work
group translates the thoughts and feelings of the group
into behavior which is adapted to reality. The work group
is described as a sophisticated, well organized, relatively
emotionless problem solving task unit.

In conjunction with the work group functioning, Bion
discusses three basic assumptions under which the group
operates in relation to its emotional climate. The three
assumptions which he observed operating within the therapy
group are the dependency assumption, the pairing assumption,
and the fight-flight assumption. The dependency assumption
is operating when the group meets to seek nurturance from
the leader on whom it depends. The pairing assumption

involves the subgrouping of two individuals in a relationship

based upon the Messianic hope that the relationship or group
will be preserved for its own sake. The fight-flight
assumption deals with the group's emotional response to
self—preservation.

Work group functioning coacts with only one of the
group assumptions at a time. According to his theory
there is no developmental pattern to the three assumptions.
Any assumption can operate within the group at any given
time depending upon the group's emotional state. Develop-
ment pertains only to the work group in terms of the degree
of sophistication of its organizational structure.

Bach (195A) utilized the dynamics of the group situa-
tion in an intensive psychotherapeutic situation to reach
the goal of free communication on a nondefensive, personal,
and emotional level. He labeled seven developmental phases
in the therapeutic growth process. The group starts with
a situation testing phase where they learn something about
each other and develop expectations about the group.

Growing out of their initial apprehensions and insecurity
about the group is the leader dependence phase. The next

two stages are characterized by the group's increasing skill
and strength in dealing with group members' dependency needs.
At first the group assumes a family quality, then a style
similar to adolescent peer groups.

The social needs and social anxieties aroused by the

group are expressed and integrated into the work of the

10

group through a play and fantasy stage. An ingroup con-
sciousness stage characterized by subgrouping and pairing
precedes the attainment of the work group phase.

Other examples of the development of therapy groups

can be found in Tuckman.

Sensitivity Groups

 

Three sets of investigators (Thelen and Dickerman,
19A9; Miles, 1953; Bennis and Shepard, 1956) have presented
very similar descriptions of the pattern of a sensitivity
group's development. The first group studied was one
attending a session at the National Training Laboratories
in Bethel, Maine and the other two were groups of students
in college sensitivity training classes. In all three the
goal of the group was to improve the interpersonal relations
of the members.

Among the three studies the Bennis and Shepard analysis
is the most detailed in its description of the development
process. The goal of the Bennis and Shepard approach was to
improve the internal communication systems of the group and
to achieve consensual validation of the group's experiences.
Two major problem areas blocking this goal are the authority
or dependence relations and the personal or interdependence
relations.

In resolving the dependence problems there are three

subphases which the group passes through. The ambiguity of

11

the group's situation at the start produces anxious searching
for a common goal and a dependency on the leader to provide
direction. The lack of response from the leader arouses
hostility and rebelliousness within the group. Struggles

for power among warring subgroups, characterize the second
subphase. Resolution of the conflict is reached through

the compromise efforts of the strong, independent members

who had been relatively passive up to this point.

Next the group turns to personal relations. Patching
the wounds of the warring period gradually leads to the
legislation of a strong intermember identity demanding
complete loyalty to the group. The degree of intimacy
required by the group forces a breakdown in the group
identity. The task demands of the final course evaluation
activates certain members toward a willingness to express
their own self concepts. Through rational discussion,
consensus is established among the group members about
each other's conceptual scheme of interpersonal relations.

Although labelled somewhat differently, the stages
discussed by Thelen and Dickerman and by Miles are com-
parable in content. These three studies directly parallel
Tuckman's four stages of forming, storming, norming, and
performing.

An alternative description of group development is
provided by the life cycle models of Mills (196A) and

Mann (1967). Both theorists believe that any developmental

12

model must follow the birth to death life cycle of the group
rather than the simple progression to some implicitly
desired state. Learning groups are the type of groups used
in the development of both models. Mills' model is derived
from his experience with many different learning groups,
whereas the Mann model comes from the observation of a
Harvard class in interpersonal relations. The Mills' five
stage developmental sequence differs very little from the
previous three studies of sensitivity groups. He added a
final stage of the termination of the group and the separa-
tion of the members, and placed less emphasis on the conflict
of the second stage. In the second stage he stressed the
group's initial attempts to replace anomie through goal
establishment and role development.

Mann concentrated upon the changing relationships
between the members of the group and their formal leader,
and not the entire set of interrelationships within a group.
The six phases of group development are defined in terms of
the group members' willingness to take over the leader's
interpretive, analytic role. The first phase of initial
complaining is characterized by nurturance and control
themes. Because of theabdicationcfi‘the leader's expected
role people respond with frustration. The response of the
members follows their own basic mode of response to a novel
situation. Phase two, premature enactment, is the beginning

of some self-awareness of the members as a group. Tension

l3

develops between those who prefer dependence on the leader
and those who prefer autonomy. A confrontation phase of
direct hostility between the group members and the leader
follows. During this period the members complain that the
leader is too passive and weak but at the same time that he
is maipulative and overbearing in relation to task goals.
A high level of enacting the leader role by the members
occurs in the internalization phase. During this period
there is a dramatic increase in identification and an
emphasis on work. The last two stages of separation and
terminal review are concerned with the death of the group.
Depression is the dominant theme in the group, spreading
to feelings of loss and abandonment in the final phase.
Generally the phase is highly emotional with changes in

emotions from mania to depression.

Task Groups

 

Fundamental Interpersonal Relations Orientation is a
theory of interpersonal relations proposed by Schutz (1958)
to analyze effective work groups. Schutz feels that every
person orients himself toward others in characteristic ways
based upon childhood experiences. Three interpersonal needs,
inclusion, control,and affection, are used to explain
characteristic behavior patterns. ‘Inclusion refers to the
need to develop and maintain a satisfactory relationship
with others. Control refers to the decision making process

between people, i.e. relations with others with respect to

1A

authority and power. Affection refers to the need for close
personal and emotional ties with others. Schutz postulates
that the development of any group follows the pattern of
concern for inclusion needs, then concern for control, and
finally concern for affection needs.
The theoretical model of task group development that
is supported by the largest amount of reported research
is the model of Bales and his associates (Bales and Strodt-
beck, 1951; Bales, 1953; Heinicke and Bales, 1953; Philp and
Dunphy, 1959; Bales, 1970). All these studies are based
upon the use of Bales' Interaction Process Analysis (1950)
of discussion groups. Bales and Strodtbeck (1951) and
Bales (1953) found that the interaction of a group within
a single meeting or session moved from problems of orienta-
tion to problems of evaluation, to problems of control.
Concurrent with these transitions, the relative frequencies
of negative and positive reactions tended to increase.
Heinicke and Bales (1953) and Philp and Dunphy (1959)
extended the analysis from a single session to four and
eight sessions respectively. The studies examined changes
in four types of acts: (1) attempted answers, (2) questions,
(3) positive socio-emotional, and (A) negative socio—
emotional. Across all sessions the attempted answers cate-
gory accounted for over 50% of the acts, the questions and
positive socio—emotional categories each about 20%, and the

negative socio-emotional category about.6%. Both studies

15

showed a similar pattern of decreasing task oriented rates
and increasing socio-emotional rates over time. Session one
was heavy on task oriented types of interaction with an inhi-
bition of affect and low rates of negative reactions. The
second session was the period of greatest conflict. In the
Philp and Dunphy group there was a sharp drop in positive
statements with only a slight increase in negative state-
ments, whereas the Heinicke and Bales group had a sharp
increase in negative statements and only a slight drop in
positive statements. The overall effect was still the same.
Sessions three and four showed greater affect with sharper
rises in positive reactions than negative, and less task
oriented activities.

Schroder and Harvey (1963) have developed a model of
group development related to their ideas of a conceptual
organizational system. The model posits the growth of
the group from an undifferentiated mass through stages of
differentiation to a state of maximal integration. Initi-
ally the group begins with some overgeneralized and absolute
definition of its situation. During the second stage some
alternatives to the original situation definition enter and
the absoluteness begins to change to some degree of conflict
between subgroups. Stage three is the beginning of integra-
tion. The members recognize some mutual dependencies and
are Open to multiple alternatives. During the fourth stage
. integration proceeds more intensively as the group's capa-

city to integrate improves.

l6

Olmsted (1962) suggested another developmental model
of a social group. He divides the model into two pre-
conditions and nine major group attributes classified as
rudimentary, emergent, and mature; and refers to the rela-
tions to group products and member satisfactions.

In any particular analysis situation the preconditions
are assumed to be constant in this model. Rudimentary
attributes are concerned with the group getting acquainted
and establishing its basic patterns of interaction. Emergent
attributes focus upon the structural development of the group
with respect to establishing standardized behavior patterns
for accomplishing its tasks. Mature attributes reflect the
results of the preceding development in the form of code of
behavior and a dominant tone of the group interaction.

I Two other group development models were deduced from the
analysis of specific types of groups (Modlin and Paris, 1956;
Jacobson, 1956). Modlin and Faris analysed the history of
a staff team of professionals working in a mental hospital.
They labeled the phases of the team's history: Structural-
ization, unrest, change, and integration. The structural—
ization phase was characterized by the team's definition of
functional roles. The period of unrest involved interdis-
ciplinary friction and personality clashes. The change phase
was the beginning of the notion of the group as a functioning
unit. Communication and group identification were improved.

The integration stage involved internalization of the group

17

in such a way that the patients were not viewed as intruders,
but temporary members.

The only group developmental model that was based upon
an industrial work group emerged from a study of the effec—
tiveness of committees in a voluntary agency (Jacobson, 1956).
Jacobson's main concern was with the structural process in
group growth. He labeled six different stages in the growth
of his groups that were associated with the activity level of
the committees.

Stage one involves an identification of the individual

 

with the group in terms of a congruency of his own
needs with the requirements of the group.

Stage two was characterized by an opportunity for the

 

members to take part in group activities.

Stagg three was characterized by the emergence of a

 

differentiated role structure.

Stage four followed with some improvement in the com—

 

munication practices of the group.

Stage five signalled the point at which leadership

 

practices had their greatest influence on the group's
growth.
Stgge six led to the existence of a relatively en—

during set of group processes.

Comparison of the Different Theoretical Approaches
The various theoretical ideas will be discussed and

compared for the purpose of describing the model of

18

development for the industrial work group. The implications
of the particular type of group for the type of model
employed as well as the specific sequence of development
will be highlighted.

A life cycle model of group development was proposed
by two different people (Mills,.l96A; Mann, 1967). Their
contention was that the death of the group is left out of
most theoretical conceptions. For both theorists the type
of group they were focusing upon was the learning group in
which the members knew upon entering the group that it was
temporary and would end at a given time. The industrial
work group, which is the concern here, has no definite time!
for its death. Thus, preparation for its demise does not
appear to be a significant factor in the developmental pro-
cess of the industrial work group. Consequently, a life
cycle model does not appear appropriate for the industrial
group.

The industrial work group is much different than either
the therapy or sensitivity group in terms of the goals of
the group and the motivation of the members for belonging to’
the group. The purpose of both the therapy and the sensitiv-
ity group is to improve the individual's ability to cope with
the problems of living and to interact less defensively with
others. The group is used as a medium to effect this indi-
vidual develOpment. In this case the task becomes the build-

ing of the interpersonal relationships. The difference

19

between the two types of groups is that the therapy group
involves remedial training for those people with severe
problems in interpersonal skills and the sensitivity group
involves advanced training for those people who want to
increase their capabilities in interpersonal skills.

One feature of therapy or sensitivity groups is the
unusual structuring of the situation to emphasize the
common problems among individuals in interpersonal relation-
ships. The leader abdicates the normal leadership role to
confront the individuals with their inefficient modes of
behaving. At the same time the individual is constantly
cajoled into analysing his own behavior in various situa—
tions. This situation tends to heighten the degree of
threat and emotionality present within the group. In the
groups' activity, specific relationships such as the member
to leader and specific emotions such as hostility are con—
centrated upon.

The industrial work group, on the other hand, is
characterized by its performance of a specific task.
Instead of the interpersonal structure bging the task, it

is serving or accomplishing the task. The group is used

 

to perform the task and not to help the individual solve
his problems.

As Tuckman (1965) points out, most of the literature
deals with therapy or sensitivity groups and not task groups.

Although his sources were primarily reports about therapy and

20

sensitivity groups, Tuckman's synthesis is supposed to apply
to all types of groups. His own discussion points out the
fact that the poorest fit of his model is with the task
groups. He deduced from his investigation of the task

group literature that only the initial stage of forming and
the final stage of performing were essential characteris-
tics of task groups. In reviewing all the literature dis-
cussed here and by Tuckman, those two stages are the ones
mentioned by almost all the theorists.

Sandwiched between the initial forming stage and the
final performing stage in most of the therapy and sensi-
tivity group models are stages related to changing charac-
teristic behavior styles of individual members in inter-
personal situations. Bion's theory of group functioning
described these individual behavior styles in terms of his
three group assumptions of dependency, pairing, and fight-
flight. In his model these three assumptions follow no
developmental pattern. Only work group functioning has the
characteristics of development. His discussion of the
structure of the group as the characteristic in the develop-
ment of work group functioning is consistent with the
emphasis upon the development of structure in the task group
models (Jacobson, 1956; Schroder and Harvey, 1963; Olmsted,
1962).

From this analysis, it would appear that a model of

the development of an industrial work group should

21

concentrate upon the interpersonal structure of the group in
accomplishing the task. The characteristics of its develop—
mental pattern should reflect primarily the progress of the
group as a group and not the progress of individuals as.
good group members.

Concentrating on those models related to task groups
there is a great deal of variation in the description of the
phases of development. Two of the models discussed do not
seem to follow the guidelines laid out above. Shutz (1958)
presented a theory of interpersonal relations whose principal
emphasis was the individual and his characteristics in the
group and not group characteristics as such. Thus, the
Schutz model is more similar to therapy and learning group
models than to task group models.

The work of Bales and his associates focuses upon the
content of interaction within the group. The specific
form of the content patterns varies with the type of problem
before the group. The Philp and Dunphy groups had a dif-
ferent distribution in the categories of acts than did the
Heinicke and Bales groups, which Philp and Dunphy attributed
to the differences in the tasks of the two groups. The task
of the Philp and Dunphy groups was to do a class project and
report it; the task of the Heinicke and Bales groups was to
discuss a human relations problem. Their work is quite
helpful in labeling the pattern of task activities to be
observed in the development of the group's structure, but

they do not directly focus upon the group structure.

22

Schroder and Harvey (1963) and Olmsted (1962) developed
models that applied to a general group type, whereas, Modlin
and Faris (1956) and Jacobson (1956) evolve models from a
particular group situation. The first two models provide a
more thorough understanding of group development at the
abstract conceptual level. The last two models provide
some data about the operations of group development in a
specific situation.

The conceptual framework for the pattern of group
development is best articulated by Schroder and Harvey
(1963). They see the group moving from an undifferentiated
mass through stages of differentiation to the point of
maximal integration. They are less clear about the specific
type of changes within this pattern primarily because they
switch their discussion throughout from the total system to
the single group and vice versa. Their description of the
change that occurs in the group from stage one to stage two
is useful in understanding the task group. In therapy and
sensitivity groups, this is the period of high emotionality
and conflict. Even though the task groups are not character-
ized by the same high emotionality, a higher degree of
tension and conflict exists here than in other stages (Philp
and Dunphy, 1959; Heinicke and Bales, 1953; Modlin and Paris,
1956). According to Schroder and Harvey (1963) this is the
beginning of the differentiation of the group. The group

begins a redefinition of its situation and the alternative

23

modes of Operating and conceptualizations of the situation
produce some tension.

The Jacobson model (1956) being the only model directly
involved with the industrial situation will serve as the basis
for the conception of the development of the industrial work
group that follows. The Jacobson model was based upon a
post hoc analysis of voluntary groups. The model here will
be modified and extended to work groups within a complex

organization.

CHAPTER II
A THEORETICAL MODEL OF GROUP DEVELOPMENT

IN AN INDUSTRIAL SITUATION

From the analysis of the literature, several guidelines
were set up for this model. The structure of the group in
accomplishing its task will be the focus of the model. The
emphasis will be upon the internal system of the group
rather than the external system (Homans, 1950). Within the
complex organization, the task of the group is fairly well
Specified for the group by its organizational context. The
process it develops in accomplishing that task is more at its
discretion.

In developing the model, it is necessary to define more
clearly the traits of the fully developed or mature work

group from the processes within its development.

Group Maturity

 

The concept of group maturity is not well defined.
Bennis and Shepard (1956) claim that maturity for a group
is analogous to maturity for a person: a mature group knows
very well what it is doing. This idea is analogous to the
syntality definition which refers to the final performance

of the group as a group (Cattell, 1948). It is necessary

24

25

to examine the ways in which individual maturity has been
conceptualized to determine whether the analogy between group
and individual maturity seems reasonable.

Developmental psychology according to the organismic
theory of Heinz Werner (1957) postulates a single regula—
tive principle. His orthogenetic principle states:

wherever development occurs it proceeds from a state

of relative globality and lack of differentiation

to a state of increasing differentiation, articula—

tion, and hierarchic integration (Werner, 1957, p.

126).

Mature organisms are characterized by the development of a
stability of structure while maintaining the capability of
reorganizing to adapt to a changing environment.

The developmental theories of Piaget (1932), Freud
(1933), and Erikson (1950) rest upon a similar principle
in discussing ways in which the individual adapts to his
environment. Evidence for this concept of maturity can also
be found in Heath's (1965) studies of mature and immature
college men.

Offer and Sabshin (1966) and Jahoda (1958) have
reviewed the literature on concept of mental health, which
is often treated as a synonym of maturity. Offer and
Sabshin synthesized the literature into four different
approaches: (1) Normality as health, (2) Normality as the
average, (3) Normality as utopia, and (A) Normality as

process. Only the last two are consonant with the ideas of

the maturity concept as used in developmental psychology.

26

Normality as utopia is conceived of as some state of self
actualization or as the utilization of full potential
(Maslow, 1954; Rogers, 1961). Normality as process is con-
sidered the end result of a set of interacting systems that
change over time (Werner, 1957; Piaget, 1932; Freud, 1933;
Erikson, 1950).

Jahoda described positive mental health as a multi—
faceted concept. She identified six positively related
elements in her definition:

1. Attitudes of the individual toward himself.

2. The style and degree of individual growth,
development, or self actualization.

3. Integration.

A. Autonomy.

5. Perception of reality.

6. Environmental mastery.

The construct of individual maturity described by these
various theorists seems to be closely related to the ideas
of group maturity in the work group of Bion, the integrated
group of Schroder and Harvey, and the enduring set of group
properties described by Jacobson. The analysis provided by
Jahoda establishes a framework for operationalizing the
concept of group maturity.

A reexamination of end states defined in the models
of group development in relation to the constructs of
individual maturity provides a construct of group maturity

conceptualized as a composite of three factors:

27

1. The problem solving process of the group.

2. The cohesiveness of the group.

3. The existence of shared norms within the group.

In all the group development models the problem
solving capability of the group represents the group's
total functional integration and its ability to deal with
the environment. Cohesiveness and shared norms within the
group provide the dual capabilities of attaining stability

while maintaining the flexibility for adaptation.

Stages in Group Development

h A key concept derived from the theorists discussing
individual maturity is that there exist critical periods
during which particular behaviors are acquired and par-
ticular relationships are important. Each developmental
stage of the individual revolves around a critical problem
to which all the individual activities are directed. For
example, in Freud's schema the critical factor during the
phallic stage is the resolution of the Oedipal complex.

The same notion of critical periods can be applied to
group development. During the developmental stages of a
(group, there exist critical periods for each of the processes
involved. Group development differs from individual develop-
ment in the sense that groups do not master one process or
set of behaviors before they can begin the next stage. Each
group process interacts with other processes and contributes

to the development of a group at all stages. Thus one must

28

keep in mind that each process has its own critical period
in the development of group maturity. This does not mean
that the other processes do not contribute to group matur-
ity at that stage but that the one particular process is
more important at that particular stage.

In the very beginning of the group the most important
process is identification. Identification comes first
for many reasons. First, identification is not a group
process as such. It describes the individual's relation
to some other person, persons, or things. This process
involves the incorporation of some need for the group
within the individual. Some degree of identification with
the group among the members is necessary for the survival
of the group (Zalesnik & Moment, 1964). In addition, the
type and degree of identification in the group determines a
direction or pattern which the other processes will follow
in their development.

All the groups described by the theorists mentioned
had to go through some type of initiation and orientation
phase. This is a period of defining the group situation for
the individual. The descriptions of this stage fit together
into Jacobson's idea of identification in which the indivi—
dual discovers if his own interests and needs are congruent
with the programs and goals of the group.

Once some degree of psychological acceptance of the

group is established among the group members the

29

interpersonal structure of the group takes primary importance.
The second critical period involves a redefining of the
group situation and the beginning of the differentiation
of the group's structure. This period is charaCterized by
the development in the group of the performance of the
appropriate task behaviors by the members of the group.
These are a modification of the general task behaviors
described by Benne and Sheats (19A8) that any group needs
to perform and not merely related to specific performance
of their organizational tasks. This stage is a combina-
tion of the ideas of Jacobson in his second and third
stages.

In the development of the group the next critical
factor is the communication structure. The communication
patterns established within the group serve as the con-
necting link between the further differentiation of the
group and the integration of the group.

Finally, leadership practices are the critical process
at stage four. The type of leadership practices referred
to here is the total group's leadership practices. The
distribution of leadership activities across the group
builds upon and expands the effectiveness of the structural
aspects of the group. As leadership functions become more
widely distributed within the group, the group approaches
the condition described by Likert and sensitivity group
theorists as one in which the group can realize its full

potential.

30

Each of these four processes are continually inter-

acting with each other as a group moves toward maturity.

One process contributes to and builds upon efforts to

improve the other. Each stage of group process is asso-

ciated with some degree or level of group maturity. The

changes in the processes of the group are linked to in—

creases in the group's total level of maturity.

From these ideas the following hypotheses are derived

to state the relationship between the elements of group

process and group maturity.

1.

Identification is more strongly related to group
maturity at level one than are role differentiation,
communication practices, and leadership practices.

Identification is more strongly related to group
maturity at level one than it is at levels two,
three, and four.

Role differentiation is more strongly related to
group maturity at level two than are identifica-
tion, communication practices, and leadership
practices.

Role differentiation is more strongly related to
group maturity at level two than it is at levels
one, three, and four.

Communication is more strongly related to group
maturity at level three than is identification,
role differentiation, and leadership practices.

Communication is more strongly related to group
maturity at level three than it is at levels one,
two, and four.

Leadership practices are more strongly related to
group maturity at level four than are identifica-
tion, role differentiation, and communication.

Leadership practices are more strongly related to
group maturity at level four than they are at
levels one, two, and three.

31

A diagram of the relationship among the elements is

presented as Fig. 1.

Levels of Maturity

l 2 3 A

q
4

I - Identification

T — Task Development

C - Communication

L - Leadership Distribution

 

Fig. l. Preportion of group maturity explained by the ele—
ments of group process at each level of maturity.

The idea expressed in these hypotheses is that all of
these processes contribute to expanding the degree of
maturity of the group over time. However, each process has
a particular period in which it makes its greatest contri-

bution to the group's maturity.

CHAPTER III

THE WORK TEAM CONCEPT AT THE RESEARCH SITE

The site Of this research is a manufacturing Operation
which employs about A50 perple in four divisions.

The company installed the Scanlon Plan in 1952 with the
aid of Dr. Carl Frost. The Scanlon Plan is a theory or
phiIOSOphy Of management, which consists of a set Of assump-
tions about human motivation and behavior, some general
principles concerning the effective management of organiza-
tions based upon these assumptions, and some specific
structural recommendations for the implementation Of these
principles in ongoing organizations. The principles of the
Scanlon Plan stress the need for: (l) a system for sharing
with the employees information about the organization's
Objectives, plans, and problems; (2) a system for increasing
organizational effectiveness through participation by the
employees in decisions that affect them and their jobs; and
(3) a system for establishing a reward structure and leader-
ship practices that foster cooperation.

These three principles generally are implemented through
a network Of production committees at the department level,

a screening committee at the divisional or corporate level,

32

33

and a monthly bonus based upon labor efficiency. The pro-
duction committees meet to discuss ways Of improving the
efficiency Of the department and to solve current production
problems. The screening committee acts on suggestions
referred to it by the production committee and reviews the
bonus performance for the month. For every month in which
the organization produces their products at less cost than
the established standard labor cost, a monthly bonus is pro-
vided to employees for their part in improving the labor
efficiency of the organization. The bonus is distributed to
everyone in the company from the tOp to the bottom as a
percentage Of their salary. A more thorough discussion of
the Scanlon Plan can be found in McGregor (1952), Lesieur
(1958), and Doyle (1970).

Over the past twenty years a multitude Of behavioral
science inputs have been assimilated into the organizational
management system Of the company where this research was
done through feedback to the organization from consultants
and research projects and through individual training of
the managers. From their experiences, a core Of people
throughout the organization has developed a "self-
consciousness" about the nature of the organization. This
core, through its internalization of the Scanlon Plan
principles, provides the network of teams with the frame—
work for the deveIOpment of group process and the capability

of adaptation to its environment.

3A

In 1967, the company hired a full time organizational
development manager and began to use more extensively a ‘
variety of behavioral science inputs. Several inputs led to
changes in the Operation of their Scanlon Plan.

First, every manager from the president to the first
line supervisor has spent a week at a Managerial Grid train-
ing program (Blake and Mouton, 196A). This training helped
the managers assess their impact upon others and gave them
their first experiences in teamwork management. It also
established a common language for the discussion of
behavioral science concepts.

Second, the company became involved with Rensis Likert
and the longitudinal organizational research project of the
Institute for Social Research at the University of Michigan
(Taylor and Bowers, 1970). Likert's (1961, 1967) conception
of the supervisor as a linking pin in an interlocking network
Of teams throughout the organization stimulated additional
thinking.

The combination Of these two factors, the continued
consultation Of Dr. Frost, and other training programs and
assessments led to modifications in their Scanlon Plan
committee structures over the last three years. Originally
the production committee consisted of two elected representa-
tives from each department and the foreman. The screening
committee included one representative from each production
committee (no foremen) who would all meet with the division

management.

35

Dissatisfaction arose within the organization over the
Operation of these committees. Management felt that the
production committees were passing suggestions along
automatically to the screening committee without taking
action on them. Some production workers were complaining
that their representative was not adequately representing
them because he did not have the same job, and did not know
nor understand their problems.

One of the foremen drew up a plan to cope.with these
problems based upon his experiences at the Managerial Grid
training program. Independently, the organizational develop-
ment manager, stimulated by the publications of Dr. Likert
and by the analysis of the data about participation at the
production level gathered from the University Of Michigan
employee survey and a Scanlon Plan employee survey from
Michigan State University developed a set of recommendations.

From their analysis a new concept of the work group in
the company emerged. Rather than a single workgroup, they
discovered that each foreman actually had three or four
distinct groups. A work team was defined as the group of
people that have a common responsibility for one part of the
organization's work. The work teams throughout the organiza-
tion were identified and established.

The production committee, at this point, became an
entire work team. Each supervisor has 1—3 work teams and

3—1A people per team. The committee meetings shifted from a

36

meeting with representatives to a meeting with the whole
group. Quarterly Division team meetings, including repre-
sentatives from each work team, the foremen, and Division
management have been set up to replace the screening com-
mittee. The function of the work team is to plan the work,
to identify and solve problems, and to evaluate the activi—
ties Of its own team. The Divisional team is involved in
divisional planning, policy making, and coordination and

serves a communication link within the division.

CHAPTER IV

METHODS AND PROCEDURES

Data Gathering Procedures and Research Design

The data for this study were gathered through the joint
efforts of the Midwest Scanlon Associates, Michigan State
University, and the Institute for Social Research (ISR) of
the University Of Michigan. Mr. Steve Iman of the Institute
for Social Research, cooperated in the study, design, and
data gathering.

At the time that this research project was being
develOped, the company studied was an active participant in
the longitudinal employee attitude survey of a number of
organizations being conducted by the ISR at the University
of Michigan.. ISR's cooperation in the data collection phase
of this project was obtained to take advantage of the Oppor-
tunity to use the ISR data that were relevant to the purposes
Of the study.

A cross sectional correlational analysis of the two
classes of variables--group process and group maturity-—was
used. Since the concept of a work team and its use was I
introduced to the whole organization at the same time, it

was assumed that differences in maturity among groups at

37

38

the time Of the study could be attributed to their rates of
development resulting from their type of group process.
Given this assumption and the exploratory nature of this
study, the cross sectional design was used to generate a set
of hypotheses for a more extended analysis of group growth
in the future.

The measures of the group process and the group maturity
variables were Obtained in the fall of 1970 from the Inter
Company Longitudinal Survey (ICLS) of ISR, which was being
given for the third time, and the Heinen survey, which was
develOped specifically for this project. Appendix A contains
a copy of the Heinen survey with its instructions. The
procedures for administering these two surveys were dif-
ferent. For the ICLS, each individual in the organization
was given a questionnaire packet which he took home. The
completed survey was returned several days later tO prede—
termined drop Off points at the plant. During the four
week period after the ICLS surveys were collected the
Heinen survey was completed by the members of the organi-
zation in meetings of their own work teams.

Two additional measures were gathered as validation
data for the group maturity measures. The two measures were
the team's perception of its own growth and the managerial
ratings of the work teams. The measure of the team's per-
ception Of its own growth was contained in the ICLS ques-
tionnaire. The managerial ratings Of the work teams were

gathered six months after the initial data collection.

39

Subjects
The unit Of analysis for this study is the production

level work team. Individual questionnaire responses were
converted into work team indices. The study was limited
to production level teams in order to exclude a possible
source of extraneous variance due to differences in the
primary purpose or function of the work team.

The data from the two separate questionnaires were
matched for each work team. In the ICLS questionnaire the
individual respondent was not identified, so the surveys
were matched in terms of the number of respondents to both
surveys in each work team. If the difference was greater
than one, that team was dropped from the analysis.

Using these criteria twenty of the thirty-two production
teams remained in the analysis (see Table l). Thirteen teams
were from the Division A and seven from the Division B. The
twenty work teams ranged in size from 3 to 11 people and
were managed by eight different foremen with each foreman

having from 1 to A teams.
Definition and Measurement of the Variables

Group Maturitereasures

Cohesiveness (Coh) is the attractiveness of the group
for its members. Operationally the variable will be defined
in terms of the members (I) perceiving themselves tO be part

Of a team, (2) preferring to remain in the team rather than

A0

TABLE 1

Work Teams in Analysis

 

 

 

 

Team Number --Employees in Team Questionnaires Returned
ICLS Heinen
Division A
1 5 A A
2 ll 9 10
3 A 3 A
A A A A
5 6 5 6
6 ll 9 8
7 ll 10 9
8 10 9 9
9 ll 8 9
10 6 5 5
11 6 6 5
12 ll 6 7
13 A A A
Division B
1A 6 A A
15 8 7 8
16 A A 3
l7 5 3 A
18 3 2 3
l9 5 5 5
20 11 7 7

 

Al

leaving it, (3) perceiving their team to be better than
other teams with respect to the way the men get along
together, the way they help each other out, and the way they
stick together (Seashore, 195A).

The problem solving process (PS) measure examined the

 

procedures a team follows in solving its problems. Four

parts Of the procedure were indexed: (l) outlining the

 

problem, which refers to the preparation and planning in

solving the problem; (2) discussing thegproblem, which con-

 

cerns the manner in which the team deals with the problem;

(3) solving the problem, which refers to the type of solution

 

reached; and (A) overallgproblem solving ability, which is

 

an expression Of competence in handling a problem. This
procedure is based upon Maier (1952).

Shared norms (SN) is a measure of the extent of agree-

 

ment among members about how the team ought to behave.
Sixteen normative items were selected on the basis of Ob—
serving the organization and identifying relevant issues

for the work teams. For several months prior to the study,
this investigator discussed work team activities with people
at all levels of the organization. These employees talked
about their own work team as well as their thoughts about
the whole work team operation. Items were selected to
represent at least three broad areas of work team responsi-
bility: (1) the total team responsibilities to individual

members; (2) the domain of activities the teams were engaged

A2

in, both some of those common to all teams and some of those
only a few engaged in; and (3) some activities that might
adequately be handled by production level teams but were now
being handled by higher staff level teams.

Five point Likert scale items were used to measure each
Of these three variables. For the Coh and PS variables, each
individual received a score of 1-5 on each item. For each
variable items were combined to give the individual a mean
Coh score and a mean PS score. Within each work team indi-
vidual means were combined into mean work team scores to
derive the work team's Coh score and PS score.

The shared norm measure was also a five point Likert
scale asking the team member whether the team should or
should not do something. The method of scoring was different
for this. Since amount Of agreement concerning norms was
the key issue, the work team variance was computed for each
of the sixteen normative statements. The work team SN score
was the sum of the item variances for all sixteen items. On
this measure a high score indicated few shared norms and a
low score indicated many shared norms.

A second category of variables served as a type of
criterion measure of the work team's maturity. The first
measure was the team's perception of its own growth (PTG)
and the second was the manager's ratings Of the team's
effectiveness (Mgr.R). PTG consists of five point Likert

scale items filled out by the work team members describing

A3

their own perceptions of the team's progress and develop-
ment.

Mgr.R are paired comparison ratings Of the production
level work teams collected six months after the question-
naire measurements. A copy Of the instructions to the
managers for making the ratings is included in Appendix B.
The production management work teams in each division
rated all the production level work teams in their own
division. In order to combine these two sets of ratings
another group of the managers was used who could rate some
teams in both divisions. Uhrbrock and Richardson's (1933)
key man techniques were used to combine the three sets of
ratings.

Using Guilford's Case III assumptions (195A), three
scales were derived for the three sets of judges. The
chi square probability of a larger value for each Of the
sets was .98, .95, and .91 respectively. The assumptions
of the Guilford Case III method were met. The third set
was used then to combine the other two sets into a single

scale.

Group Process Measures

The identification process (Id) represents the work
team member's relation to the group. Two factors were
assessed in this relationship: (1) his evaluation of the
group as instrumental to his goal attainment; and (2) his
perception Of the value congruence between himself and the

group.

AA

The Id variable was also comprised of five point Likert
scale items. The procedure for establishing a work team Id
score is the same as that for the Coh score and the PS score.

The task development variable (TD) was the number of
role behaviors being performed in the team by members other
than the supervisor. Each work team member was given a list
of twenty role behaviors adapted from the Benne and Sheats
(l9A8) role classification. Group task role behaviors and
group maintenance role behaviors that applied to the work
team situation were used. Certain representative role be-
haviors that referred to activities linking the work team
with other teams were used in place of individual roles.

To complete the questionnaire role items, each person
was asked to check off the person or persons in the team
that were performing each of these twenty behaviors. If
no one performed the activity, respondents were instructed
to check no one; if more than one person performed the
activity, respondents were to check each person who per—
formed the activity.

Since each respondent could check more than one person,
an agreement cut Off score of 50% was established as a basis
for a person being counted as performing a role behavior in
the team. That is, at least 50% or more Of the team mem—
bers must have said person A performed role behavior 1 be-
fore person A would be counted as performing role behavior 1.

Once it was determined who performed which role behaviors,

A5

the number of role behaviors being performed by people other
than the supervisor were counted. Since the possible range
of this score was 0-20, the team's TD score was divided by
four to put it on the same scale as the five point Likert
measures.

The communication measure (Comm) was defined as the
degree of interconnectedness among the members of the work
team. Comm was derived using the quasi-sociometric pro-
cedures developed by Weiss and Jacobson (1955). Each person
recorded the names of the people they countacted regularly
in the company, and indicated the frequency and importance
Of the contact and number of years they had the contact.

The number Of high frequency, high importance recip-
rocal contacts that existed within each work team was
counted. The ratio of the number of these reciprocal con-
tacts to the total number of possible reciprocal contacts
within the team is the Comm measure. The ratio score was
multiplied by five to put it on the same scale with the
other measures. .

Leadership distribution (LD) is defined as the extent
to which all the members of the team are performing the
leadership activities. The measure of leadership.distrir
bution was Obtained from the individual team member's
responses to questions about who performed which role be-
haviors in the work team. For each Of the twenty role be-

havior items, the individual response was recoded into a

A6

1-5 score. The recoding procedure for each item followed
this formula:

- no one performs the behavior.

— the supervisor alone performs the behavior.

the supervisor and one other performs the behavior.
- the supervisor and several others (more than two)
perform the behavior.

the entire work team performs the behavior.

- someone in the team performs the behavior, but

not the supervisor.

.trerI—l
I

OUT
I

After the recoding, two Of the twenty items were dropped
because they were behaviors performed almost exclusively
by one member of the team, who was not the supervisor.
For the remaining eighteen items, each individual received
a mean LD score. The individual LD scores were then com-
bined into a mean LD score for the work team.

Table 2 presents a summary of the sources used in the
measurement of each of the variables in the study. The items
used to measure each Of the variables and an analysis of

their empirical characteristics can be found in Appendix C.

A7

TABLE 2

The Data Sources of Variables used in the Study

 

 

Variable

ICLS

Data Source

Heinen

Other

 

Maturity

Coh
PS
SN

Process

Id
TD

Comm
LD

Criteria

PTG
Mgr. R

><><><

Paired
Comp

 

CHAPTER V

RESULTS

Group Maturity Measures

The first step in examining the relationship of the
type of group process to the stages or levels of group
maturity was to compute, for each work team, an overall
group maturity score. This total score was a composite of
the team's cohesiveness, problem solving, and shared norm
scores. Each team's cohesiveness, problem solving, and
shared norm scores were transformed into Z scores and then
summed. In summing the three scores, the sign Of shared
norm Z scores was reversed because a low score indicated
high shared norms and vice versa. Table 3 lists each team's
Z score for cohesiveness, problem solving, shared norms, and
total. A constant was added to the total score to make the
scores all positive. Among the twenty groups then the low
total score was .01 and the high total score was 6.73 with
a mean total score of 3.38 and a standard deviation Of 2.08.

The total group maturity measure and each of its com-
ponents were compared to the two independent ratings of the
team's development by means Of Pearson product moment cor-

relations (see Table A). The first rating was a measure

A8

“9

TABLE 3

Z Scores of Group Maturity Variables

 

 

 

 

 

 

+p

Problem Shared
Team Cohesiveness Solving Norms Total
1 -0.916 -0.085 —0.0080 -1.009
2 -0.777 -0.659 —0.27A8 -l.7ll
3 -l.293 0.085 0.7151 -0.A93
A -0.559 -0.829 -0.2259 -l.6lA
5 -0.1A2 -0.l9l —l.l670 -l.500
6 0.015 —l.255 -2.1006 -3.3Al
7 -0.A60 —0.212 0.3016 —0.370
8 1.503 1.957 -1.1A18 2.318
9 —0.976 —0.617 1.1286 -0.A6A
10 0.988 0.7AA 0.9596 2.692
11 -0.380 0.021 1.0960 l.A55
12 0.531 0.021 -0.ll92 0.A33
13 1.186 0.553 -l.952A -0.213
lA 1.305 0.872 1.20A2 3.381
15 -l.730 —l.A89 0.7566 —2.A62
16 -0.559 0.553 1.0530 1.059
17 1.027 1.553 0.6A69 3.227
18 1.682 -l.510 0.030A 3.222
19 0.095 -l.297 -0.3726 —l.575
20 —0.519 —l.23A -0.5297 -2.283
TABLE A
Correlations Among Group Maturity Measures
and the Ratings Of the Work Team
Total Coh PS SN
PTG .69* .69* .80* .09
Mgr. .20 .36 .A2+ .36
*p

50

Of the team's own perception of its development (PTG) and the
second measures were ratings by the managers Of the team's
development (Mgr. R).

Total group maturity, cohesiveness, and problem solving
were all significantly correlated with perceived team growth,
but none of the measures were significantly related to the
managerial ratings. The problem solving measure approached
a significant correlation with the managerial ratings

(p < .10).

Relation Of Group Process to Group Maturity

 

A 2 x 2 treatments x levels repeated measures analysis
of variance was used to test the hypotheses relating group
process variables to group maturity. The twenty production
level work teams were rank ordered on their total group
maturity score and divided into four levels of five work
teams each. For each work team the repeated measures were
their identification, task development, communication, and
leadership distribution scores. Separate analyses of
variance were also computed for each of the components Of
the group maturity measure. For each component the twenty
production level work teams were again rank ordered and
divided into four levels of five work teams. For all four
analyses, each of the levels was not always composed of the
same block of five teams. Some differences in the rank order
of the maturity measures determined the shifts in level of

a particular group.

Tables 5—8 present the AOV tables for each of the group
maturity measures. Tables 9—12 present the mean summary
tables for each of the AOV tables. For each of the AOV's
there was only a singificant difference between types of
group process. There were no significant differences
between levels of maturity nor any significant interaction
effects. With respect to the types of group process the
highest mean was the identification process, With the task
development and leadership distribution process about equal,
and the communication process the lowest. An inspection
Of Tables 9-12 shows that this same pattern among the group
process variables exist within levels Of maturity as well as
across all four levels.

Each work team's identification score, task develop-
ment score, communication score, and leadership distribu-
tion score were transformed into a proportion of the team's
total group process score. One way AOV's of these propor-
tions within each group process type across the four levels
of group maturity were used to determine the level Of
maturity at which the process has its greatest relationship.
Tables 13—16 have the one way AOV's for each of the group
process types with the four different measures of group
maturity. Table 17 contains the mean summary tables for
the proportions of group process x group maturity measures.
The tables show that the only significant differences are

between shared norms and task development. Task

52

TABLE 5

AOV of Group Maturity Total Score x Group Process

 

 

 

 

Source SS df MS F
Between teams AA.5771 19
A (Levels of Maturity) 8.02A2 3 2.67A7 1.1708
Teams wn. blocks 36.5529 16 2.28A6
Within teams 62.37A8 60
B (Group Process) 30.6320 3 10.2107 16.1128*
AB 1.3239 9 .1A71 .2321
B x teams wn. blocks 30.A189 A8 .6337
TABLE 6

AOV of Group Maturity Cohesiveness Score x Group Process

 

 

 

Source . SS df MS F
Between teams AA.5771 19
A (Levels of Maturity) 6.7126 3 2.2375 .9A55
Teams wn. blocks 37.86A5 16 2.3665
Within teams 62.37A8 60
B (Group Process) 30.6320 3 10.2107 16.6760*
AB 2.3506 9 .2612 .A266
B x teams wn. blocks 29.3922 A8 .6123

 

*p r .01

53

TABLE 7

AOV of Group Maturity Problem Solving Score x Group Process

 

 

 

 

Source SS df MS F
Between teams AA.5771 19

A (Levels of Maturity) 8.7912 3 2.930A 1.3102
Teams wn. blocks 35.7859 16 2.2366

Within teams 62.37A8 60

B (Group Process) 30.6320 3 10.2107 18.1A59*
AB A.7323 9 .5258 .93AA
B x teams wn. blocks 27.0105 A8 .5627

TABLE 8

AOV of Group Maturity Shared Norm Score x Group Process

 

 

 

Source SS df MS F
Between teams AA.5771 19
A (Levels Of Maturity) 1.6028 3 .53A3 .1989
Teams wn. blocks A2.97A3 16 2.6859
Within teams 62.37A8 60
B (Group Process) 30.6320 3 10.2107 17.90Al*
AB A.366A 9 .A852 .8508
B x teams wn. blocks 27.376A A8 .5703

 

*p < .Ol

5A

TABLE 9

AOV Mean Summary of Group Maturity
Total Score x Group Process

 

 

 

 

 

 

 

Id TD Comm LD Total
Level 1 15.93 13.00 7.10 11.80 A7.83
Level 2 15.32 12.00 7.70 12.30 A7.32
Level 3 18.35 17.00 10.70 16.76 62.81
Level A 19.1A 1A.25 8.70 13.30 55.39
Total 68.7A 56.25 3A.20 5A.l6 213.35

TABLE 10

AOV Mean Summary Of Group Maturity

Cohesiveness Score x Group Process

Id TD Comm LD Total
Level 1 15.11 1A.50 6.A0 12.96 A8.97
Level 2 16.92 13.25 8.85 13.68 52.70
Level 3 16.80 12.25 6.60 13.0A A8.69
Level A 19.91 16.25 12.35 1A.A8 62.99

Total 68.7A 56.25 3A.2O 5A.l6 213.35

 

AOV Mean Summary of Group Maturity Problem

55

TABLE 11

Solving Score x Group Process

 

 

 

 

 

 

 

Id TD Comm LD Total
Level 1 15.83 11.50 8.70 11.00 A7.03
Level 2 15.7A 1A.00 A.A0 l3.A9 A7.63
Level 3 18.03 16.50 l2.A0 16.37 63.30
Level A 19.1A 1A.25 8.70 13.30 55.39
Total 68.7A 56.25 3A.20 5A.20 213.35
TABLE 12
AOV Mean Summary of Group Maturity Shared
Norm Score x Group Process
Id TD Comm LD Total
Level 1 18.10 10.25 7.90 12.51 A8.76
Level 2 15.A8 16.25 7.80 13.81 53.3A
Level 3 17.09 15.50 8.95 13.52 55.06
Level A 18.07 1A.25 9.55 1A.32 56.19
Total 68.7A 56.25 3A.20 5A.16 213.35

 

56

TABLE 13

One Way AOV's of Proportions of Group

Process x Total Group Maturity

 

 

 

 

 

 

 

Source SS df MS F
Id

Levels .01A8 3 .00A9 .3952
Error .1979 16 .012A
TD

Levels .0017 3 .0006 .1111
Error .0868 16 .005A

Comm

Levels .0015 3 .0005 .0A59
Error .1737 16 .0199
L0

Levels .0025 3 .0008 .615A
Error .0209 16 .0013
TABLE 1A
One Way AOV's of Proportions Of Group
Process x Cohesiveness

Source .88 df MS F
Id

Levels .0061 3 .0020 .1550
Error .2066 16 .0129
TD

Levels .0135 3 .00A5 .957A
Error .0750 16 .00A7

Comm

Levels .0056 3 .0019 .1792
Error .1696 16 .0106

LD

Levels .0029 3 .0010 .7692
Error .0205 16 .0013

 

57

TABLE 15

One Way AOV's of Proportions Of Group

Process x Problem Solving

 

 

 

 

 

 

 

Source SS df MS F
Id
Levels .0189 3 .0063 .5207
Error .1938 16 .0121
TD
Levels .0109 3 .0036 .73A7
Error .0776 16 .00A9
Comm
Levels .02A0 3 .0080 .8A21
Error .1512 16 .0095
LD
Levels .0063 3 .0021 .9091
Error .0171 16 .0011
TABLE 16
One Way AOV's of Proportions of Group
Process x Shared Norms
Source SS df MS F
Id
Levels .0530 3 .0177 -7700
Error .1597 16 .0100
TD
Levels .0396 3 .0132 .2581*
Error .0A89 16 .0031
Comm
Levels .0036 3 .0009 .0833
Error .1726 16 .0108
LD
Levels .0003 3 .0001 .071A
Error .0231 16 .001A

 

*p ,«

AOV Mean Summary Of Proportion Of Group
Process x Group Maturity

58

TABLE

17

 

 

 

Id TD Comm LD

Total

Level 1 1.8887 1.2283 .6A88 1.23A5
Level 2 l 6662 1.2671 .7560 1.3109
Level 3 1.5399 1.3509 .7333 1.3759
Level A 1.8228 1.2513 .6769 1.2A89
Coh

Level 1 1.602A l.A52A .6119 1.3335
Level 2 1.8A27 1.12A2 .7362 1.2970
Level 3 1.7500 1.2620 .6A15 1.3A68
Level A 1.7225 1.2590 .825A 1.1929
PS

Level 1 1.9001 1.1153 .8051 1.1799
Level 2 1.7019 l.AA31 .A362 l.Al88
Level 3 l.A928 1.2879 .8968 1.3226
Level A 1.8228 1.2513 .6769 1.2A89
SN

Level 1 2.1A83 .9351 .61A7 1.3020
Level 2 l.A670 1.5310 .6938 1.3082
Level 3 1.5862 1.398A .7593 1.2563
Level A 1.7161 1.2331 .7A72 1.3037

 

59

develOpment's highest mean proportion of group process
measure occurs in level two of the shared norms maturity
measure.

None of the AOV results support any of the eight
hypotheses. A close inspection of Table 17 shows that for
the shared norms and problem solving measures but not the
total or cohesiveness measure, the identification, task
development, and communications measures have their highest
proportions of total group process in the appropriate
levels. However, the differences between levels is non-
significant.

Another estimate of the relationship between the group
process variables and the group maturity variables was
obtained through the Pearson product moment correlations of
the variables with each other. Table 18 gives the correla—
tions between the group maturity measures. Cohesiveness
and problem solving were strongly related to each other,
but neither was correlated with shared norms. Table 19
shows that task development, communication, and leadership
distribution were all highly correlated with each other,
but none of them were correlated with identification.
However, identification was the only variable among the
group process measures that was correlated with any of the
group maturity measures (see Table 20). It was correlated
with the total, cohesiveness, and problem solving maturity

measures, but not shared norms.

60

TABLE 18

Correlations Between Group Maturity Variables

 

 

 

 

Coh PS SN
Coh 1.00
PS .7A** 1.00
SN .2A -.1A 1.00
TABLE 19

Correlations Between Group Process Variables

 

 

 

 

Id TD Comm LD
Id 1.00
TD .28 1.00
Comm .30 .65** 1.00
LD .39 .8A** .60** 1.00
TABLE 20

Correlations Between Group Maturity
and Group Process Variables

 

 

 

Id TD Comm 7 LD
Total .66** .19 .1A .25
Coh .76** .12 .19 .17
PS .69** .21 .12 .3A
SN .11 -.O8 .01 .OO

 

**p < .01

CHAPTER VI

DISCUSSION

According to the model proposed in this dissertation,
the group processes of identification, task development, com—
munication, and leadership distribution, each have a critical
period of importance in the development of an industrial work
team's maturity. Identification is the most important pro-
cess for the group's development in stage one, task develop—
ment in stage two, communication in stage three, and leader-
ship distribution in stage four. The model is not directly
supported by the results of the analyses performed. In the
analyses of variance, the proportion of error variance was
always relatively high. Given the small number of teams and
some of the methodological procedures used in this pilot
study, there were many opportunities for the error variance
to increase. Some of the methodological difficulties of this

study will be discussed.

Group Maturity Measures

The group maturity measure was described as being com-
posed of three components, problem solving, cohesiveness, and
shared norms. However, only problem solving and cohesiveness

correlated highly with each other and with other variables;

61

62

the shared norms component does not. The difference between
the first two components and the last is in the procedures

of measurement. Problem solving and cohesiveness were five
point Likert scales upon which the individual respondent
marked the degree to which he felt that the particular stim—
ulus element existed in the team. Thus a high maturity score
would reflect team members' perceptions of high amounts of
these variables existing within the team.

An analysis of the shared norms measure indicates that
it may have been possible for a team to get a high score on
shared norms that, in fact, did not reflect high maturity of
the team. In filling out the questionnaire, each individual
in the team indicated his opinion about what things the team
should or should not do. The sharedness of the norms was
derived from the variance among their responses. It-is
possible that a low variance score could have been obtained
among a set of people entering the team with similar values
about organizational behavior without any knowledge of
others' viewpoints as well as among a team who, through
continued interaction, have arrived at a formal acceptance
of an explicit set of rules for group behavior that express
their group value system. In terms of governing the behavior
of the team, these two different types of shared norms would
produce very different effects. According to the theory of
group development, the first type described would not be

classified as having shared norms, but rather as having a

63

correspondence among individual values because the norm
sharing was not part of the group process and did not regu-
late group behavior. It is not the contention here that
individual value patterns do not affect the behavior in the
team. However, it is the contention that the total team's
behavior will only be affected by individual value patterns
as the various members of the team become aware of each
others' values and translate them into the team's value
system through continued interaction.

In order to discover the norms of a particular group,
four types of information are necessary: (1) the indivi—
dual's position on an issue to indicate his own values;

(2) the individual's perception of others' positions on an
issue to indicate the perceived agreement between group
members; (3) some information about the accuracy of group
members' perceptions of each other; and (A) an indication of
the saliency or relevancy of particular issues to the group.
The only type of information directly available in this
study concerns the individual's own perception on certain
normative issues. The measure of shared norms used may or
may not include the other types of information.

Using these four types of information, the development
of norms within a group might follow a pattern such as:

(l) the individual members have opinions on certain issues,
but no knowledge of the other members' opinions on the same

issues nor any indications of the importance of these issues

6A

to the other members; (2) the individual members begin to get
some indication of others' opinions on selected issues, but
not necessarily a very accurate reflection of all the others'
Opinion; (3) through the group process each of the members
gets a very accurate understanding of the group's position

on a few salient issues; and (A) further group process

leads to the internalization of a set of group values that
gives the members the basis for stating and following the
group's position on not only the most salient issues for the
group, but also new issues that the group has not encountered
or is facing the first time. Since a high group maturity
score, using the shared norms measure, could be obtained for
a group at any of these developmental stages, it is very
probable that the low order correlations between shared norms
and the other components of maturity are a result of this
confounding.

Two independent measures of the development of the group
were used as validation measures of the group maturity
measures. The correlations were high and significant be—
tween the team's own perception of its growth and the total
group maturity, cohesiveness, and problem solving; but were
lower and nonsignificant between the management ratings and
the group maturity measures. The correlation between the
managerial ratings and problem solving did approach signifi-
cance (p < .10). Since the cohesiveness and problem solving

variables are based upon the team's perception of itself, it

65

would be expected that the correlations would be higher with
the team's own ratings than with the managerial ratings.

Two other factors may account for the differences in
the pattern of correlations with the two criterion measures
of group maturity. Perceived team growth, cohesiveness, and
problem solving are all variables measured by a set of atti-
tude statements about the group contained in the same ques-
tionnaire. The correlations among these variables may be
increased by a general halo of positive attitude toward the
group expressed in the respondents' answers to each question.

The nonsignificance of the correlations among the
managerial ratings and the group maturity measures might be
accounted for by measurement error in the managerial ratings.
Since no manager could rate all the teams, three sets of
raters were used to combine the team ratings of the other
two groups of raters in the manner described in the methods
chapter. With the small number of teams the combination of
the three sets of paired comparison ratings may have pro-
duced some additional error in the resulting scale values
for each of the work teams.

The methodological problems encountered with the group
maturity measures suggests the need for additional strong
behavioral criterion measures of group maturity for valida-
tion purposes.

In addition to the problems concerning the construct of

group maturity, there were another set of problems

66

encountered in the statistical analyses. The theory pre—
dicted an interaction between levels of maturity and types
of group process. The requirements of the theory and of the
analysis of variance design demanded that the twenty teams

be divided into four levels of five teams on the basis of
their group maturity scores. All twenty teams were rank
ordered on each of the four maturity measures. The distri-
bution of team scores on each maturity measure was separated
into a level after every fifth team. The actual distribution
of scores did not lend themselves to these types of
divisions. Across the three components and the total matur-
ity scores, the natural breaks in the distribution varied.
For all four measures, the difference between the teams
separating two levels of teams was always smaller than the
differences between one pair of teams within the same level.
For instance, the difference in the total maturity score
between the two teams, which are rank ordered on total
maturity ten and eleven and which divide level two and level
three, was .09; whereas the difference between the two teams,
which are rank ordered on total maturity ten and eleven and
which both fall within level three, was .16. There always
existed within each level at least one pair of teams which
had a greater difference in their maturity scores than the
pair of teams dividing the two adjoining levels. The distri—
butions of each of the component maturity measures seem to

fit more readily into three levels than four.

67

Another potential problem in the measurement of group
maturity is the representativeness of this sample of twenty
teams to the total hypothetical population of industrial
work groups on a maturity dimension. Acting as if these
twenty teams were in fact representative of all the levels
of maturity of the total population of groups is probably
a false assumption. From the distribution of group maturity
scores and from observations of the teams' behavior, a more
realistic assumption would be that these twenty teams form a
relatively homogenous subset of teams that has been sampled
from the hypothetical population of teams at some relatively
low level on an absolute maturity scale.

Within the limits of this dissertation, the actual
representativeness of this sample can not be determined.
Future research must not only be directed towards further
development of the construct of group maturity, but also the
collection of normative data on the maturity level of groups

over time.

Group Process Measures

 

Turning now to the group process variables, there were
some similar problems in the measurement of the variables and
the statistical analyses. The pattern of correlations among
the group process measures showed that the task development,
communication, and leadership distribution measures were
highly interrelated, but not with the identification measure.

Theoretically this pattern can be understood. The three

68

related variables were all measures of the structural pro-
cesses of a group; the identification variable was a measure
of the individual's relation, acceptance, and involvement
with the group.

The operational independence of the task development
variable and the leadership distribution variable is ques-
tionable. Essentially the same data is summarized in
slightly different ways. The task development variable uses
the matrix of who is performing what behaviors in the team
and summarizes how much is being done by the team members.
The leadership distribution score recodes each individual's
perception of how much of the team is performing each leader—
ship role and summarizes that for the entire team. Since
practically every role behavior was a leader role behavior,
the two measures result in very similar values.

Differences in the four process variables were treated
as four repeated measures of group process. The identifica-
tion score was derived from a summated set of five point
Likert scale items. The communication score was a ratio of
the actual reciprocated sociometric choices to the possible
number of these choices. The task development score
involved the counting of the number of role behaviors being
performed in a team. The leadership distribution score was
a summated score of team member reports of how many team

members were performing each role.

69

Multiplicative transformations were used on the communi-
cation and task development scales to make the variances
approximately equal and the scale values one to five. In
fact, with both these variables the scale values ranged
from zero to five. The zero points of these four scales
do not appear to be equivalent, therefore the comparability

of the four scales is questionable.

Relation of Group Process to Group Maturity

 

The only relationship found between the group process
and group maturity measures was the correlation between
identification and the cohesiveness, problem solving, and
total group maturity measure. Since the identification
measure was also an attitude scale measure contained in the
same Likert questionnaire as were the cohesiveness and prob-
lem solving scales, there is a possibility that a common
halo effect could have contributed to the correlation.
However, the cluster analysis procedures give some evidence
that these are independent but correlated variables.

Taking into account the previous discussion of the
various types of method error variance that might have
influenced this study, another explanation for the findings
might be offered. If this sample of teams does in fact
represent a set of teams in the early stages of development,
then the pattern of results conforms to the theory. Among
the group process measures, the set of twenty teams have

relatively low scores on the structural variables of task

70

development, communication, and leadership distribution.
The identification scores for the entire set of teams were
relatively higher. In examining the correlations with the
group maturity measures, the higher correlations with iden-
tification and the relatively low correlations with the
structural measures would conform to the theoretical pre-
dictions for groups in the initial stage of development.
The non zero correlations among the group maturity
measures and the structural measures of group process can
be explained in the following terms: (1) some teams are
beyond stage one and (2) each of the processes operates to
some extent in every stage. If the above assumption is
correct, then the results of this study do support the

theory indirectly.

Comments About the Theory

 

The inconclusiveness of the results of this study sug—
gest an examination of the theory. The theory hypothesizes
that the four processes of identification, task development,
communication, and leadership distribution each have a
critical period in the development of an industrial work
group. The theory, however, does not stipulate what actually
occurs within each stage. Neither description of the charac-
teristics of the critical process or the relationships among
the processes within a specific stage, nor a description of

the changes in the processes from one stage to the next is

71

explicit in the theory. This impreciseness in the theory
may account for some of the problems in the data analysis.
A diagram of the theoretical model was presented in
the form of a series of pies (see Fig. 2). The diagram
illustrated the proportion of group process related to a
given level of maturity. In the pie diagram the whole pie
represented the total amount of maturity of a group at some
stage of its development, but the slices referred to the
proportion of group process associated with group maturity.
Neither the theory nor the diagram indicated whether the
total amount of group process increases, decreases, or re-
mains the same as the group matures. At first, one assumes
that the amount of group process increases in a direct
linear relationship with maturity; however, the theory only
discusses the changing relationships between each of the
group process variables and group maturity as maturity

increases.
Levels of Maturity

1 2 3 A

V L 1
4 .

 

 

 

I - Identification

T - Task Development

C - Communication

L - Leadership Distribution

Fig. 1. Proportion of group maturity explained by the ele-
ments of group process at each level of maturity.

72

Using the numbers 1 to 5, with 1 representing a low
score and 5 representing a high score, two alternative
examples of data can be given that would confirm the idea
that each group process has a critical period in the develop-
ment of the group, but would yield very different conclu-
sions about the relationship among the four processes in the
development of the group.

Example 1.

Level of Maturity

1 2 3 A
Group Processes
Id A.00 3.00 3.00 1.00
TD 3.00 A.00 2.00 3.00
Comm 2.00 2.00 A.00 2.00
LD 1.00 1.00 1.00 A.00

Example 2.

Level of Maturity

1 2 3 A
Group Processes
Id A.00 A.00 A.25 5.00
TD 3.00 A.25 A.25 5.00
Comm 2.00 2.75 A.50 5.00
LD 1.00 1.75 2.75 5.00

In Example 1. the total amount of group process remains the

same at each of the four levels, whereas in Example 2. the
total amount of group process increases from one level of
maturity to the next. The two examples also have different

ways of indicating the contributions of the critical process.

73

In Example 1. the critical process always has a higher score
and a higher proportion of total group process in its char-
acteristic stage than the other processes have. However,
this means that in Example 1. each process has a lower score
in each of its non critical stages. For instance, identifi-
cation would have to decrease as the group matures for the
model of Example 1. to be true. Although such an event is
possible, it does not seem to follow logically.

The second example defines the critical process in
terms of the amount of change in the process variable from
one stage to the next. According to this example, all four
group processes increase as the group matures, yet the
increase from one stage to the next is the greatest for the
critical process. This model resembles the developmental
pattern of many child behavior theorists and the pattern
Jacobson (1956) described among the voluntary organization
task groups.

This discussion of the group's development presumes
that the changes in group process are ones of degree.
Another alternative that would conform to the theory could
be the hypothesis that the kind of identification, task de-
velopment, communication, and leadership distribution would
change from one stage to the next. The descriptions of
groups developing in the literature tend to support the idea
that changes in levels of group maturity are associated with

qualitative changes in group process.

7A

This type of conception of the group's development
pOints to another basic distinction between the model
developed in this study and the one presented by Tuckman.

The model used here emphasizes that each stage in the devel-
opment of a group is characterized by a particular critical
process variable; the Tuckman model identifies a certain
theme of activity associated with each stage in the develop—
ment of a group. A combinatiOn of the two approaches may

be helpful. Group development may be characterized by
different qualitative patterns of group process with a spurt
of activity in one process at each stage.

Future research should be directed at articulating the
characteristics of each of the group processes within a stage
and the changes in group process from one stage to the next.
The first step may require the observation of a set of work
groups over time to describe their pattern of development.
This description should concentrate upon identifying exactly
what happens within a group and not upon searching for evi-
dence that confirms or disconfirms a particular theory. The
emphasis used here was upon structural variables. Some other
group variables not discussed in this model that might affect
group's development were the leadership style of the formal
leader, the group atmosphere, and the group goals. The des-
criptions of the group's behavior should attempt to identify
what variables are actually relevant to group development,
and also what type of changes occur among these variables

over time.

75

Another step in the explication of the theory involves
the distinction between group maturity and group processes.
The construct of maturity must be clearly defined so that the
patterns of group process associated with the development of
maturity can be distinguished from the patterns of group
process not associated with development of maturity.

After these aspects of the theory have been developed,
one can turn to such questions as:

1. Are there different developmental patterns for

groups with different pruposes or different kinds

of task?

2. Will increases in group maturity always result in
better achievement of organizational goals?

3. What happens to a group that follows a different
sequence in its development or stops somewhere in
its development?

These are only a few examples of the type of question that

might be asked.

Summary

This study was designed to provide some preliminary in-
formation with which to build a model of the development of
an industrial work group. The limited amount of empirical
data available about the development of an industrial work
group made necessary the exploratory nature of this dis—
sertation. The data gathered in the study pointed out prob-
lems in both the measurement and articulation of the theory

and suggested some directions for future research.

BIBLIOGRAPHY

Bach, G. R.

Ronald Press,

Bales,

the study of small groups.

R. F.

Intensive group psychotherapy.<

BIBLIOGRAPHY

New York:

 

195A.

Interaction process analysis:

a method for

 

 

Wesley,

Bales,
T.

R. F.
Parson,

1950.

 

1953, pp. 111-161.

Bales,
York:

Bales, R. F.
solvin

1951,

Benne,

Bennis, W.
ment.

Bion, W. R.
19A8,

Bion, W. R.
19A8,

Bion, W. R.
1949,

Bion, W. R.
19A9,

Bion, W. R.
1950.

Bion, W. R.
1950,

R. F.

g.
16.,
K. D.
members.

3’

The equilibrium problem in small groups.
R. F. Bales,
in the theory of action.

and E. A.
Glencoe,

Personality and interpersonal behavior.

Cambridge, Mass.:

Addison-

In
Shils, Working papers

 

111.: Free Press,

New

 

Holt,

Rinehart and Winston,

and Strodtbeck, F. L.
Journal of Abnormal and Social Psychology,

1970.

Phases in group problem

 

A85-A95.

and Sheats,

P.
Journal of Social Issues, 19A8, A, A1-50.

Functional roles of group

 

G. and Shepard, H. A.
Human Relations, 1956, 9, A15-A37.

 

Experiences

i, 314-320.

Experiences

1, u87-A96.

Experiences

g, 13-22.

Experiences

295-303.

Experiences

;, 3—1u.

Experiences

3. 395-Ao2.

in groups:

in

in

in

in

in

groups:

groups:

groups:

groups:

groups:

76

A theory of group develop-

1. Human Relations,

 

 

 

II. Human Relations,
III. Human Relations,
IV. Human Relations,

 

V. Human Relations,

 

VI. Human Relations

 

77

Bion, W. R. Experiences in groups: VII. Human Relations,
1951, A, 221—227.

 

Blake, R. R. and Mouton, Jane, S. Managerial Grid. Houston,
Texas: Gulf Publishing, 196A.

 

Burns, T. and Stalker, G. M. The management of innovation.
Chicago: Quadrangle, 1961.

 

Cartwright, D. and Zander, A. Group dynamics. New York:
Harper and Row, 1968.

 

Cattell, R. B. Concepts and methods in the measurement of
group syntality. Psychological Review, 19A8, 55,
u8-63. —’

 

Doyle, R. J. A new look at the Scanlon Plan. Management
Accounting, 1970, 52 (3), A8-52.

 

 

Erikson, E. H. Childhood and society. New York: Norton,
1950.

 

Freud, S. New introductory lectures on psychoanalysis.
In Standard edition, Vol. 22, 1933.

 

Guilford, J. P. Psychometric methods. New York: McGraw-
Hill, 195A.

 

Hare, A. P., Borgatta, E. F., and Bales, R. F. (Eds.) Small
groups: studies in social interaction. New York:
Knopf, 1966.

Heath, D. H. Explorations of Maturity. New York: Appleton-
Century—Crofts, 1965.

 

Heinicke, C. and Bales, R. F. Developmental trends in the
structure of small groups. Sociometry, 1953, 16 (1),
7-38.

 

Homans, G. D. The human group. New York: Harcourt, Brace
and World, 1950.

 

Hunter, J. and Cohen, S. Package: a cluster analysis pro-
gram. Unpublished manuscript. Michigan State
University, 1970.

Jacobson, E. The growth of groups in a voluntary organi-
zation. Journal of Social Issues, 1956, 12 (2), 18—23.

 

Jahoda, Marie. Current concepts of positive mental health.
New York: Basic Books, 1958.

 

78

Katz, D. and Kahn, R. L. The social psychology of organi-
zations. New York: John Wiley and Sons, 1966.

 

Lesieur, F. G. (Ed.) The Scanlon Plan. Cambridge, Mass :
M.I.T. Press, 1958.

 

Lewin, K., Lippitt, R., and White, R. Patterns of aggres-
sive behavior in experimentally created social climates.
Journal of Social Psychology, 1939, 10, 271-279.

 

Likert, R. New patterns of management. New York: McGraw-
Hill, 1961.

Likert, R. Human Organization. New York: McGraw-Hill, 1967.
Mann, R. D. with Graham 8. Gibbard and John J. Hartman,
Interpersonal styles and group development. New York:

John Wiley and Sons, 1967.

Maslow, A. Motivation and personality. New York: Harper,
195A.

McGrath, J. E. and Altman, 1. Small group research. New
York: Holt, Rinehart, and Winston, 1966.

 

McGregor, D. The human side of enterprise. New York:
McGraw—Hill, 1960.

 

Miles, M. B. Human relations training: how a group grows.
Teachers College Record, 1953,-5§, 90—96.

 

Mills, T. M. Group transformation: an analysis of a learn—
ing group. Englewood Cliffs, N. J.: Prentice-Hall,
196A.

 

 

Modlin, H. C. and Faris, Mildred. Group adaptation and
integration in psychiatric team practice. Psychiatry,
1956. 12. 97-103.

 

Nelson, C. W. Development and evaluation of a leadership
attitude scale for foremen. Unpublished doctoral dis-
sertation, University of Chicago, l9A9.

Offer, D. and Sabshin, M. Normality. New York: Basic
Books, 1966.

 

Olmsted, D. W. A developmental model of the social group.
The Sociological Quarterly, 1962, 3, 195-207.

 

Philp, H. and Dunphy, D. Developmental trends in small
groups. Sociometry, 1959, 22, 162-17A.

 

79

Piaget, J. The moral judgment of the child. lew York:
Harcourt Brace, 1932°

 

Roethlisberger, F. J. and Dickson, W. J. Management and the
worker. Cambridge, Mass.: Harvard University Press,

1939.

 

Rogers, C. R. On becoming a person. Boston: Houghton
Mifflin, 1961.

 

Schroder, H. M. and Harvey, 00 J. Conceptual organization
and group structure. In 00 J. Harvey (Ed.), Motivation

and social interaction. New York: Ronald Press, 1963,
3.38—.1660

 

 

Schutz, W. C° FIRO: a three dimensional theory of inter—
personal behavior. New York: Rinehart, 1958.

 

 

Shepard, H. A. Changing interpersonal and intergroup rela-
tionships in organizations. In J. G. March (Ed.),
Handbook of Organizations. Chicago: Rand McNally, 1965.

 

Taylor, F. W. The principles of scientific management. New
York: Harper, 1911.

 

Taylor, J. Co and Bowers, D. G. The survey of organizations:
toward a machine-scored, standardized questionnaire
instrument. Technical Report, September, 1970, Insti-
tute for Social Research, University of Michigan,
Contract number N0001A-67—A—0181-0013, NR 170-719/
7—29—68 (Code A52), Office of Naval Research Group
Psychology Programs.

 

 

 

Thelen, Ho and Dickerman, W. Stereotypes and the growth
of groups. Educational Leadership, 19A9, 6, 309—316°

 

Tuckman, B. W. Developmental sequence in small groups.
Psychological Bulletin, 1965, 6§_(6), 38A-399.

 

Uhrbrock, R. S. and Richardson, M. W. Item analysis.
Personnel Journal, 1933, 12, lAl-15A.

 

Weick, K. E. Laboratory experimentation with organizations.
In J. G. March (Ed.), Handbook of Organizations.
Chicago: Rand McNally, 1965, 1911-2600

 

Weiss, R. S. and Jacobson, E. A method for the analysis of
the structure of complex organizations. American
Sociological Review, 1955, 29, 661-6680

 

80

Werner, H. The concept of develOpment from a comparative
and organismic point of view. In D. B. Harris (Ed.),
The concept of development. Minneapolis, Minn.:
University of Minnesota Press, 1967.

 

Zalesnik, A. and Moment, D. The dynamics of interpersonal

 

behavior. New York: John Wiley and Sons, 196A.

APPENDICES

APPENDIX A

HEINEN WORK TEAM SURVEY

WORK TEAMS

 

This is part of our program to get additional data
about how a work team develops. With it we hope to find
things that we should start doing, other things that we
should stop doing and things that we need to do better.
There are no right or wrong answers, only your answer is

important.

81

82 ”

No matter who is in your work team, who are the people you
work with. LIST on the next page the names of the peOple in the

company with whom ypu work most closely. "Work most closely" means

 

the pepple with whom you have regular Contacts on any aspect of
your 392, Name as many or as few people as accurately describe your

working contacts.‘

Check how frequently you contact in person, on the phone; or by
letter or memo each of the persons you name.
g
Put a check mark after each name in the column which indicates

the importance of this work contact.

Check the column which indicates how long you have worked closely

with this person.

83

Time

Impcrtance

Frequency

'ames
of
People

of
Contact

Known

Contact

. O, 8.8!!

 

...aa,,e-s.a.-.w....ma.>-.-a-l

t!..l "v

 

 

 

 

 

 

_ --
_.—J -—.—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ncmmz.o~ ow memmz-m. ....... a: -- .2:
mum-wxn-mioplumwsalr in: II: :1! -_.l-.: :1...-
me 2atoezmpwcoe N 2::.-::.:-2 -
u 2: .wmmp-co mcucoe m ...... -:i .2: - -tI :u.-:: -::22:::
I
.2 EDS-.29., - i: .i- in 2..- ..: 2.... I:-
:.wrupw4 12::-
- . - umszmaom
- sac . -
:- - tea i.e., 2 - 2..- . iii-:-
IIIIII. - Iii..-- .-'.iniI-I-. 53-... _
sudMXiumo was.“ ~mco>mm . ,:::.:::- l - -
- --..s...-s§wm.,..chest-5 ,
. xfixmmz mmemp m :0 N . - at: :1;
- -, -2 .z-a-n-xswmo -:-.:|::I:
seven mmswe chm>mm 1:- 2. .2: --- :a-.:-.
fi .
.
2-- 2d:-:22:::a 52L129-i-2223.-:: 2 ......... 22:-2:::-

 

 

 

are
mari
per
act

Att
Lee

(302

17.

18.

19.
20.

2:.

22.

23.

24
25

26
27

2E

(A,

81:

Below is a list of activities that might be performed in a work team. Not all_of these activities

are performed in every work team. For those activities that are performed in your 5913 tgaub put a check
mark under the name of the person who does it. If more than one person does it, check each person that
performs the activity. If no one in your team performs the activity, leave it blank. Each of these

activities can be performed in a team meeting or daily on the job.

NOTE: See the attached list for the number code of the people in your work team.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EXAMPLE:
Work Team Members
1 - 2 3 4 5 6 l7 8 l9 l0 iii 112 l3 l4 {is
Attends monthly work team meetings. x x x x x x x x x x x x x x x
Lectures to civic groups about the
Scanlon Plan -
Goes to management grid training. x
. Work Team Members
HORKTEAMNO. l 2 3 4 5 6 7 8 i9 l0 hi 12 13 l4 l5

 

l7. Records work team minutes

18. Gives recognition for a job well
done.

 

l9. Coordinates work team activities

 

20. Coordinates our team with other
production work teams.

 

 

2l. Coordinates our team with Engineer-
' ing, maintenance, and other staff
teams.

, ...._----..-— -

 

22. Suggests new ways for solving .-
production problems. 3

 

23. Puts suggestions into effect. 2

 

24. [Emphasizes team work. ’

 

25. Deals with conflicts of opinion ;
within the team.

 

26. Spends time chatting with and
counseling work team members.

 

27. Gathers information for work
team activities.

 

28. Encourages the work team members
to improve their present per-
formance.

 

29. Encourages the work team members
to talk together and share ideas.

 

30. Tells the work team members how 2
well they are dOing their job.

 

3T. Obtains materials to perform the
Job.

 

 

32. Trains new team members. :

 

33. Keeps the team moving toward their
goal.

 

34. Acts as a representative to the
Employee's Committee.

 

35. Encourages the work team members
to set Specific goals for their
work.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

36. Makes comments about the effective- .

—n‘-1- -‘ «Lg—.— _4‘:.~-n J L L J J

 

 

 

 

team to do; other things you nay not we. c
Other. that is your own Opinion about what t

l0.

ll.

12.

l3.
l4.

15.

l6.

855

Below is a list of things your wpr

k
H:

U

Strongly
prefer
the team

not do it

(1)

Meet regularly once each month.
Rate team members for merit increases.

Decide what training and work Team Development
activities the team needs.

Tell each team member in detail how well he
is doing his job.

Interview and select new team members.

Discipline team members whenever company
policies are violated. _~

Allow team members a great deal of freedom in
they way they do their work.

Get Engineering, iaintenance. and other staff
people involved in production problem solving
meetings.

Have social get togethers. parties. picnics. etc.

Get together with other shifts to solve inter-
shift problems.

Decide what new products the company should
develop.

Take the responsibility to train team members
to get ahead in the company.

Go to other teams for help in solving problems.

Set the standards for jobs within the work team.

Decide what to do with people whose jobs are
eliminated.

Determine how lay-offs should be handled during
periods of an economic sloxdown or a strike.

team_is doing or might do. Some of these things you may prefer your
vei‘ teap_to do; and still others you may not care ore way or the
:e team_a§_§_whole should do, not just you individually or management?

 

ALTERNATIVES
Somewhat Doesn't Somewhat Strongly
prefer matter prefer prefer
the team to the team the team
not do it me do it do it
(2) (3) (4) (5)
+———

 

 

m“%-__ ._____.____
. MM-~-.”

APPENDIX B

INSTRUCTIONS TO MANAGERS FOR PAIRED

COMPARISON WORK TEAM RATINGS

In front of you is a deck of cards. On each card is
listed a pair of work team names. For each pair you are to

circle (or underline) the name of the work team which in

 

your opinion is most effective. We want you to judge the

 

overall effectiveness of the teams. You should consider
their ability to work together, their creativeness in

solving their problems, and their skill at handling conflict.

86

APPENDIX C
ITEM ANALYSIS

The characteristics of the items used in measuring each
of the variables from the ICLS questionnaire were examined
through the use of cluster analysis techniques (Hunter and
Cohen, 1970). The cluster analysis was based upon a sample
of 306 individuals from all of the company employees, who
had returned completed questionnaires and not merely limited
to those llU individuals in the twenty work teams in the
analysis. The total organizational pool of subjects was
selected for purposes of generalizability and stability of
measurement. The measures used in the study were intended
to reflect characteristics of work teams in the whole
organization rather than the twenty teams with complete
data. Also, the characteristics of cluster analysis
require a sample of several hundred to insure the stability
of the measures in the analysis. For both these reasons the
item analysis used the total organization subject pool.

The items used as measures of the variables for this

study that were included in the ICLS questionnaire were:

87

88

Cohesiveness

 

1.

To what extent do you feel that you and the other
persons in your work group belong to a team that works
together?

How much do the members of your work team want to stay
in your work team rather than join any other work team?

How does your work team compare with the other work
teams at on:

 

the way they get along together?
the way they stick together?

the way they help each other on the job?

Problem Solving

 

 

*ll. To what extent do persons in your work group keep each
other informed about important events and situations?
*12. To what extent are persons in your work team willing to
listen to your problems?
13. To what extent does your work team plan together and
coordinate their efforts?
in. To what extent does your work team make good decisions
and solve problems well?
15. In discussing a problem in your work team, to what
extent do team members feel free to suggest ideas
that are different from the majority opinion?
16. In arriving at a decision to a work team problem,
how often does your work team outline a detailed
action plan?
17. To what extent does the work team feel free to change
a previous team decision?
Identification
21. To what extent do persons in your work team help you
find ways to do a better job?
22. To what extent do persons in your work team provide

the help you need so that you can plan, organize, and
schedule work ahead of time?

89
23. To what extent are your interests and needs taken into
account when work activities are assigned?

*2A. To what extent are you important to the smooth opera-
tion of your work team?

25. To what extent do the members of your work team see
eye to eye on most matters?

26. To what extent do the members of your work team agree
with you on what are the basic purposes of your work
team?

Perceived Team Growth

 

*31. To what extent do you have confidence and trust in the
persons in your work group?

32. On the basis of your experience and information, how
would you rate your work group on effectiveness? How
well does it do in fulfilling its mission or achieving
its goals in comparison with other work groups in the
company?

33. How do you feel about the progress your work team has
made in the past year?

‘3A. To what extent do you believe you have ideas for

improving your work team that have not been expressed
and fully discussed.

‘TBS. To what extent do you think you and the people you
work with need training in how to build a more effec-
tive work team?

*items have been dropped from the scales through item
analysis.

Table 21 gives the inter-item correlation matrix for
these items and Table 22 presents the median and the range
of the inter-item correlations, item—cluster correlations,
and the item communalities for each of the scales contained
in the ICLS questionnaire; For all the variables the inter-
item correlations, item-cluster correlations, and the item

communalities were moderately high to high.

90

 

Ill"

 

 

 

 

 

on. am. mm. mm. mm. om. mm. mm. Om. ea. om. ow. am. am. AH.::wm.. mm
mo.- mo. mm. OH. mo. so. NH. mm. mm. ow. AH. no. mo. mo. mo. mm. am
am. mm. mm. mm. mm. am. am. am. ow. mm. mm. ow. ca. am. mm. he. mm
ma. om. mm. mm. mm. mm. Hm. me. am. mm. mm. Ha. Ha. me. mm. mm. mm
HH. em. me. He. om. mm. mm. om. me. me. He. oe. oe. mm. am. mm. mm
am. He. am. mm. mm. mm. ow. mm. oe. am. we. oe. cm. am. mm. me. mm
mm. He. am. am. mm. mm. mm. mm. mm. am. we. mm. mm. mm. 8H. Ha. mm
oo.H em. mo. ma. ma. om. ma. mm. mm. ea. ma. mm. mm. mm. am. am. am
oo.H mm. mm. mm. mm. mm. He. am. am. am. am. mm. om. Hm. oa- mm

oo.H mm. mm. am. am. ea. mm. Ha. Ha. Hm. Hm. am. ma. mm. mm

oo.H mm. am. am. mm. mm. ma. mm. mm. mm. mm. mo. mm. mm
octa mm. mm. mm. mm. am. mm. mm. mm. mm. mm. am. he.

oo.H ow. am. mm. mm. mm. mm. mm. om. mm. mm. on

oo.H am. om. mm. mm. mm. mm. mm. mm. om. ma

oo.H me. am. me. am. am. mm. .mm. mm. ea

oo.H om. me. me. mm. mm. mm. as. ma

oo.H mm. mm. mm. mm. «a. He. NH
oo.a ma. mm. mm. am. he. -mm.

oo.a we. we. ma. me. m

oo.H as. am. me. a

oo.H mm. we. m

oo.H nm. N

i i oo.H H
am mm mm am am 0H ma eH ma NH HH m a m m H meaum

 

mCOHHwMHWHHOU E®#HI.H®HCH meMUW maHUH

Hm mam/NH.

91

 

oo.H Om. av. mm. ON. mm. mm. mm

Oo.H mm. mm. ma. ma. OH. am
OO.H 5%. mm. mm. av. mm

oo.H mm. mm. mm. mm

OO.H mm. mm. Hm

OO.H mm. mm

OO.H mm

 

mm Vm mm mm Hm mm mm mEmuH

 

 

Apmscflucoov

Hm Mdmdfi

92

TABLE 22

Median and Range of Inter—Item Correlations,
Item-Cluster, and Item Communalities
of Scales in ICLS

 

 

Variable Inter-Item r's Item-Cluster r's Item Communalities
Scales medianlxwvhigh median low high median low high

 

Coh 52 nu 71 8o 55 83 64 31 69
PS 35 30 68 62 58 68 39 3A A6
Id 37 3A 68 67 58 68 A5 33 A7
PTG A1 38 M7 66 58 70 um 35 ug

 

Alphas were computed for each of these scales for both
the total organizational sample and also for the twenty pro-
duction level work teams. The alphas generally are only
slightly lower in the smaller production level work team

sample (see Table 23).

TABLE 23

Alphas of ICLS Scales

 

 

 

Variable Total Twenty Production
Scales Organization Teams

Coh 83 83

PS 77 71

Id 79 83

PTG 68 68

 

93

The normative behavior measure and the leadership dis-

tribution measure were composed of items contained in the

Heinen questionnaire. The sixteen normative items used were:

1.

2.

\OCID‘QGUW

10.

ll.

l2.

13.

1A.
15.

16.

Meet regularly once each month.

Decide what training and work team development activi-
ties the team needs.

Get engineering, maintenance, and other staff people
involved in production problem solving meetings.

Get together with other shifts to solve intershift
problems.

Set the standards for jobs within the team.

Go to other teams for help in solving problems.
Decide what to do with people whose Jobs are eliminated.
Determine how layoffs should be handled.
Interview and select new team members.

Discipline team members whenever company policies are
violated.

Rate team members for merit increases.
Tell each member in detail how well he is doing his Job.

Allow team members a great deal of freedom in the way
they do their work.

Have social get togethers, parties, picnics, etc.
Decide what new products the company should develop.

Take the reSponsibility to train team members to get
ahead in the company.

Table 24 gives the norm inter-item correlations.

9A

 

oo.a mm. ON. ma. mm. ma. OH. ma. ma. Hm. mm. mm. mm. mm. Hm. mm.

KO
r—l.

 

oo.H mm. Hm. HH. mo. mH. om. om. om. Hm. «H. mH. eH. mH. NH. mH
oo.H HH. Hm. mH. aH. mH. mH. Hm. mH. so. 8H. we. MH. HH. eH

oo.H mH. HH. mm. mm. mH. mm. mm. mH. Hm. mm. mm. mm. MH

oo-H we. om. mm. om. am. mH. am. am. mm. mm. sH. NH
oo.H am. Hm. mm. mm. om. Hm. am. mH. mm. mm. HH
oo.H mm. mm. mm. mo. mH. oH. mH. 8H. mo. oH

oo.H Ha. Ha. mm. mm. mH. mH. mm. Hm. m

oo.H on. mH. om. mo. no. mm. MH. m

oo.H mm. mm. mo. no. em. 0H. s

oo.H em. em. mm. pH. pH. 6

oo.H am. mm. mm. mm. m

oo.H Ha. om. we. a

oo.H om. am. m

oo.H mm. m

oo.H H

6H mH eH mH NH HH 0H m m a e m a m m H msmuH

 

 

mcoHumHmHHou EouHiumucH m>HumEuoz

wm mqmflB

ll.
l2.

13.

IA.

15.
l6.
17.
18.

95

The eighteen leadership distribution items were:
Suggests new ways for solving production problems.
Puts suggestions into effect.

Gathers information for work team activities.
Obtains materials to perform the Job.

Trains new team members.

Tells the work team members how well they are doing
their job.

Keeps the team moving toward their goal.

Encourages the work team members to set specific goals
for their work.

Gives recognition for a job well done

Spends time chatting with and counseling work team
members.

Emphasizes team work.
Deals with conflicts of opinion within the team.

Encourages the work team members to improve their
present performance.

Encourages the work team members to talk together and
share ideas.

Makes comments about the effectiveness of team meetings.
Coordinates work team activities.
Coordinates our team with other production work teams.

Coordinates our team with engineering, maintenance,
and other staff teams.

Table 25 gives the leadership distribution inter-item cor-

relations.

Table 26 contains the median and range of the inter-item

correlations and the alphas for both measures. The

Leadership Distribution Inter-Item Correlations

\ C)
O\.

TABLE 25

 

 

 

Items 1. 2 3 u 5 6 7 8 9
1 1.00

2 .MO 1.00

3 .31 .u0 1.00

M .u3 .uo .3u 1.00

5 .35 .21 .28 .M5 1.00

6 .20 .28 .u0 .M1 .33 .00

7 .35 .27 .32 .M2 .39 .21 1.00

8 .27 .29 .M5 .37 .30 .M6 .3M .00

9 .30 .27 .25 .35 .11 .33 .18 .37 1.00
10 .29 .30 .u6 .M2 .23 .M1 .2u .39 .30
11 .39 .u1 .38 .M3 .25 .3M .uu .37 .35
12 .31 .26 .31 .37 .20 .39 .3M .34 .35
13 .33 .30 .M9 .30 .30 .M1 .36 .51 .32
1a .26 .38 .u2 .u1 .25 .uu .37 .M5 .30
15 .M3 .32 .38 .51 .37 .34 .M6 .35 .35
16 .32 .28 .MO .32 .26 .32 .M3 .32 .35
17 .23 .26 .MO .u6 .2u .u3 .32 .M2 .36
18 .35 .36 .38 .50 .26 .u7 .25 .39 .26

 

97

 

 

 

TABLE 25
(Continued)
Items 10 11. 12 13 14 15 16 17 18
10 1.00
11 .35 1.00
12 .46 .H2 1.00
13 .H2 .36 .35 1.00
14 .MM .37 .42 .51 1.00
15 .36 .HS .MO .33 .39 1.00
16 .29 .29 .3U .36 .38 .39 1.00
17 .U2 .37 .41 .39 .30 .29 .50 1.00

18 .33 .25 ‘.32 .26 .21 .28 .28 .M1 1.00

 

98

inter~item correlations are somewhat lower for these scales
than for those scales included in the ICLS questionnaire, but

the alphas are the same or a little higher.

TABLE 26

Median and Range of Inter-Item Correlations
and Alphas for Normative and
Leadership Distribution

 

 

 

 

Variable Inter-Item r's Alpha
Scales Median Low High

Norms 22 2 7O .78

Leadership 35 ll 51 .90
Distribution

 

Table 27 presents for each of the variables, the mean,
the standard deviation, and the range of scores of the
twenty work teams in the analysis. All the scales are five
point scales except for the shared norms (SN) measure which
is a variance score and the managerial ratings (Mgr.R) which

are scale scores derived from paired comparison ratings.

99

TABLE 27

Variable Means, Standard Deviations, and Ranges
for the Twenty Production Work Teams

 

 

 

Variable Mean Standard Low High
Deviation Score Score
Coh 3.65 50 2.78 9.50
PS 3.29 .97 2.59 9.21
SN 25.85 6.85 17.68 39.98
1d ' 3.99 .92 2.71 9.20
TD 2.81 1.19 .25 9.75
Comm 1.71 1.96 0.00 5.00
LD 2.71 .69 1.08 9.10
PTG 3.99 .37 2.95 9.18
Mgr. R 6.90 2.11 2.59 12.26

 

"I1111111119111111s