.__ I I ..
a»:
...~—._~__ .-
‘53:?

 

‘0.

 

".1

fizp‘. .

31113 ‘3 l~‘ ‘1! 1
“NW ’3'? ~ I ,, .3

‘. E. :5: '1‘1‘311‘1‘ii1‘1‘1' «‘3‘1111‘11rwt1‘1‘éf
.2: .1' El"? W
.3; i1{,; I f‘ I ‘ ‘: . '57; 113‘}; L

‘r ‘H ..‘:.;l:.I 133:" .3 .1 .. ‘
:3 | 111‘ “'11u.-§‘L!fi 3 ' 33}. ,3'15‘ "
‘33. 41:11.21 'i ““ ’ ‘1“ 1‘1E““1::“ ‘
‘ 1." v 1:11 51" 1111.1 ‘1 1111111:

‘; :3 1' 311.31!“ "51 ““1“

3.1 '.:-' 1 ‘ “353;; ifi‘ir". ggixjg;
Is. 5‘. 9 1.311153335:f,."'.‘r
1“; i535}. “1 3-3:;
1 1. ‘ i“ " 1 ' Mm :'

‘1‘ 3-“ 1 h 1191‘? .11“'-,-.13‘11““
" 3 ) '1 II ‘ l u n i r

' v.3 1 «'13: HI - .. -.
A' ‘ I ~ HIM, 1"." f: 1“??? 1‘ ‘1'": 1111”}, ‘1’};

. .‘,' ‘ £1“. “ “1 1 3:13“ ‘
51 ,3 1‘“ ‘1 ‘11". “111mm,:

- .aI.‘

.311"?wa . .

-_..-. .n>-~
‘~.- ‘,~' 21.?

“\‘.

fir“:

-2“:
03“
- .ggcé

r43?
v-‘= —:;~

4.7»
,r'i‘ :. :-
$3: . "-

-. .... 4n."

       
 
     

      
  
    
 
     
      
 
 
     
  

   

    
  

              
    

g! r" -,. ‘7 ',, '1' ‘LJ‘?‘. "L‘1‘l . ‘l'Ii‘ 3‘7". —
f ’93“ .5 I‘: %. 21711591111515“ 13.52:: 11535-31? 151593111“ ‘
‘ “is .. .33 . My"! ‘1- ; ‘1‘1‘.‘:1““ éfiigéfim‘gfi‘ ”Whig?"

    
   

fiv‘g’fifi.
g I .711‘1‘33‘ 11:111W11‘VI9
1131111111“. 1110115?“ I 11‘1”" ‘1?“ ‘ 1‘3,” I!

$43311 . '1111‘31131'
‘1 ["3 :‘-‘I3. gag-$515131 i

.3.“

1'5
”33“ A ‘3}:-

'FrA

. ‘3'."
31' 1““ .5 111““ ”151 3 ““1131;
.. m1.“ “ . M1 MI '31“ 3‘3"! 333‘. jag; “1 22.13: 39%;
iii 1"““1‘11‘11513‘ 11:43] fI‘I'HI 11‘3““11fl I3" f“ 1611113?“ 212;"! A; j 11c“?! 1‘?
31"!“ '1'. 1 ; .

"—v

«9-—

1 r31“.‘.“-‘“““
{1533311 .43

I' .. -:. ‘1
{3115‘11‘33115111‘ J‘g‘h'fi‘ 1‘ "“

.1; -

  

"s‘.’

,
"
,fl.
_
o'- ...' :0
.:.{x:.' _. ,
.._J.4¢3(.'
-.

, -
{Eagzé av
f..." n...

-. _ ~13?

I .

‘ I..

       
   
 
     
    
   
  
  
  

9...?

'13::-

1"“
?%

 
 

"3-!

u. .-
.J
1..

. .I 33333., ,. , .. @353.
3,; 1,13” I ‘1': I n 3; ‘ 1. Md 5 '

L H ‘}!1‘1111“1t‘.‘11111 ’3': ' ’ 11112111511"f‘,““11““§'ki‘£:“‘i“‘ 3.1132413?! . gilfigfi'
"1" 3333 I1 .~, [5.3 “"351 i "3-; 3.33435 fig}: ‘5‘" .33. .
I“‘11‘1.I1"31'S:‘1'L1‘5I1‘311‘1 ' I111§1 11111111151111.11111‘

1" “1‘5“1 1 1-5111 W1 3‘ 1" .1111““1 “1‘ ‘1111351 I: 11111"1L‘I51‘711 1
x 1 111““ :jle “If; ‘il‘lpfi $711,; '14 ":ll 3";111h“31‘111r;‘§§3.
. #11“ h ““311111":1“~&‘111 V

‘1‘I I‘ I , H.141
1313513?in . .323333, 53;; :3...
1.015;:311‘ .' 5511‘ r 11.111111,
9.11 I 1““‘1“1“‘“‘“ 5:},
. 1153‘ ..5“1‘52 '139‘33
‘51'13‘p‘é‘it $“y“I‘ 1‘1‘51‘1511111‘figfii‘ . 1“
‘1““11‘1ii‘3‘d ‘111‘1513H1fl-L’ 4,3

  
  

3:» HI
' 5312115

  
   
   
    
 
   

    
 

   
  
 
 
  
 
   
 
   

      
  
  

  

  
 
       
 

  

   

  
  

    
    
    
  
  
 
   
   
 
   

    
 

 
 
  
   
 

 
 
    
 
     
     
 

    

  

 

      
     

.

I g, '4
" ‘ 1‘ E{ '121 I! .
‘1‘ 1115151111151‘I121“111£§1§l ”511‘95‘1' 1‘11 1‘11““
' 1' I . - "J‘ . .‘ 'I
1;); it“ ‘1‘11‘53'3‘ia111‘ .1“? {33.813

1 ’) ; I‘D?!

 
   
   
 

 
    
     
   

     

         
     
      
    
    
     
 
 

       
  

    

{c251
T‘?
:::::.
:1;
.r}.

        

   
  

   
  
    
  

II I. I
_ _. "may; . .- ' _. v
' .27." .;.;‘,§8:r . , .- . 5 ,1 ~
{i 3:35: 1. '. . * #3.. .3.- - $11?
. m." .g- fl: " '. . .u-av A“
"-z ' Y" ‘2. '43:“ ”‘1‘.
w . . ,

 

     
    
     
     

          
 

 
      
 

 

  

 

 
 

.- I— .< .1‘.‘ '1 1". £3...
0‘ “‘33, ., 2... 3.21315; 1 l"1111‘f1 1|” “my?” 1'1""; 9", ‘ .‘-“..‘:‘::.
'1. ‘3 wt. . . 'I I. “1:" I. ."C . . ”it" "' fv“=-5*”I”’ fit -
. . '.. :51353fisz«211?1‘s“ ‘1 15531.3 “333:1 ‘ (-11‘1‘13““‘1"5,‘ 1 :11’1‘I‘1E‘151‘11‘1‘“1‘11“1" '. » ‘3‘ :I9'7ii‘z‘é‘: ""3.
, *3' ".2 .42;.33533333353135w -"I11:,va‘.‘1151‘11‘11i;{n‘133 33:53333333‘3I3‘I5 " ~53!“ . '
223-},13 ';'II' I... 3 ‘r'. ‘.;".":‘-..§»~“' “.Itdzi‘ 'z‘t’l‘l . ‘ " 5' ‘ 5‘33 . ’ 1'4" :'
513‘ 5. 3‘ 311:! 1"”:‘(‘;‘|‘E1'1}13é:121‘i‘:1;}:: 111113252Lis‘1111‘3'111‘; {FF/1“!” 1111‘1‘1‘1‘ L11": 3:! ‘F‘Tlf" L51 '31:; 133-11 {1‘91“} 1 .' W11“? 3‘
.5 ”1.1.”! '.-.I.’: 1M" [11333533913 5":‘1I‘I1‘1‘ ll11"“ !' fit". 13 MI ‘1“"13i‘1‘" " “M1191“: ‘11111‘111‘11 313 61131 {11911114
.'. ‘I. “"33 4.3333313135355123 1‘1"f‘o‘11‘11““11l“‘““““"111E‘!‘5 32‘1EQI533‘7113’1 $511? ‘1‘!
‘1'1‘1'LI"‘I "5 ‘ ‘1'“1'HHI' " ’ . ' ' i' '1 .':.I .
3139:3‘»?II‘I..E‘E‘I‘I‘LI‘IIIE-33‘15522'I‘E‘Ilz331155‘g35‘3‘ 31311111115111? iii 13. 33%»
251mm! I"":':"'-' 111"". ‘55 ‘11 £3111“ - - ,.. 523‘: .1:' we 1.‘ ; I 1
II1:1.»11311433311193.331I {33H IIWJ‘ ,. 1 ' 'Izi‘ -’: ".II'I- J" u.
z5.3%..='.'.33333.33335153333?)3"'1‘.“ . 31.33%. {3‘3 333 “I I 3‘ I335... . «I?
J;‘I;—:-::‘::.£j;3:.l3323333333333.3333:293r3 ‘3 my? I a. .Y .3.
3523133433! 3111311 I 3‘53! I e‘ 3113' . - . ”3151-. .31“..wa "3‘2"
1551'i‘11‘3‘1'1fl131“11‘5‘.l111113‘1"311 341. "191:. @3131 - . I: «r V-
‘13‘5‘1‘.:"111.“1‘1“3 311115"'III“1“11‘11‘ “15‘: “131,35 15¢,th 7.3!. .
1133'? 31131111. 1 I'3 1111513 5 I, ‘13‘1111‘1‘1"“‘1"‘““
III W l Eh: 11} 1.11 ill.” [I \ A1111“;
“‘I153‘1m ‘n11m11 1111111;1-‘J1L!‘1 941.1111 ' KI. rl

AET UNIVERSITY LIB

é“ IIIIIIIIIIIIIII II II IIIIIIIIIIIIII

(‘0‘qu 3129301694362

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the
thesis entitled
GRASSROOTS CHANGE: VARIABLES
AFFECTING ROLE INNOVATION

presented by

Ann E. Williams

has been accepted towards fulfillment
of the requirements for

M.A. Psychology

degree in

 

 

 

5%! profl

Date 22 I621]???

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

LIBRARY
Mlchlgan State
UnlversIty

.—

 

 

 

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1M chIM.pG§-p.14

GRASSROOTS CHANGE: VARIABLES AFFECTING ROLE INNOVATION
By

Ann E. Williams

A THESIS

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

MASTER OF ARTS

Department of Psychology

1997

ABSTRACT

GRASSROOTS CHANGE: VARIABLES AFFECTING ROLE INNOVATION

By

Ann E. Williams

This study provided a construct definition and a conceptual model involving the
predictors and outcomes of role innovation. The study considered the impact of climate
for innovation and goal focus on role innovation. The process of role innovation was also
assessed by studying the impact and contribution of horizontal member exchange and role
negotiation. The effect of mastery and performance learning orientation as well as
individualism/ collectivism and two final outcomes, performance and task satisfaction,
were also investigated. The participants were 228 undergraduates who worked in dyads
on a Tinker Toy building task. Coders later assessed videotapes of the production
sessions to determine whether or not certain role innovations had occurred. The results
indicated that climate for innovation affected role innovation, but few of the other

primary hypotheses were supported.

TABLE OF CONTENTS

LIST OF TABLES ...................................................................................... vi
LIST OF FIGURES .................................................................................... vii
INTRODUCTION ...................................................................................... 1
CONSTRUCT DEFINITION ..................................................................... 4
Jobs versus Roles .............................................................................. 4
Role Theory ...................................................................................... 5
Defining Role Innovation ................................................................. 7
Existing Definitions .......................................................................... 8
Construct Definition ......................................................................... 9
Distinction from Related Constructs ................................................ 10
LITERATURE REVIEW ........................................................................... 13
ORGANIZATIONAL FACTORS .............................................................. 18
Climate for Innovation ...................................................................... 18
Goal Focus ........................................................................................ 2O
PROCESS OF ROLE INNOVATION ....................................................... 25
LMX and TMX ................................................................................. 27
Horizontal Team Member Exchange (HMX) ................................... 30
Role Negotiation ............................................................................... 31
RESEARCH MODEL AND HYPOTHESES ............................................ 33
Climate for Innovation ...................................................................... 35
Goal Focus ........................................................................................ 36
Interaction between Climate and Goal ............................................. 37
HMX ................................................................................................. 41
Role Negotiation ............................................................................... 42
Performance ...................................................................................... 42
Task Satisfaction ............................................................................... 43
Individual Differences ...................................................................... 43
Individualism/ Collectivism ............................................................. 44
Learning Styles ................................................................................. 47
Overall Model ................................................................................... 49

METHOD ................................................................................................... 5 1

Participants ....................................................................................... 51
Design ............................................................................................... 51
Rewards ............................................................................................ 52
Change Memo ................................................................................... 53
Task .................................................................................................. 53
Manipulations ................................................................................... 54
Climate for Innovation ............................................................ 54
Goal Focus .............................................................................. 55
Procedure .......................................................................................... 56
Measures ........................................................................................... 57
Individualism/ Collectivism ................................................... 58
Mastery/ Performance Learning Orientation .......................... 60
Climate for Innovation ............................................................ 62
Goal Focus .............................................................................. 62
HMX ....................................................................................... 62
Role Innovation ...................................................................... 62
Role Negotiation ..................................................................... 65
Performance ............................................................................ 66
Task Satisfaction ..................................................................... 67
Job Change ............................................................................. 67
RESULTS ................................................................................................... 68
Data Analysis .................................................................................... 68
HMX ................................................................................................. 70
Role Negotiation ............................................................................... 71
Role Innovation ................................................................................ 73
Performance ...................................................................................... 75
Task Satisfaction .............................................................................. 77
Additional Analyses ......................................................................... 80
DISCUSSION ............................................................................................. 90
Primary Research Questions ............................................................. 90
Final Outcomes ................................................................................. 96
Individual Differences ...................................................................... 99
Limitations ........................................................................................ l 01
Contributions .................................................................................... 105
Future Research Directions ............................................................... 107

APPENDIX A: Organization Descriptions ................................................ 110

APPENDIX B: Goal Manipulation ............................................................ 116
APPENDIX C: Standard Operating Procedures ......................................... 118
APPENDIX D: Progress Reports ............................................................... 123
APPENDIX E: Change Memos .................................................................. 125
APPENDIX F: Scales ................................................................................. 127
APPENDIX G: Videotape Coding Sheet ................................................... 136
REFERENCES ........................................................................................... 138

LIST OF TABLES

Table 1: Rotated Factor Matrix for Individualism] Collectivism Scale ..... 59
Table 2: Rotated Factor Matrix for Mastery & Performance

Learning Orientation Scale ........................................................... 61

_ Table 3: Means, Standard Deviations, Reliabilities and

Intercorrelations for Model Variables ......................................... 69
Table 4: Hierarchical Regression Analysis Results for HMX .................. 71
Table 5: Hierarchical Regression Analysis Results for

Role Negotiation ......................................................................... 73
Table 6: Hierarchical Regression Analysis Results for

Role Innovation .......................................................................... 75
Table 7: Hierarchical Regression Analysis Results for

Performance ................................................................................ 77
Table 8: Hierarchical Regression Analysis Results for

Task Satisfaction ......................................................................... 79
Table 9: RM-MANCOVA Results ........................................................... 81
Table 10: Summary of Hypothesis Support ............................................. 89

vi

Figure 1:
Figure 2:
Figure 3:
Figure 4:
Figure 5:
Figure 6:
Figure 7:

Figure 8:

LIST OF FIGURES

Model of Role Innovation ........................................................... 34
Interaction of Climate and Goal Focus ....................................... 40
Model of Role Innovation with Hypotheses ........................... 50
Climate Effect on Role Innovation ............................................. 83
Climate Effect on Performance .................................................. 84
Climate Effect on HMX ............................................................. 85
Goal Effect on Performance ....................................................... 86
Goal Effect on Role Negotiation ................................................ 87

vii

INTRODUCTION

In today’s dynamic, unpredictable environment, businesses and organizations
cannot just proceed “as usual.” As the realities of the changing workforce, worldwide
economy, foreign trade and cut-throat competition begin to impact the marketplace,
organizations need to find innovative ways to survive the fray. Kieman (1993) said it
succinctly, “In a world of ceaseless turbulence and discontinuous change, innovation has
become one of a handful of critical preconditions for survival, as the rapidly shrinking
half-lives of many Fortune 500 companies bear ample testimony.” (p. 11)

Innovation is critical for organizational survival today, yet true innovative power
lies within individuals. Kanter (1983) noted, “...individuals actually need to count for
more, because it is peOple within the organization who come up with new ideas, who
develop creative responses, and who push for change before Opportunities disappear or
minor irritations turn into catastrophes.” (p. 18) In order for organizations to gain
competitiveness through innovation, individuals need the freedom to create and innovate.

The current business environment rarely supports creativity and innovation.
Organizations often try to control the behavior of their employees - including innovative
behavior. The nature of the creative process makes it difficult for organizations to control
innovation; employees cannot innovate upon command (Farr & Ford, 1990).
Furthermore, many organizations have internal controls that maintain the status quo and
discourage creativity and innovation. For example, most organizational socialization
programs teach newcomers to conform to traditional practices and norms, not to develop

new and different ideas (Van Maanen, 1978).

2
Demanding innovative behavior from employees might prove to be an impossible

task. However, many employees already perform unobtrusive but useful innovations in
their current positions. This grassroots innovation occurs as employees change or modify
their personal roles and tasks and are initiated by the employee without organizational
intervention. This grassroots form of innovation, called role innovation, is initiated by
the individual and is not directly controlled by the organization. Role innovation
encompasses many behaviors including an employee changing his or her tasks, work
situations and interpersonal interactions in the workplace (West 1987b; Farr & Ford,
1990). Workers develop new strategies for completing their tasks, re-negotiate their
position with co-workers and even develop new goals. Role innovation is often
overlooked as a value-added activity, probably because it is not as startling or dramatic as
other forms of innovation - such as unveiling a new product (Ettlie & Reza, 1992).

Although the immediate impact of role innovation may be small and focused on
the employee’s role, the effects are likely to aggregate and make an impact at the
organizational level. By role innovating, employees may improve their own productivity,
performance and satisfaction. Employees who role innovate are also might find new and
better ways to do their jobs. Eventually, these improvements could be spread to the rest
of the organization. It is time to recognize the grassroots innovation that is already
occurring in the workplace and recognize its potential benefits.
Structure of Paper

In this research, I tried to contribute to the role innovation literature by providing

empirical data exploring the predictors and outcomes of role innovation. I hoped to make

3
the following contributions. First, I provided a construct definition of role innovation.

Second, I reviewed and discussed existing research on role innovation. Third, I presented
a laboratory study based on a conceptual model that addressed two primary research
questions surrounding role innovation. The first research question was: Will the
organizational factors of climate for innovation and goal focus (group versus individual
goals) affect role innovation? The second question was: How do HMX and role
negotiation contribute to the role innovation process in interdependent workgroups?

In order to answer these questions, I used the following structure for the paper. I
started by developing a new definition of role innovation that pulled from existing
research and added neglected elements. I highlighted the important features of the new
definition and distinguished it from related constructs. Next, I discussed climate and goal
focus, two organizational factors expected to influence role innovation. In order to
investigate the role innovation process, I used the Horizontal -Member Exchange (HMX)
construct (Nason, 1995) and role negotiation to explore the interaction element of role
innovation. After discussing the central constructs, I presented the model and hypotheses
that guided my study. After discussing the theory, I presented the method, results and

discussidn of the study.

CONSTRUCT DEFINITION

When approaching a relatively fuzzy construct such as role innovation, the vital
step was to develop a conceptually distinct and bounded definition. It was important to
answer the question: What is role innovation? In this section, I defined the process of
role innovation. Since employees differ in their roles, situations and needs, it was a
challenge to define role innovation in a way that captured the variety of possible
innovative behaviors. Researchers studying role innovation seem to have an implicit
understanding of the concept, yet few provide explicit, measurable definitions.
Jobs versus roles

Role innovation is a complex construct that is difficult to define. However, the
role and role-making literature (e. g. Katz & Kahn, 1978) offer a good starting place. The
role literature provides a foundation for defining role innovation. An important step is to
realize that roles are distinct constructs from jobs. Ilgen and Hollenbeck (1991) provided
a thorough comparison of jobs and roles. They defined a job as a set of task elements
grouped together under one job title and designed to be performed by one individual. A
job can be described by its tasks, goals, skill requirements and boundaries. This
objective, bureaucratic feature of jobs makes them quasi-static and establishes them as a
basic unit of the organization.

Although they include the elements of jobs, roles encompass a broader set of
elements such as social relationships. Ilgen & Hollenbeck (1991) wrote, “Roles are
larger sets containing emergent task elements plus those elements of the jobs that are

communicated to the job incumbent through the social system and maintained in that

5
system” (174). This definition of roles highlighted two important points. The first point

is that roles contain emergent task elements which are responsibilities and duties that
develop over time for an employee. Emergent task elements may become expected duties
for the employee, but they are not included in the formal job description. The second
point is that roles are based in, and maintained by, the social system. Roles are not
objective, they are based on interpersonal relationships and social influences in the
workgroup. Essentially they are the interconnections among jobs and the people who
occupy them. In summary, roles have less precise boundaries than jobs, include formal
and informal task elements and are part of the social system of organizational structure.
Role theog

One of the important characteristics of roles is their social nature. Katz and Kahn
(1978) discussed the importance of the social system and it’s influence on individual
behavior. They described organizations as being a contrived system of roles. Roles are
more than a set of tasks, because roles differ based on social elements. Roles are not
defined in an objective vacuum; they are dependent on the interaction and input of other
people. The other people in an organization who affect an employee’s role are termed the
role-set (Katz & Kahn, 1978). The role set consists of interrelated individuals in a social
system usually bound by workflow or official sanction. Although Katz and Kahn (1978)
recognized the existence of many types of roles and groups in organizations, this paper
will focus on workgroups (supervisors and co-workers) connected by interrelated tasks.

Members of the role-set develop role expectations (expected patterns of behavior and

6
responsibility) for each member of the role-set based on each one’s needs, skills and

interactions.

Katz and Kahn (1978) described the process of a role episode in a social system.
The role episode is a cyclic, on-going process that involves role-sending and role-
receiving. Role-sending involves one person sending a message with certain intentions.
The sender can intend the message to merely inform the receiver or to actually influence,
direct or control (Katz & Kahn, 1978). The message reflects the sender’s perceptions of
the receiver’s role in the group. Role-receiving occurs when the other person receives the
message and interprets it. Every individual has subjective perceptions of the organization
and his or her place in it.

People can interpret messages in very different ways. This leads to the
deveIOpment of different roles and expectations (Katz & Kahn, 1978). The messages and
intent of the messages interact with perceptions of the message to develop roles for
people. Role taking reflects the interaction between the focal person (employee) and
other members of the role set (workgroup). For example, a member of the role-set, such
as a supervisor, could request that the focal person perform a task. In the role episode,
the focal person receives the role from the sender and determines how to respond based
on the content and attitude of the request. The same role sent from the supervisor could
be interpreted differently by different focal people. This interpretation is influenced by
each focal person’s role expectations, personal attributes and interpersonal factors (Katz
& Kahn, 1978). Roles are largely based on perceptions of different people in the

organization.

7
The role-set perceives each employee within it as an individual with varying

strengths, weakness and relationships. For instance, employees have differing latitudes to
make mistakes or unusual suggestions. In 1958, Hollander suggested that workers earn
idiosyncratic credits when they are good, contributing members to the group or
organization. As they accumulate these credits, employees can exchange them for the
opportunity to exhibit unusual behavior without incurring sanctions for violating norms.
He also indicated that workers must establish credibility before implementing changes or
exhibiting unusual behavior (Hollander, 195 8). This is one example of employees having
different roles, expectations and demands.

More recent research has focused on how roles develop between supervisors and
employees (Graen & Scandura, 1987) and between same-level team members (Seers,
1989). This research will be discussed later in the paper as part of the description of the
new construct of Horizontal Team Member (HMX) exchange. The key ideas presented
by the role literature is that roles are socially determined, encompass task elements of
jobs and vary between members of the role-set.

Defining Role Innovation

In order to effectively study role innovation, a solid definition of it was needed.
The definition had to explain the construct in a thorough, concrete and researchable
manner and distinguish role innovation from related constructs. The previous discussion
of roles pointed to two critical factors that needed to be incorporated into a definition of
role innovation. The first factor was the idea that roles include the elements of a job

(Ilgen & Hollenbeck, 1991), therefore role innovation needed to incorporate job elements

8
by being task-focused. The second factor was that roles are socially determined (Katz &

Kahn, 1978), and therefore role innovation needed a social element. I examined the
definitions of role innovation in the current literature and tried to demonstrate how they
fail to include one or both of these important elements.

Existing definitions

Past researchers provided definitions of role innovation that contained different
elements and levels of detail. Van Maanen and Schein (1979) discussed role innovation
as one possible outcome of socialization. They conceptually defined role innovation as
the rejection and redefinition of the major premises concerning the missions and
strategies that were followed by the majority of role occupants to both practice and justify
their present role. This definition focused on strategic, occupational-level aspects such as
changes in missions and goals. It is not task-focused and does not explicitly include a
social element. The definition reflects some of Schein’s earlier interest in role innovation
at the professional-organization level (Schein, 1971). Schein (1971) described role
innovation in the case of a professional who wanted to redefine the norms of his or her
profession.

West (1987b) presented a more simplistic definition. He defined role innovation
as the introduction of significant new behaviors into a role. This definition did not
specify that role innovation should be task-focused and contain a social element. The
lack of specificity in the definition would make it hard to operationalize. Farr and Ford
(1990) defined role innovation as “the intentional introduction within one’s work role of

new and useful ideas, processes, products or procedures.” (63) This definition does

9
provide one important distinction. It distinguishes between role innovation and the

emergent elements included in the definition of roles. Role innovation behaviors are
intentional, whereas the emergent elements pertaining to roles often develop rather
haphazardly. This definition also dictates that the individual must initiate role
innovation; it is not a organizationally imposed event. This definition focuses on the
active involvement of the individual and the outcomes of role innovation such as ideas,
processes, products or procedures.

Nicholson (1984) also presented a definition of role innovation, but he labeled it
role development. He defined role development as a proactive strategy in which a person
tries to change role requirements to better match his or her needs, abilities and identity.
This definition captures the proactive nature of role innovation and the motivation for it.
However, the definition neglects to identify a level of focus (task-focus) or the
importance of social influence.

Construct Definition

A definition of role innovation should incorporate traditional elements of the
definition as well as be task-focused and include social elements. For the purposes of this
paper, role innovation was defined as:

“A job holder’s proactive process of changing role expectations and

interaction within a workgroup with the intention to introduce new and

useful ideas, processes or procedures for completing tasks.”

With this definition, I have tried to make six key points. The definition captured

the two critical elements of role innovation: task-focus and social influence. The first key

10
element of the definition is that role innovation is task-focused. Role innovation occurs

as the employee is trying to find a better way to do the job. Role innovation would result
in changes to the job tasks through behaviors such as reorganizing processes, introducing
new ideas and re-defining work relationships with other people. Second, the social nature
of role innovation indicates that it can also influence interactions with the workgroup
(role-set). The social influence is reflected by changes such as re-defined interpersonal
interactions, shifts in responsibility, and related role innovations by others in order to
coordinate an interrelated process.

The third key point is that role innovation is a voluntary behavior initiated by the
individual, not one dictated by the organization. This is the “grassroots” nature of role
innovation. Role innovation springs from the fundamental unit of the organization - the
individual employee. Fourth, role innovation is an intentional behavior. The
intentionality of role innovation helps distinguish it from the emergent elements included
in the basic definition of roles. The fifth point is that the intent of role innovation should
be to improve or benefit the job or work situation. Sometimes changes have unexpected
negative outcomes, but role innovation must be distinguished from negatively-intentioned
role changes such as sabotage or withholding effort (Kidwell & Bennett, 1993). Finally,
it is important to note that role innovation is a process.

Distinction from related constructs
When defining a construct it is also important to set boundaries that distinguish it

from related concepts. Though many similarities exist, I will argue that role innovation is

11
a distinct concept from the standard treatments of the three following concepts: extra-role

behavior, job design (job enlargement), and product innovation.

Extra-role behaviors (ERBs) are positive, discretionary behaviors which are not
included within a formal job description (Van Dyne, Cummings and Parks, 1995). ERBs
are usually defined as positive, helping behaviors which are directed at another person or
the organization. ERBs include non-required behavior such as helping co-workers,
maintaining positive attitudes, keeping workgroups members informed, and participating
in company politics and events. Because role innovation focuses on task changes, not
helping behaviors, it is distinct from extra-role behaviors. The recipient of the behavior
also differs. Role innovation is focused on the individual’s own job. People role
innovate in order to improve their own position, relationships or job tasks. Extra-role
behavior is focused outwardly at co-workers or the work group (Van Dyne et al., 1995).

Since role innovation adds and/ or changes the tasks of a job, it is similar to job
design efforts focused on job enlargement. Based on job design principles (Hackman &
Oldham, 1980), job enlargement involves combining jobs and adding ancillary duties to
jobs (Campion & McClelland, 1991). Role innovation differs from job enlargement in
three distinct ways. First, job enlargement is usually a change that is designed and
implemented by management (e.g. Campion & McClelland, 1991). This formal, top-
down implementation is strikingly different than the grassroots, informal nature of role
innovation. Second, job enlargement means adding new tasks to a job, whereas role

innovation sometimes just involves altering existing tasks. Third, the job design concepts

12
focus on single jobs or a class of jobs. Role innovation focuses on tasks and creates a

link among roles.

The study of individual innovation often centers on product invention or
development (e.g. Scott & Bruce, 1994; Ettlie & Reza, 1992). Research and development
laboratories create new technological marvels and products. Product innovation is
generally limited to a few research and design and manufacturing organizations (e.g.
Galbraith, 1982; Van De Ven, 1986). Role innovation can potentially occur in all types
of organizations within all job categories. Another fundamental difference is that product
innovation is initiated by the organization and is often someone’s assigned job (Staw,
1990). The definition of role innovation stresses that it is a grassroots effort initiated by
the individual which is outside of the job description.

Though I have tried to show that role innovation is distinct from these three
concepts, I recognize that some similarities exist. Role innovation and extra-role
behavior both concern behaviors that are not a formal part of the job. People do not
engage in them with the expectation of a formal reward. It is possible that people who
engage in extra-role behaviors will also role innovate when they are dissatisfied with their
role. Job design and role innovation both focus on improving the job. Product
innovation and role innovation both involve the introduction of something new into the

organization.

LITERATURE REVIEW OF ROLE INNOVATION

When studying role innovation, the critical question is: Why would an employee
role innovate? Why would an employee expend extra effort, thought and time in order to
change his or her work role when no formal reward was attached? This study does not
presume to account for all sources of motivation for role innovation. Undiscussed,
additional motivations for role innovation might include avoiding task interaction with
poor performers (Farr & Ford, 1990; West, 1987b), minimizing interpersonal conflict and
desiring to implement a new technology or technique.

However, much of the literature on role innovation focused on issues of poor task
design and altering the method of accomplishing tasks. Therefore, I will discuss most of
the literature by linking it to the concept of task frustration. Task frustration occurs when
a worker experiences distress about performing the tasks of his or her job. Task
frustration can be caused by poorly defined tasks, outdated procedures, time-consuming
interactions, blockages toward goals and many other circumstances. All of these task
frustrations could encourage a worker to role innovate. By role innovating, the worker
would hope to reduce frustration and improve his or her job situation. Since I am
focusing on role innovation in the workplace, task frustration is a major theme in this
paper and in the role innovation literature.

Poor job design can quite easily lead to task frustration. Many existing jobs were
not developed using formal job design techniques. On the contrary, jobs often develop
and stabilize for arbitrary reasons (Ilgen & Hollenbeck, 1991). Since little planning

occurred, jobs often have unproductive idiosyncrasies and illogical processes or demands.

13

14
Even jobs that originally received job design planning exist in a business environment

with rapidly changing demands (Kieman, 1993). The job design often does not keep pace
with the changing demands. Consequently, most jobs have some poorly designed,
unmotivating features (Hackman & Oldham, 1980). Logically, the person best qualified
to recognize frustrating problems is the person closest to them - the employee.

Changing task design was the primary focus of some of the first research done on
role innovation (West, 1987b). West (1987b) conducted a longitudinal field study of
1700 male and female British managers to measure changes in work-roles and outcomes
such as satisfaction and personal change. West (1987b) sent the managers a survey that
asked them to indicate how differently they perform the job than the previous job holder
did. This study found a positive correlation between role innovation and satisfaction and
also with personal change. However, the key fact here is that West (1987a) perceived
role innovation essentially as ‘making a change to a task’ and divided it into six
categories of behavior. The six role innovation categories were: (1) Setting new work
targets/ objectives; (2) Deciding the methods used to achieve work targets/ objectives; (3)
Deciding the order in which different parts of the job are done; (4) Choosing whom you
work with in order to carry out your work duties; (5) Initiating new procedures or
information systems; (6) Developing innovative ways of accomplishing targets/
objectives (West, 1987a).

West (1987a) found strong reliability and validity for the measure of role
innovation containing these six behaviors. This work contributed significantly to the role

innovation literature by defining task change behaviors and providing a measure.

15
However, the self-report surveys assessed role innovation by comparing the current

employee with the previous job holder (West, 1987b). My conceptualization of role
innovation considers it as a personal process in which the employee changes his or her
own role. Of course, the personal change occurs within a social context, so the impact on
co-workers must be considered. Since I define role innovation to be task-focused, task
frustration should encourage an employee to role innovate. Following many traditional
models, I believe that task frustration is a major catalyst of role innovation.

Staw & Boettger (1990) researched a task-focused construct called task revision,
under the construct of counter-role behavior, that has many similarities to role innovation.
“Counter-role” behavior was defined as behavior that differs from expectations (Staw &
Boettger, 1990). Counter-role behavior included a range of actions including deviance,
dissent, role innovation and task revision. Each of these behaviors can be appropriate in
certain situations. The researchers indicated that when a role is incorrectly specified, then
counter-role behavior (role innovation/ task revision) can result in excellent performance
(Staw & Boettger, 1990). Role innovation was used by employees to fix a frustrating
task and improve their performance. This is another example of the link between role
innovation and task frustration.

Staw & Boettger (1990) defined the construct of task revision as, “...taking action
to correct a faulty procedure, inaccurate job description, or dysfunctional role
expectation.” (537) This definition of task revision captures the task-focus of my role
innovation definition, but does not explicitly encompass the social element. Staw &

Boettger (1990) tested their task revision construct in two laboratory studies. They used a

16
communications officer exercise in which subjects were instructed to re-write a recruiting

brochure with differing types and sources of information. The first study indicated that
subjects did more task revision when alternative solutions were highly salient and when
goals were general rather than specific (Staw & Boettger, 1990).

In the second study, the researchers delved deeper into their finding that specific
goals, such as those advocated by the goal setting literature (e.g. Locke & Latharn, 1990),
can operate as a control mechanism to suppress useful forms of counter-role behavior
(Staw & Boettger, 1990). They supported their previous findings in that specific goals
only encouraged revisions in the target areas whereas ‘do your best’ goals encouraged
revisions in all areas. One of the general findings of the research was that there was a low
base rate of task revision, and control mechanisms such as goal setting and conformity, to
norms restricted it even further (Staw & Boettger, 1990).

Along with task frustration concepts, other reasons for role innovation have been
investigated. Role innovation can occur with individuals who wish to have more control
over their tasks (West & Rushton, 1989). Higher need for control has been associated
with higher self-reports of role innovation (West & Rushton, 1989). Need for control is
affected by the amount of discretion in the role. Low discretion roles offer fewer
opportunities for the employee to change the role, whereas some high discretion roles
almost demand it (Nicholson, 1984). West and Rushton (1989) studied 145 student
nurses in a low discretion atmosphere. The low discretion atmosphere did limit role
innovation and forced the student nurses to adapt to the environment through personal

change (West & Rushton, 1989).

17
Some of the earliest research investigated the effects of different socialization

practices on role innovation. Schein (1968) defined organizational socialization as “the
process of ‘learning the ropes’, the process of being indoctrinated and trained, the process
of being taught what is important in an organization...” (p. 2) Van Maanen and Schein
(1979) described six structural dimensions of the socialization setting which could foster
either conformity or innovation. Each dimension has two polar organizational strategies
for socializing newcomers. The six dimensions are: formal-informal, individual-
collective, serial-disjunctive, investiture-divestimre, sequential-non—sequential and fixed-
variable (Van Maanen & Schein, 1979). An organization can influence the development
of it’s workers by choosing to use different combinations of these strategies. For
example, a socialization process which is individual, informal, disjunctive and involves
investiture processes might lead to more innovative behavior. Jones (1986) found that an
“institutionalized strategy” consisting of a collective-fonnal-sequential pattern does lead
to poor affective adjustment and less innovative behavior than the individualized
approach. A recent empirical study also found that institutionalized socialization
(collective, formal, sequential, fixed, serial) led to less attempted and less actual role
innovation (Ashforth & Saks, 1996).

The role innovation literature is in its infancy. Researchers are struggling with the
boundaries and definition of the construct. Most of the empirical research on role
innovation has treated it as one of many outcomes and has not focused on role innovation
as a process of its own. This studied investigated two organizational factors expected to

influence role innovation as well as investigating the process of role innovation.

ORGANIZATIONAL FACTORS
Using the construct definition of role innovation and the existing literature, this
study tried to address two research questions. The first question was: Will the
organizational factors of climate for innovation and goal focus (group versus individual
goals) affect role innovation?

Climate for Innovation

 

Organizations differ on the amount of pressure for conformity or for innovation
that employees face in the workplace. Although role episodes occur between members of
the workgroup, the climate of the organization affects the freedom of these relationships.
Climate is often viewed as a mediator between the organizational context and individual
response (Kozlowski & Hults, 1987; James & Jones, 1974). Climate links the
organization and individual by relaying the norms and expectations of the organization.

Climate at the individual level refers to the person’s cognitive interpretation of the
organizational situation and has been labeled “psychological climate” (James, James, &
Ashe, 1990). Employees use information provided by the psychological climate to
determine what behaviors are acceptable and gauge the possible outcomes of actions
(Scott & Bruce, 1994). Organizational climate encompasses situational characteristics
such as context, structure and processes. The important feature of organizational climate
is that it is built through consensus of individual’s psychological climate. The consensual
nature of organizational climate provides a basis for collective responses (Kozlowski &
Hults, 1987). Organizational climate provides a framework for considering motivation at

a collective level (Kozlowski & Hults, 1987).

18

19
Some organizations have a climate that encourages innovation and which should

also encourage role innovation. Climate is a broadly-defined term that does not allow
easy conceptualization. It is critical to define the dimension (i.e. climate for safety) that
is being studied by the climate measure (Kozlowski & Hults, 1987; Schneider, 1985).
This study focused specifically on the climate for innovation.

The climate for innovation has a strong influence on innovative behavior
(Kozlowski & Hults; 1987; Kanter, 1983). Kanter (1983) noted that innovative
companies have a ‘culture of pride, climate of success’. Past research concerning
technical innovation and updating indicated that climate can either facilitate or discourage
innovative behavior, updating activities and job performance (Kozlowski & Hults, 1987).
Without a supportive climate, role innovation becomes an employee’s struggle against
the organization to make desired changes. In an empirical study of innovation, Scott &
Bruce (1994) found a significant correlation between innovative behavior and support for
innovation within a research and development laboratory.

Organizations with a climate for innovation create a norm for innovation.
Burningham and West (1995) conducted a study to measure predictors of team-level
innovation using a sample of oil company workers in the United Kingdom. They
determined that support for innovation was one of the principle predictors of group
innovation. Other predictors included vision, participative safety and task orientation
(Burningham & West, 1995). West (1990) distinguished between articulated and enacted
support for innovation. The support included verbal support, cooperation and providing

time and resources (West, 1990). Companies must do more than say they support

20
innovation. They must provide support, resources and encouragement for it. The

importance of climate for innovation also applies to role innovation. Employees must
feel that change and variation are acceptable within the organization. Organizational
norms that encourage change may even encourage members to role innovate in order to
be culturally appropriate (Farr & Ford, 1990).
Goal Focus

In addition to climate for innovation, goal focus is an organizational factor that
should influence role innovation. In this study, goal focus distinguished between group
level goals and individual level goals. Organizations institute a goal focus among
workers either by setting individual goals and rewards or by setting group goals and
rewards. This study’s role innovation definition emphasized both individual task
elements and social interaction, and these two distinct aspects of the role innovation
definition are partially captured in the dichotomy between individual and group goals.
This makes goal focus a uniquely suited context for considering different influences on
role innovation. From a task perspective, goals can help focus worker’s attention on
different job elements (Locke, Durham, Poon & Weldon, in press) and may impact role
innovation of job tasks. From a social perspective, group goals inherently have a stronger
social element than individual goals.

Before considering the impact of goal focus on role innovation, it is important to
consider the current goal literature. Most of the current goal literature focuses on goal-

setting tactics and processes. This section should show that goals are important to

21
performance and that group and individual goals have different impacts on outcomes

especially for interdependent tasks.

Considerable work has been done on goals and goal setting in the past decade (see
Locke & Latham, 1990 for a discussion). The relationship between goals and
performance has been found in many studies which indicates the importance of goals for
organizations. Much of this research has focused on goal-setting (Latham & Locke,
1991) which examines the specificity and difficulty of goals as well as the process for
setting the goals and following up on them. The main findings of this goal-setting
research include: (1) specific, difficult goals lead to higher levels of task performance
than “do your best” goals; (2) feedback strengthens the goal-performance relationship; (3)
goals impact performance most when commitment to the goal exists; (4) assigned goals
are just as effective as participatively set goals if a rationale is given for the goal; (5) self-
efficacy effects performance and goal setting (Locke et al., in press).

Although this goal-setting research has contributed significantly to the goal
literature, it does have its limitations. Two limitations are that most of the research has
been conducted at the individual level and most of the research has been done using
independent tasks (Locke et al., in press). It is important to begin conducting more goal
studies involving group goals and interdependent tasks. Increased teamwork and
interdependence in today’s organizations create a need for group level goals. The
individual goal-setting literature provides a starting place for research on group goals.
However, group and individual level goals should encourage different types of behavior

especially with interdependent tasks. The need for group-level research is seen by the

22
increased use of groups and teams by American organizations. The last decade has seen a

shift in organizational structure in the United States to more group-level work and task
interdependence. Many organizations today are stressing teamwork, groups and
cooperation in the workplace (Guzzo & Shea, 1992; Wagner, 1995; Shamir, 1990,
O’Leary-Kelly, Martucchio & Frink, 1994).

Interdependent tasks, such as those faced by most teams, require a different type
of goal setting than that of individual, independent tasks. However, the distinction
between independent and interdependent tasks and workgroups has not always been clear.
Before researching group goals, it is important to decide what type of interdependence the
group will have. In the research, the definition of interdependent workgroups has varied
(Keman, Bruning & Miller-Guhde, 1994). In one interdependent task study, participants
worked together on group goals but were not allowed to communicate (Mitchell & Silver,
1990). In another study (Weldon & Weingardt, 1988), participants were allowed to
communicate and individual contributions and interaction remained anonymous.

Mitchell and Silver (1990) made a further distinction between goal interdependence and
task interdependence. They described goal interdependence as group members sharing a
common goal whether or not they actually work together. Task interdependence occurs
when group member must work together to accomplish a group goal. In order to reflect
the reality of most groups in organizations, interdependent tasks should have group
member interaction and communication (Keman et al., 1994). This type of
interdependent task creates a perfect opportunity to study group level goals. All member

of the group should have an investment in the task and goal.

23
Although less research has been done on group-level goal setting, some progress

has been made. Preliminary results indicate that group goals can improve performance
on interdependent tasks (Weldon & Weingardt, 1988). Consistent with individual goal
setting, studies have shown that groups perform better when they have specific, difficult
goals (Weingardt & Weldon, 1991; Locke et al., in press). A meta-analysis that included
26 effect sizes from 10 studies supports the idea that groups with goals perform better
than groups without goals (O’Leary-Kelly et al., 1994). Further research indicates that
the relationship between group goals and performance is influenced by several variables
such as task complexity, group member effort and planning (Weingardt, 1992).

Most of the goal research has focused on goal setting and related issues. Few
studies have examined the differential impact of individual versus group goals. However,
Mitchell and Silver (1990) examined the interaction between group and individual goals.
They found that group goals plus individual goals resulted in high performance on
interdependent tasks. However, group goals worked as well as group plus individual
goals and both performed better than individual goals. They concluded that individual
goals can cause competitive feelings and behavior which can lower performance on an
interdependent task (Mitchell & Silver, 1990).

In summary, this section has shown that goals are important to performance.
Significant amounts of research have shown the optimal way to design and set individual-
level goals. Some of this research has also been translated to work with group-level
goals. However, the critical point is that individual and group levels goals should cause

people to behave differently. This difference should occur as people with individual

24
goals should focus primarily on the individual aspect of the task and give only secondary

attention to the group-level, interdependent aspect of the task. People with group level
goals should focus on the group-level goal but also complete individual tasks since they
are necessary to the group-level task.

Since goals affect behavior, they should effect role innovation as well. People
given group goals should focus on the group task. If the task is interdependent, people
with group goals should be more likely to discuss issues with other group members and
to seek solutions to group problems. They would be more likely to make role innovations
that improve the task for the group as well as for themselves. However, with an
individual goal, people working on interdependent tasks might only role innovate to help
themselves and not in order to benefit the group. For interdependent tasks, group goals
are likely to encourage more role innovation, since changes could be made to individual

as well as group elements of the task.

PROCESS OF ROLE INNOVATION

The second research question is: How do HMX and role negotiation contribute to
the role innovation process in interdependent workgroups? This process reflects the
social nature of role innovation - other workers should be (but sometimes are not)
considered and consulted when a worker makes a task change, especially on an
interdependent task. Besides role negotiation, the process of role innovation involves
other steps. The first process step occurs when the worker solves the problem and
proposes the role innovation idea.

In order to role innovate, the employee must be able to devise a solution for the
task frustration that he or she has encountered. The problem solving step in role
innovation should be rather simple. All problem solving situations require some ability,
but the solutions to most role innovation problems should be fairly obvious (Farr & Ford,
1990). Some researchers (Staw & Boettger, 1990) even believe that role “innovation” is
somewhat of a misnomer, since the appropriate behavior is rather obvious and does not
always require much creativity. Farr & Ford (1990) recognized that the innovative aspect
of role innovation must be considered on a continuous scale. The ‘newer’ and ‘more
useful’ that the change is; the more innovative it can be considered. Since role innovation
solutions are often obvious, most individuals should be capable of initiating them (Farr &
Ford, 1990). Other researchers would add that all creativity and innovation result more
from hard work than from creative genius (Staw, 1990).

Usually one worker will decide to role innovate and devise a solution. However,

the new idea or behavior usually has implications for all members of the workgroup —

25

26
especially when an interdependent task is involved. The social nature of role innovation

indicates that discussion should accompany proposed changes, since most changes will
directly or indirectly impact the entire workgroup. The success of this negotiation
depends heavily on the quality of the relationship between the initiator and the others in
the workgroup. Role innovation moves from being one worker’s idea to being a
discussion and negotiation point for the group. The discussion functions as both a
“reality check” that keeps changes within organizational norms and a time for the role
innovator to persuade others in the workgroup to accept the change.

Negotiation is important since workgroup members are dependent on one another
on the job. When jobs are highly interdependent, related activities should be considered
before role innovation occurs (Farr & Ford, 1990). Part of the negotiation phase must
include support for good ideas and the winnowing of weak ideas (Staw, 1990). In some
organizations, proposed ideas must have ‘idea champions’ that support and promote the
idea in order to make it a reality (Kanter, 1983). At the role innovation level, the idea
champion would be the role innovator. The key point to remember here is that role
innovators usually face some level of opposition, but they must be willing to persevere to
get the change implemented (Staw & Boettger, 1990). If the workgroup disputes a
proposed role innovation, the employee must decide whether their arguments are valid or
not. This decision will help him or her decide whether or not to proceed with the role
innovation. The entire negotiation process depends on the relationships that the role
innovator has with the workgroup. One way to study the process of role negotiation is to

examine the quality of the relationship between the role innovator and workgroup. Role

27
innovators with positive relationships should have more success in implementing ideas.

Members with poor relationships should experience difficulty during the negotiation
phase and therefore be less likely to implement their proposed role innovation.

In recent years, researchers have studied relationship quality between workgroup
members by using leader-member exchange (LMX; Graen & Scandura, 1987) and team-
member exchange (TMX; Seers, 1989). Building on this research, Nason (1995)
developed a new construct called horizontal member exchange (HMX) to study the
quality of relationship between same-level workgroup members.

LMX and TMX

Recent research indicates that the quality of a role relationship can be defined and
measured through leader-member exchange (Graen & Scandura, 1987) and team-member
exchange (Seers, 1989). These theories provide an explanation for the potential success
and failure of role negotiation attempts. Leader-member exchange (LMX) is a theory that
applies role theory to the relationship between supervisors and subordinates. LMX
involves the negotiation process which affects the quality of working relationships
(N ason, 1995). Leaders exchange resources (i.e. information, assignment, support) with
subordinates in order to elicit desired behavior. This exchange process creates a unique
relationship between the leader and each subordinate (Dansereau, Graen & Haga, 1975).
Subordinates with positive LMX relationships become part of the leader’s in-group
(preferred members) and receive special attention and responsibilities. Leaders often ask

in-group members to perform coveted extra duties that the leader does not have time to

28
perform (N ason, 1995). On the other hand, leaders usually ask out-group members to

perform only the routine tasks contained in the accepted job description (Nason, 1995).

Graen and Scandura (1987) added a role routinization stage to the Katz and Kahn
role theory. In the role routinization stage, the LMX interactions become predictable and
follow an ordered sequence (Graen and Scandura, 1987). Role innovation would cause a
disruption of the role routinization stage as the employee pushes to re-negotiate the role.
Early research measured LMX by considering the amount of negotiating latitude that the
subordinate had in defining the role (Nason, 1995). Out-group members did not possess
much influence with the leader, so their negotiating latitude was low. This low
negotiating latitude limited the subordinate’s ability to affect his or her role with the
leader. In-group members had high negotiating latitude and therefore had more power to
define their role with the leader. It is important to note that out-group members may have
to be more assertive in order to obtain desired behaviors (N ason, 1995).

Seers (1989) developed the team-member exchange (TMX) construct based on the
concept of LMX. TMX does not reflect a dyadic relationship such as that between a
supervisor and subordinate. It states that the entire peer group transmits role expectations
to the focal team member (N ason, 1995). TMX captures the quality of the relationship
between the target member and the rest of the peer group. TMX reflects the relationship
between peers with relatively equal status, so the exchange processes should be more
balanced (Seers, 1989). It measures the peer group perceptions of the target member’s
willingness to help and share with other members (Seers, 1989) This same level of

information-sharing, help and recognition should also be sent to the target member from

29
the peers. This type of information sharing and support affects the effectiveness of the

member’s working relationship with the peer group (Seers, 1989).

An empirical study by Seers (1989) indicated that TMX had differential validity
from LMX and that TMX accounted for significant proportions of variance in work, co-
worker and general satisfaction beyond the LMX construct (Nason, 1995). The research
also showed that the quality of the member’s relationship with the peer group accounted
for variance in the supervisor’s rating of individual performance (Seers, 1989). Nason
(1995) stressed that role behavior serves as the primary currency of team-member
exchange. Interdependent team members rely on compliant role behavior in order to
accomplish their tasks (Nason, 1995). One major limitation of TMX is that it departs
from the traditional dyadic framework of role theory to explore the group interaction
(N ason, 1995). This limitation impairs the ability to understand team member role
behavior and attitudes in the traditional role theory context (Nason, 1995).

Researchers have considered the effect of LMX and TMX on innovative behavior.
Scott & Bruce (1994) found that LMX accounted for 39% of the variance in support for
innovation through a LISREL goodness-of-fit measure, but they did not find a significant
correlation between TMX and innovative behavior. The non-significant TMX results
may be due to conceptual problems, since TMX does not reflect a true role relationship.
It is based on an average relationship between a focal member and his or her peers

(Nason, 1995).

30
Horizontal Team Member Exchange (HMX)

In response to the limitations inherent in TMX, Nason (1995) developed a new
role relationship measure called horizontal team member exchange (HMX) to study
dyadic relationships between team-members. These horizontal relationships differ
between team-members and can affect team and individual level outcomes (N ason, 1995).
HMX has an advantage over TMX in that it follows the traditional boundaries of role
theory (N ason, 1995). HMX relationships develop with the same cycle of sending,
receiving and responding to role expectations detailed by Katz and Kahn (1978). Since
HMX defines the relationship between two people, it retains within-group variance
(N ason, 1995). TMX loses much of the within-group variance by aggregating all
responses to a peer group level.

Through LMX, leaders have the opportunity to offer institutional rewards (pay,
raises, responsibility, information, etc.) in the exchange relationship. HMX relies on
cooperation as the exchange currency for the relationship (N ason, 1995). This permits a
very balanced exchange situation. HMX is appropriate for considering team
relationships, since all members rely on each other to complete an interdependent task -
the leader does not necessarily have extra power at the task-level. Cooperation is
essential when teams are performing interdependent tasks (Nason, 1995). Although
HMX affects the interaction of dyads, the quality of HMX is defined and measured at the
individual level (N ason, 1995). The exchange relationship quality “...is an individual-
level perception of the quality of a working relationship that develops over a series of

dyadic interactions (Nason, 1995 p.41). Nason’s (1995) model of HMX shows that team

31
members indicate their perception of the relationship quality through their

communication behaviors.

In summary, HMX provides a means of measuring the quality of the relationship
between same-level workgroup members. Since workgroup members work
interdependently, their exchange currency is cooperation. Members with better HMX
relationships would have more negotiating power for implementing a role innovation than
a member with a poor relationship. HMX provides a method for capturing the role
negotiation process.

Role Negotiation

HMX provides a good measure of the interaction between co-workers. It can help
determine which workers have the high quality relationship necessary to initiate a role
innovation. But HMX just gauges the quality of the relationships. It is also critical to
consider the actual role negotiation behavior. The process of role negotiation would
probably start with the individual presenting the role innovation idea to the workgroup (or
some sub-set of the workgroup). After the presentation, the workgroup would discuss the
idea in either a formal or informal format. The workgroup would need to consider the
impact of the changes on all inter-related tasks or relationships (Farr & Ford, 1990).

Final approval decision or rejection of the idea could be made by the group as a whole or
by the supervisor. Depending on the situation and the complexity of the change, this
process could take minutes, days or longer. The proposed model assumes that the role
innovation process begins after a role has been established through the original role

making process described by Katz and Kahn (1978). The negotiation that occurs during

32
the role innovation process is intended to revise the existing role. HMX gauges the

quality of the relationships between co-workers and role negotiation refers to the actual
discussion behavior that occurs between co-workers.

Although LMX and TMX indicate that some people will be more successful than
others at role innovation, the whole issue can be avoided by bypassing the role
negotiation stage. Some employees will not consider or consult the workgroup when
making a change. This decision might be acceptable if the employee is working in an
autonomous position, but interdependent jobs could easily suffer from undiscussed role
innovation (Farr & Ford, 1990). By bypassing the scrutiny of workgroup and
organizational norms, the employee’s un-negotiated role innovation can resemble
rebellion (Van Maanen & Schein, 1979). If the employee initiates a role innovation that
negatively impacts the workgroup, then serious problems could arise. For example, some
role innovations require adjustments by other workgroup members. By not allowing
these adjustments, the role innovator could cause confusion and performance problems
for the workgroup. Role negotiation should be a critical part of the role negotiation

process for interdependent tasks.

RESEARCH MODEL AND HYPOTHESES

Very little empirical research exists on role innovation. As a whole, the topic has
not received much attention. Lack of a solid construct definition and research
methodology has hindered work in the area. This paper presented a construct definition
of role innovation which provided boundaries and guidelines for research and will now
present the conceptual model of the influences and process of role innovation (see Figure
1). This model incorporates the organizational variables, HMX and role negotiation
discussed in previous sections and also introduces several individual difference variables
that might influence role innovation. The model assumes that the role innovation process
has been initiated by a catalyst such as task frustration. The model starts with the
influence of organizational factors and ends with improved task performance. (See

Figure 3 at the end of this section for the model marked with hypotheses).

33

34

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

Leaming Style
Orientation
lndividualism/
Collectivism
rf
Climate for P8 ormance
Innovation { l
I v
I Horizontal R l R l
Member __. o e __+ o e.
I, Exchange Negotiation Innovation
Goal I T
Focus
Task

 

 

     
 

 

Satisfaction

Figure 1: Model of Role Innovation

35
After the role innovation process is initiated (often with task frustration as a

catalyst), the organizational factors of climate for innovation and goal focus should have
an effect. Climate for innovation indicates whether or not a workplace will support,
encourage and accept role innovations. Goal focus indicates whether or not the
organization encourages cooperation and teamwork which affects the social and
negotiation aspects of role innovation. When a role innovation idea survives the
organizational factors, the model predicts that the worker will negotiate the proposed
change with other workgroup members. This is an important process step, especially
when workgroups and tasks are highly interdependent. This negotiation is dependent on
the quality of the relationship (HMX) between the workers. After negotiation, the role
innovation is implemented. Assuming that the role innovation corrected a flaw in the
task, performance on the task should improve. Specific hypotheses will be developed for
each of these ideas.
Clim_ate for Innovation

In the discussion of climate, I argued that climate for innovation has a strong
influence on innovation. Climate relates the norms and values of the organization to
individual workers. Climate indicates what behaviors or types of behaviors are
acceptable. This impact of climate on innovation should also extend to role innovation.
Role innovators face issues of introducing a change into an organizational context, and
some organizations will support a worker’s role innovation and appreciate the proactive
job involvement. Other organizations might perceive role innovation as a threat to the

status quo and attempt to block or discourage it. Role innovation is a grassroots, personal

36
action initiated by the individual. However, all individuals function within organizations

that affect their behavior and decisions. Therefore it is expected that climate for
innovation will affect role innovation.
Hypothesis 1 .' Participants in a climate that supports innovation will engage in more
role innovation than participants in a climate that does not support innovation.
Goal Focus

My construct definition stresses both the task-focus and social nature of role
innovation and goals incorporate both of these elements. As for the task focus, goals
have been shown to focus attention on tasks and generate higher performance (Locke et
al., in press). Although the social element is a critical part of all roles and role
innovation, most earlier researchers did not recognize the social element. The importance
of the social elements might be investigated by considering the differential impact of
group and individual goals. Research has concluded that individual goals can create
competition and detriment performance on interdependent tasks (Mitchell & Silver,
1990). This competitive feeling should cause people to focus only on their own situation
and not the social situation. Therefore, workers with individual goals are expected to
only role innovate their own specific task and not consider the group needs.

Group goals focus participants on the shared task and should encourage
c00peration which should improve performance on interdependent tasks. This
interdependence and c00peration should stimulate more role innovation. Workers will

have an opportunity to role innovate their own tasks as well as role innovate the group

37
tasks. Group goals expand the scope for role innovation which should encourage more

overall role innovation.

Hypothesis 2: Participants with group goals should role innovate more on the
interdependent task than participants with individual goals.

Interaction between Climate and Goal

It is expected that a climate supporting innovation and group goals will result in
more role innovation. However, it is also possible that the interaction between climate
and focus can affect these findings. This difference is expected to occur due to the
control systems inherent in groups. Specifically, groups have considerable power for
keeping members within the norms and standards of the group. If climate for innovation
can act as a group norm, then groups might expect members to behave in the manner
suggested by the climate. This group norm effect is expected to be much stronger in
groups assigned group goals.

It is a well-researched fact that groups exert affective and behavioral control over
individual group members (see Hackman, 1992 for a discussion). Groups often strive for
uniformity among members. This uniformity can be created through socialization (see
Fisher, 1986 for review) and through on-going pressure and norms (Hackman, 1992).
Role constraints and standard operating procedures can also stifle variation (Brett, 1984).
Uniformity can help a group to run smoothly and with minimal conflict (Hackman,
1992). However, the power of group conformity can have negative effects as well. Janis
(1982) coined the term “groupthink” to describe how the pressure to conform can keep

individuals from voicing critical dissenting opinions. Janis (1982) cited the disastrous

38
Bay of Pigs decision during the Kennedy administration as an example. In that case,

dissension was discouraged during planning meetings and critical objections were never
presented (Hackman, 1992).

A classic study by Asch (1951) demonstrated the power of the group to cause
individual conformity. In this study, subjects had the simple task of matching the length
of a line to one of three presented lines. Subjects had to state their opinion out loud in a
roomful of confederate subjects. The subject was often faced with a unanimous majority
of “peers” who were all giving the incorrect answer. Approximately one third of the real
subjects chose the obviously wrong answer in order to agree with the majority (Hackman,
1992). All of this research demonstrates the power of the group to keep the individual
from deviating from the norms or standards. These powerful group norms could have an
added impact on the groups with group-level goals.

The research shows that groups create norms which affect the behavior of
individuals. However, these norms can affect a variety of behavior. For example, if the
organization had a climate for innovation, the group might make innovative behavior a
norm. The group might pressure individuals to conform to the norm and encourage
creativity, innovation and change. The power of group conformity would be to
encourage change instead of maintaining the status quo. Some organizations, such as
3M, do have a climate that support innovation and actively encourage and reward the
creativity of employees (Kanter, 1983). Farr and Ford (1990) described this as a norm for
innovation. With a norm for innovation, members would role innovate in order to be

culturally appropriate. If climate is treated as a norm, it should have a stronger impact on

39
workers with group level goals. This could create an interesting interaction between the

two variables.

As Figure 2 shows, participants with group goals are expected to role innovate
less than participants with individual goals when in the traditional setting. However,
participants with group goals are expected to role innovate more than participants with
individual goals in the innovative climate. This interaction is expected due to the extra
control power of group norms among the group goal participants. Workers with group
goals would move toward the innovative or traditional norms established by the climate.
This would create more extreme changes in role innovative behavior for workers with
group goals depending on climate.

Hypothesis 3: Climate for innovation and goal focus should interact. In a climate
supporting innovation, participants with group level goals are expected to engage in
more role innovation than participants with individual level goals. In a climate that does
not support innovation, participants with group level goals are expected to do less role

innovation than participants with an individual level goal.

40

Group Goal
Amount of
Role Individual Goal
Innovation

 

 

Climate Climate
Against Supporting
Innovation Innovation

Figure 2: Interaction of Climate and Goal Focus

41

As the use of teams and workgroups increases, workers are often finding
themselves working on interdependent tasks. This interdependence means that the
behavior of one worker will often affect the work of the others in the workgroup. Role
innovation occurs within these interdependent situation as is seen by the social element
within the construct definition. One method of studying this social element is by
considering the quality of the relationship between co-workers. Quality of the
relationships can be conceptualized by using horizontal team member exchange (HMX).

Organizational climate should have an impact on HMX, since some organization
will encourage help and positive interaction and others will discourage it. Innovative
organizations should encourage teamwork and positive interaction. They are more likely
to want employees to work together to solve problems and reach the goals of the
organization.

Hypothesis 4: Participants in innovative organizations should have better quality
relationships (higher HMX scores) than participants in traditional organizations.

HMX impacts the negotiation and decision making, since workers with higher
quality relationships and more negotiating power are more likely to find approval for
their ideas. Workers with group goals have more incentive to interact, so they should
develop better quality relationships. Individual goals have been shown to foster
competitive feelings which would lead to poorer quality relationships.

Hypothesis 5: Participants with group goals should have better quality relationships

(higher HMX scores) than participants with individual goals.

42
People with higher quality relationships should be more likely to actually discuss

issues with their workgroup. They are more certain of being received well and having
their ideas supported. This discussion refers to the actual role negotiation behavior.
Hypothesis 6: Participants with higher HMX scores should role negotiate more than
participants with lower HMX scores.

Role Negotiation

When the task is interdependent, a role innovator should discuss and justify a
potential change to the other workers. If they agree to the change, the other workers may
have to role innovate themselves in order to accommodate the change. Role negotiation
can be studied by documenting the amount of discussion or actual role negotiation
behavior occurs. Workers with group goals have more opportunity to interact and
therefore have more Opportunity to discuss issues.

With an interdependent task, people have more opportunity to work together as
well as do their own work. This opens up additional opportunities for role innovation.
The worker can change their own task as well as change the group task or interaction.
Hypothesis 7: Participants who do more role negotiation should role innovate more than
participants who do less role negotiation.

Performance

The final outcomes for the model indicate that task performance should improve
after role innovation. This is based on the idea that frustrating tasks are often caused by
poor design. When the design is changed (hopefully improved) the employee should

encounter fewer roadblocks and therefore perform better. Staw & Boettger (1990) did

43
indicate that when a role is incorrectly specified, counter-role behavior (i.e. role

innovation/ task revision) can result in excellent performance.
Iflpothesis 8: Participants who role innovate more on the task should perform better on
the task than participants who role innovate less.
Task Satisfaction

If the employee has role innovated a task to suit his or her needs, then the worker
should find the changed task to be more satisfying. This is expected to occur if the role
innovation process was successful and the new behavior improved the task. The
participant has taken control of improving his or her task by role innovating. People who
do not role innovate are likely to still be frustrated by the task
Hypothesis 9: Participants who role innovate more should report more task satisfaction
than participants who role innovate less.
Individfiual Differences: Role innovation has been conceptualized in this study as a job
holder proactively changing his or her job without direct orders from the organization.
Not all people in all situations will be able to role innovate. The main focus of this study
was to examine the process and the organizational factors that encourage or discourage
role innovation. However, common sense indicates that some people may be naturally
more inclined to role innovate than other people. This study explored two individual
difference variables that seemed like they might relate to role innovation. The individual
differences will deal with two important aspects of role innovation: interaction with other
people (individualism-collectivism) and approach to learning the task (mastery versus

performance learning style).

44

Individualism-Collectivism

The first individual difference variable, individualism-collectivism, focuses on the
social aspect of role innovation and determines how people interact with others. This
could distinguish between people who like to work in groups and people who do not like
to work in groups. Individualism-collectivism has many applications in the field of
psychology. The tendency toward having an individual or collective orientation often
depends on the person’s culture (Wagner, 1995), organizational climate and specific task
goals and rewards. An individual’s personal orientation could influence his or her
tendency to role innovate. Role innovation incorporates some individualistic aspects and
some collectivist aspects.

lndividualism-collectivism is often studied at a cultural level. This cross-cultural
research often compares the general tendency of citizens to approach the world with a
collective or individualistic viewpoint (see Wagner, 1995 for a review). Some cultures
such as the United States and Australia tend to instill individualism in people while
countries such as Japan and China tend to encourage collectivism (Wagner, 1995). These
cultural values are instilled in people and become ingrained behavioral patterns.
However, individualism and collectivism can vary at an individual level, since people
adopt different cultural attitudes and vary on basic tendencies.

Many current organizational theories stress individualism. (Wagner, 1995)
Wagner and Moch (1986) define individualism as the condition in which personal
interests are accorded greater importance than are the needs of groups. A certain level of

individualism is required for role innovation, because the employee must be concerned

45
about his or her situation and engaged enough by the task to make a change when

necessary. The construct definition of role innovation provided in this paper encouraged
individualistic tendencies. The definition stressed the point that role innovation focuses
on the employee’s own task and the employee is attempting to improve his or her own
situation. The definition also indicated that the individual initiated the change, not the
organization or group. When an employee role innovates, he or she is changing the status
quo and claiming some autonomy in the organization. Hui and Villareal (1989) surveyed
Chinese and American students and found that collectivism was negatively correlated
with autonomy and positively correlated with need for affiliation.

However, the last decade has seen a shift in organizational structure in the United
States to more team-based work groups and even to organizational interdependence.
Some researchers even suggest that highly individualistic orientations may be
inappropriate for contemporary organizations (Wagner & Moch, 1986). More
organizations are stressing teamwork, groups and cooperation in the workplace (Guzzo &
Shea, 1992; Wagner, 1995; Shamir, 1990). One method of fostering cooperation and
teamwork within an organization is to instill a collective orientation among employees
(Wagner, 1995). Wagner and Moch (1986) define collectivism as occurring when the
demands and interests of groups take precedence over the desires and needs of
individuals. Wagner (1995) conducted a study to determine the impact of a collective
orientation on cooperation and teamwork. He monitored the progress of 492 students
working to prepare a group presentation over a period of weeks. The students rated each

other on a cooperation scale and completed a individualism-collectivism survey about

46
themselves. The results indicated that collectivists cooperated more on highly

interdependent tasks than individualists did (Wagner, 1995). These results highlight the
importance that collectivism can have on encouraging cooperation in today’s team-based
workplace.

Collectivism addresses two important features of role innovation. First,
collectivism captures the social interaction nature of role innovation that is often
overlooked by researchers. Second, it highlights the fact that role innovation must be
negotiated with members of the role set. Collectivism and individualism both capture
unique aspects of the construct definition of role innovation. With interdependent tasks,
people with collectivist outlooks should focus on group goals. If people focus on group
concerns, then the quality of their interaction should improve as reflected by HMX. This
focus on group goals should also encourage more role negotiation and role innovation as
workers try to improve their own tasks as well as the tasks of the entire workgroup. This
focus on the group should encourage more group discussion (role negotiation) and more
focus on improving group performance (role innovation).

Hypothesis 10: Participants with a collectivist outlook will have higher quality
relationships (higher HMX scores) than participants with an individualistic outlook
Hypothesis 1]: Participants with a collectivist outlook will role innovate on the

interdependent task more than participants with an individualistic outlook.

47
Learning Styles

Another important individual difference could be how the person perceives,
approaches and learns the task. Mastery and performance learning orientations
incorporate some of this difference. People with mastery orientations believe that effort
leads to successful outcomes and that their ability for a task can be improved (Ames,
1992; Dweck, 1986). They engage in learning activities in order to understand the new
task or concept and enjoy the challenge of learning. In contrast, people with performance
orientations are less interested in learning the task and more interested in performing
better than their comparison group. Performance oriented people believe that successful
outcomes come from high ability and that their ability for a task is fixed and cannot be
improved (Ames, 1992). Research has indicated that mastery and performance learning
orientations are two conceptually and statistically separate dimensions (Button, Mathieu,
Zajac, 1996).

Much of the research on learning orientations was conducted in classroom
settings. These results indicated that classrooms emphasizing mastery goals encouraged
students to use more effective learning strategies, prefer challenging tasks, approach the
class more positively and believe that success follows from effort. Furthermore, mastery
goals encouraged students to persist when faced with errors or difficulties. In contrast,
classrooms emphasizing performance goals encouraged students to focus on their current
ability level, evaluate their ability negatively, and attribute their failures to lack of ability.

Performance oriented goals lead students to avoid challenging tasks and have less

48
persistence when dealing with difficulties (Ames & Archer, 1988; Dweck, 1986; Dweck

& Leggett, 1988; Elliot & Dweck, 1988).

In the classroom studies, learning orientations were usually manipulated as
situational variables (Dweck, 1986). However, the learning orientations can also be
considered individual differences. A recent paper classified learning orientation as, “...a
somewhat stable individual difference variable that may be influenced by situational
characteristics (Button, et al., 1996).” The dispositional learning orientation predisposes
people to act a certain way, but the response patterns can also vary across situation. In
this paper, learning orientation was treated as an individual difference.

Other research indicated that mastery orientation is related to the belief that
success requires interest, effort , and collaboration (Duda & Nicholls, 1992). This
indicated that mastery orientation could be related to HMX, since HMX is an indicator of
relationship quality. People with mastery orientations should be more interested in
working on the task and more likely to work well with their partners. In contrast, people
with performance orientations might feel more competitive, show less collaboration with
their partners, and not improve the HMX score after interacting. Consistent with
previous research, performance orientation is not expected to have a significant
relationship with HMX, but mastery orientation is expected to have a relationship.
Hypothesis 12: Mastery learning orientation will be positively related to HMX.

Other researchers have discussed the relevancy of learning orientations to role
innovation (Farr, Hoffmann, & Ringenbach, 1993), however this link was not empirically

tested. Their proposition was based on research that indicated that a mastery orientation

49

had caused people to accept more challenging tasks, persist when faced with difficulties,
try to understand the task, and increase their level of competency (Ames, 1992; Dweck,
1986). Since people with mastery orientations are more engaged in the task and more
interested in challenge, they were expected to role innovate more on the task. However,
performance orientation was not expected to relate to role innovation, since people with
performance orientation would only role innovate if it made them look good.
Hypothesis 13: Mastery learning orientation will be positively related to role innovation.
Overall Model

As noted earlier, this model presents role innovation as a process. The process
should be affected by organizational variables (i.e. climate and goals) and individual
differences (i.e. individualism/ collectivism and learning style). The model shows how
quality of co-worker relationships (HMX) can influence the role negotiation for
interdependent tasks. Therefore HMX and role negotiation both mediate the path to the
primary dependent variable - role innovation. Role innovation was also predicted to lead
to secondary outcomes of - performance improvement and task satisfaction. In this
complex process model, several variables act as dependent variables as well as
independent variables. The model itself essentially acts as an overall hypothesis
concerning the expected role innovation process and relationships. The overall process
model will be tested to see if the proposed model provides a good description of the

relationships between the variables.

 

 

50

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Pre
Learning Style ‘3
Orientation
12
Pre
Individualism/ I I
Collectivism
TI
1
Climate for
Innovation W $ l
3 TI 12 T2 I
. Horizontal R l R l
'0 ' Member _, 0 9 ° 9
I. Exchange Negotiation Innovation
TI
Goal 5 T
Focus

 

 

 

Figure 3: Model of Role Innovation with Hypotheses

Performance

     

Satisfaction

METHOD

Participants

The participants were 232 undergraduate students from a large midwestem
university who fulfilled a course requirement by being in the study. Four subjects (two
dyads) were eliminated from the sample. For one dyad, a research assistant failed to
collect all of the necessary scales. For the other dyad, one participant sat through the
sessions but refused to perform the task. The final sample used in the analysis consisted
of 228 participants with cell sample sizes ranging from 56 participants to 58 participants.
Seventy-seven percent of the participants were female, and 54% of the participants were
19 years of age or under.
mg

This research project was designed to study some of the factors that influence role
innovation. Participants took part in a manufacturing simulation that used Tinker Toys to
approximate an assembly line. Participants used a standard operating procedure manual
to build a specified product. The procedure presented for the task was designed to be
frustrating and therefore encourage role innovation. The study investigated how two
organizational factors, climate for innovation and goal focus, influenced the role
innovation process. The design was a 2 x 2, between subjects fully crossed design that
crossed climate for innovation and goal focus. Based on the conceptual model presented
earlier, this study also incorporated role negotiation, HMX and three individual

differences.

51

52
The first production session of the study provided the participants with an

opportunity to learn the task. The second production session involved an unexpected
change to the production schedule that was intended to make role innovation more salient
to the participants. The production sessions were videotaped and later coded for pre-
determined role innovation behaviors and role negotiation.

A preliminary power analysis was conducted (Cohen & Cohen, 1983) using pilot
data to determine the sample size necessary to detect a moderate effect size with a power
of .80 and a significance criterion of .05. The analysis was based on an R2 of .30662
which was determined using pilot data that regressed five variables onto role innovation.
The power analysis indicated that 144 participants were necessary to examine the
hypotheses, so the final sample of 228 participants should have been sufficient.

However, the disappointing results indicated that power was lower than expected
due to lower effect sizes. With correlations ranging from r = .01 to r = .34 (N = 228), the
actual power ranged from less than .25 to .99. However, with the average correlation
being about .10, the average power for the study was less than .50.

Rewards: In order to make the goals more salient, the participants had an
opportunity to win money for good performance. The reward was designed to motivate
the participants to pay attention and make a sincere effort at doing the task. It was
mentioned at the beginning of the experiment, and a poster (unique to each goal
condition) reminding participants of the reward was posted on the wall. The details of the

reward are discussed in the goal manipulation section.

53
Change Memo: The change memo was designed to stimulate role innovation by

increasing the pressure on the participants. The change memo was presented between the
first and second sessions of the experiment. At the start of the second session,
participants received a memo (see Appendix E) from the company headquarters. The
memo demanded more production in the same amount of time. The new demands were
designed to encourage role innovation as participants sought better procedures for
meeting the new production requirements.
Egg

The task was a laboratory-based manipulation that used Tinker Toy model
building to approximate a two person work team (dyad) in a manufacturing situation.
The task was consistent across all conditions. Participants received a manual that
contained information about the organization, goals for the production session, standard
operating procedures for doing the task and pictures of the components to be built.
Participants within each dyad had different assignments (see Appendix C for the standard
operating procedures) but had the same amount of work. Worker #1 built Component #1
which was the base of the final unit. Worker #2 built Component #2 which was the upper
section of the final unit. Both workers contributed to the building of a smaller component
(Component #3). When all three components were completed, the two workers built the
final unit together by combining the three components. The task had independent
elements (Component #1 and Component #2) and interdependent elements (Component

#3 and final unit construction). However, even the independent elements required some

54
interaction. Each worker had a small task that he/ she was responsible for completing for

the partner.

Although the tasks for the workers were different, each worker had an equivalent
assignment determined by the number of Tinker Toy connections. One connection
occurred every time that one Tinker Toy was attached to another Tinker Toy. For
example, inserting a rod into a spool counted as one connection, and putting a spool on
each end of a rod counted as two connections. Each participant built a component that
consisted of the same number of connections. Although the components looked different,
each participant’s component required the same number of connections to complete. The
participants completed the same task in production session one and two.

Manipulations

Q]i_n_rate for Imogtion: Although the task remained the same, participants were
assigned to a condition presenting one of two company profiles. One company profile
presented a climate that supported innovation and one company profile presented a
climate that was very traditional. The manipulation occurred through descriptions of
different elements of the organization. The participants received a manual that described
the organization, and the experimenter also read the description out loud. The company
vision statement was also repeated between the two production sessions as part of the
change memo. The two conditions differed on the following elements: company name,
identification of the part being built, description of the type of employees that performed
well at the organization, description of the employee of the month, and the company

vision statement. The organization descriptions appear in Appendix A.

55
Goal Focus: The type of goal given to the participants was also manipulated, and

participants received either individual or group goals. The procedure manual informed
the participants about what components they must complete in order to perform well.
Participants in the individual goal condition had a goal to produce as many individual
components as possible, and participants in the group goal condition had a goal to
produce as many units of the final unit as possible. The goal for the first 20 minute
production session was presented in the manual (see Appendix B) and read out loud by
the experimenter. The goal for the second session was presented through the change
memo which was also read out loud. The goal for each condition was reinforced by goal
posters on the wall. After the first session, participants completed a Progress Report
which indicated how they were doing toward meeting the goal (See Appendix D).

As part of the design, a reward was possible for participants who performed well
at the task. The reward differed depending on the goal condition. The five dyads that
scored the highest in the group goal condition were entered in a random drawing for a
$40 reward that was split equally between the two participants ($20 each). The five
participants who scored the highest (regardless of the partner’s performance) in the
individual goal condition were entered in a random drawing for a $20 prize. The

drawings and distribution of the prizes occurred after the data collection was completed.

56
Procedure

Two participants signed up for each session and were randomly assigned to
conditions and roles within the dyad. Participants were given subject numbers as they
arrived which were put on the consent forms and answer sheets. Worker #1 had a 100
number (i.e. 101) and worker #2 had the corresponding 200 number (i.e. 201).
Participants were given the consent forms and the video camera was mentioned. The
experimenter collected the completed consent forms, explained the performance reward,
and read the Reward Details sheet out loud. The answer sheets were handed out and the
following scales were administered: mastery and performance learning orientation, and
individualism and collectivism.

The experimenter then handed out the manuals for that condition and read the
following: “This is a simulation of a manufacturing company. You will be playing the
part of employees working for a company named ‘insert name of company for
condition’ which builds ‘insert name of product for condition’. You will use the
Tinker Toys to build the product. The manual contains several important things. The
first thing it contains is information about the organization itself. Second, it contains
details about the manufacturing division. This section also explains what your goals are
for the production session. The manual also contains an overview of your job and the
standard operating procedures that explain how to do your job. The procedures are
supplemented by pictures that show how the components and the final product should

look. I am now going to read some of the information to you. Please follow along in

57
your manual.” The experimenter then read the company description, the goal sheet and

job details page for the condition.

The participants received three minutes to look through the manual and then they
began their first 20 minute production session after the experimenter left the room. After
20 minutes, the experimenter stopped the camera and administered the following scales:
self-efficacy, goal commitment, task frustration, and HMX. The participants filled out
the Progress Report forms concerning their performance. While the participants took a
break, the experimenter counted the assemblies completed and recorded the results on the
performance sheets and disassembled the parts for next production session.

The experimenter announced that the company’s main client had requested some
changes to the production schedule, passed out and read the change memo, and asked the
participants to complete the self-efficacy scale. The experimenter started the second
session, stopped it after 20 minutes, and administered the following scales: goal
commitment #2, task satisfaction, HMX #2, and job change. The experimenter handed
out debrief forms, answered questions, released the participants and then counted the
number of assemblies produced. The experiment took approximately one and a half
hours to complete.

Measures

Except for performance, role innovation and role negotiation, the measures were

self-report scales. All of the original scales and coding schemes are included in

Appendix G.

58
Individualism/ Collectivism (1C). Individualism and collectivism was preliminary

measured by the 20-item IC scale developed by Wagner (1995). The items were rated on
a seven-point Likert scale ranging from Strongly Disagree (1) to Strongly Agree (7).
Items were recoded so that higher ratings indicated stronger collectivism. The overall 20-
item scale had a coefficient alpha of .8019. However, Wagner (1995) did a factor
analysis in which these 20 items fell into five distinct factors. A factor analysis on this
data, reported in Table 2 found consistent results. A principle axis factor analysis using
Varimax rotation without a set number of factors found five factors with eigenvalues over
one. The five factors accounted for 54.8% of the variance. The first factor contained four
items and corresponded to the norms scale developed by Wagner and Moch (1986). The
second factor contained five items with four of the items from a scale by Erez and Barley
(1987) and one item by Triandis, Bontempo, Villareal, Asai and Lucca (1988). The third
factor contained three items with two of the items from the values scale by Wagner and
Moch (1986) and one item from Erez & Earley (1987). The fourth factor contained five
items from Triandis et al. (1988). The fifth factor contained three items that
corresponded to the Wagner and Moch (1986) beliefs scale.

Since the 20-item scale consisted of five distinct factors, the decision was made to
use only one of the factors. The factor was chosen based on conceptual relevance as well
as the robustness of the scale. The five factors were analyzed for content and reliability.

The third factor (Wagner and Moch, 1986; Erez & Barley, 1987) was chosen to be used in
the analyses. The three item scale asked about the individual’s attitude concerning

working in groups which seemed appropriate due to the individual versus goal

59

Table 1: Rotated Factor Matrix for Individualism/ Collectivism Scale

lR Only those who depend on themselves get ahead in
life.
2 R To be superior, a person must stand alone.

3 R If you want something done right, you’ve got to
do it yourself.
4 R What happens to me is my own doing.

5 R In the long run, the only person you can count on
is yourself.
6 R Winning is everything.

7 R I feel that winning is important in both work and
games.
8 R Success is the most important thing in life.

9 R It annoys me when other people perform better
than 1 do.

10 R Doing your best isn’t enough, it is important to
win.

I I I prefer to work with others in a group rather than
working alone.

12 R Given the choice, I would rather do a job where 1
can work alone rather than doing a job where l have to
work with others in a group.

13 Working with a group is better than working
alone.

14 People should be made aware that if they are
going to be made part of a group then they are
sometimes going to have to do things they don’t want
to do.

15 Only those who depend on themselves get ahead
in life.

16 People in a group should realize that they
sometimes are going to have to make sacrifices for the
sake of the group as a whole.

17 People in a group should be willing to make
sacrifices for the sake of the group’s well-being.

18 R A group is more productive when it’s members
do what they want to do rather what the group wants
them to do.

19 R A group is most efficient when it’s members do
what they want to do rather what the group wants
them to do.

20 R A group is more productive when it’s members
follow their own interests and concerns.

R . . .
indicates recoded items

.05
.03

-.04
-.05

-.04
.16

-.04
-.06

-.02

.07

.05

-.05

.10

.25

.07

.12

N

I"I"
U! 0

fl

I):
a.

.33

-.05
.23
.14
.19

.13
.14
.21
.04

-.02

-.08

-.06

.02

.14

Factors
3 4
.18 .07
.06 .14
.16 .14
.03 .l 1
.09 .09
.13 ._15
-.02 ,z__
-.08 ,g_
.29 fl
.12 ._6_6_
& .08
.8_1 .12
,§_8_ .06
-.05 -.06
-.01 .06
.09 .10
.10 -.01
-.03 .05
.00 .15
.06 -.04

.04
.08

.14
.00

.07
.00

.00

.10

-.O3

.09

.00

.07

-.Ol

.03

.07

60
manipulation in the context of the interdependent task. The chosen individualism!

collectivism scale consisted of three items (items #11, 12, 13) and had a coefficient alpha
of .8994.

Mastery and Performance Learning Orientation. Mastery and performance
learning orientation was measured by the 16-item scale developed by Button et al. (1995).
The items were rated on a five-point Likert scale ranging from Strongly Disagree (1) to
Strongly Agree (5). The mastery sub-scale consisted of the first eight items and had a
coefficient alpha of .7708. The performance sub-scale consisted of the last eight items
and had a coefficient alpha of .8329. The mastery and performance sub-scales were not
significantly correlated (r = .0946; p > .05). A factor analysis was computed on these
data (see Table 2) in order to confirm the independence of the mastery and performance
sub-scales. A principle axis factor analysis using Varimax rotation with two factors
specified found that the items loaded on the predicted factors. The two factors accounted
for 44.8% of the variance. The first factor corresponded to the mastery scale, and the

second factor corresponded to the performance scale.

61
Table 2: Rotated Factor Matrix for Mastery & Performance Learning Orientation Scale

Factors

1 2
M l. I do my best when I’m working on a fairly difficult task. .03 ._2_Z
M2. When I have difficulty solving a problem, I enjoy trying
different approaches to see which one will work. -.10 ._31
M3. I try hard to improve on my past performance. .14 _._5_8_
M4. The Opportunity to do challenging work is important to me. .00 ._ZQ
M5. The Opportunity to extend the range Of my abilities is -.03 13
important to me.
M6. The Opportunity to learn new things is important to me. .03 ,_6_4_
M7. I prefer to work on tasks that force me to learn new things. .04 fl
M8. When I fail to complete a difficult task, I plan to try hard the
next time I work on it. .23 ,_5_Q
P1. The things I enjoy the most are the things I do the best. 66 .07
P2. I feel smart when I do something without making any
mistakes. £2 .04
P3. I prefer to do things that I can do well rather than things I do
poorly. ._7_2 -.03
P4. I like to be fairly confident that I can successfully perform a
task before I attempt it. 62 -.02
P5. I am happiest at work when I perform tasks on which I know
that I won’t make any errors. J6 -.05
P6. I feel smart when I can do something better than most people. ._5_9 .00
P7. The Opinions others have about how well I do certain things
are important to me. éé .13
P8. I like to work on tasks that I have done well in the past. d .08

62
Climate for Innovation. Climate was coded as 1 = traditional organization and 2 =

innovative organization. This coding was chosen so that if the hypotheses were
supported there would be a positive correlation between the innovative climate and role
innovation.

Goal Focus. Goal focus was coded as l = individual goal and 2 = group goal.
This coding was chosen so that if the hypotheses were supported there would be a
positive correlation between the group goal and role innovation.

Horizontal Member ExchangeIHMX). Horizontal member exchange was
measured using the HMX scale developed by Nason (1995). It measured the quality of
the relationship between the partners. It contained statements such as: “My partner and I
would probably get along well as teammates.” The items were rated on a seven-point
Likert scale ranging from Strongly Disagree (1) to Strongly Agree (7). The 14-item scale
had a coefficient alpha Of .9610.

Role Innovation. Both Of production sessions were videotaped and independent
coders watched the tapes and coded instances Of role innovation behavior. Development
of the role innovation measure involved defining role innovative behaviors, developing a
coding scheme and modifying the final measure in order to get a robust scale.

The six role innovation categories identified by West (1987a) were used to
develop the expected role innovation behaviors for this task. West (1987a) perceived role
innovation essentially as ‘making a change to a task’ and divided it into six categories Of
behavior. The six role innovative categories identified were: (1) Setting new work

targets/ Objectives; (2) Deciding the methods used to achieve work targets/ Objectives; (3)

63
Deciding the order in which different parts of the job are done; (4) Choosing whom you

work with in order to carry out your work duties; (5) Initiating new procedures or
information systems; (6) Developing innovative ways Of accomplishing targets/
Objectives (West, 1987a). These categories were used to develop a list Of expected role
innovations for the task. However, the fourth category, ‘Choosing whom you work with
in order to carry out your work duties,’ was dropped, because it was not applicable to this
laboratory study. It was replaced with a general social category called “Redefining
relationships with others.”

One of the challenges of this study was Operationalizing role innovation into
concrete behaviors that could be objectively coded. First, the role innovations were
separated into the six role innovation categories from West (1987a) with the modified
social category. Next, the role innovation categories were broken down into specific
task-related behaviors expected in each category. Third, the behaviors were described in
action-oriented terms which were the indicators of whether or not the role innovation had
occurred. Last, the behaviors were listed on a coding sheet. The coding sheet containing
the 15 expected role innovation behaviors is included in Appendix H.

Using the coding sheets, the coders determined whether or not the worker
engaged in each Of the 15 expected role behaviors. Participants received a mark for each
role innovation that they performed in a production session. Since the first production
session was a learning and baseline Opportunity, the role innovation score used in the

analyses was only the score from the second production session.

64
After the videos were coded, the value Of the scale was tested by examining the

internal consistency and inter-rater reliability. An initial internal consistency analysis
(coefficient alpha = .4681) indicated that the 15 behaviors did not hold together well for
measuring role innovation. Some of the 15 behaviors had little or no variance due to low
activity/ coding rates. Through a series of internal consistency analyses, the role
innovation variable was modified to be a five item measure that consisted of socially
oriented behaviors (coefficient alpha = .6682). The five items were behaviors #3, 4, 5, 10
and 14 from the coding sheet (Appendix G).

After identifying a reliable scale, the consistency of the coding for that scale was
considered. Most of the sessions were only coded by one rater, however, a select few
(n=12) of the sessions were coded by two coders in order to establish inter-rater
reliability. Three analyses were done to measure how consistently the videos were coded.
First, an intercorrelation Of r = .7542 (p<.01) for the role innovation scores for the two
coders showed reasonable similarity. Second, an internal consistency alpha analysis with
n=12 and treating the two raters as items found a strong coefficient alpha of .8585. The
first two analyses were done on summed scores of the number of the role innovation
behaviors done. The third analysis considered the rater agreement for each Of the 5 items.
Specifically, it measured whether or not the raters marked the same behaviors, not just
whether or not the overall sums were similar. The analysis was simply a percentage of
the number Of items that were rated the same way, and the two raters agreed on 82% of
the behaviors. In order to account for within-rater variance, the role innovation scores

were standardized within raters.

65
Some convergent validity was found by comparing the coded role innovation

measure with a self-report measure collected in the study. The self-report measure of
“Job Change” was developed for this study and gathered at the end. The coded role
innovation measure and the self-report job change measure had a moderate and
significant correlation (r = .28, p < .05). In summary, role innovation was a five-item,
standardized measure based on the coded behaviors from the second production session.

Role Negotiation. Role negotiation was preliminarily measured through the 15
behavior coding scheme used to measure role innovation. The shared coding scheme led
to an artificially inflated relationship between the two variables (r = .57; p < .05). Role
negotiation was defined as occurring when a participant discussed a potential role
innovation with the partner. If the two workers discussed a role innovation, then the dyad
received one point for role negotiation. The participants in each dyad received the same
role negotiation score. This variable was a dyadic-level variable that was disaggregated
to the individual level for these analyses. Unlike the role innovation score, the role
negotiation score was a sum of the score for the first production session plus that for the
second production session. This was done to assure that the complete negotiation pattern
was captured. If a negotiation was successful in the first session, then it might not have
been repeated in the second session. Even so, all of the negotiations could have impacted
behavior in the second session.

An initial intemal consistency analysis (coefficient alpha = .4864) indicated that
the 15 behaviors did not hold together well for measuring role negotiation. Some of the

15 behaviors had little or no variance due to low activity/ coding rates. Through a series

66
of internal consistency analyses, the role negotiation variable was modified to be a four

item measure that consisted Of socially oriented behaviors (coefficient alpha = .6041 ).
These four items are behaviors # 3, 4, 5, and 14 from the coding sheet (Appendix G).

As with role innovation, the same 12 sessions existed for computing inter-rater
reliability and similarity. Three analyses were done to measure how consistently the
videos were coded. First, an intercorrelation of .7239 (p < .05) for the role negotiation
scores of the two coders showed reasonable similarity. Second, an internal consistency
analysis with n = 12 and treating the two raters as items found a strong coefficient alpha
of .89. The first two analyses were done on summed scores of the number of the role
negotiation behaviors done. The third analysis considered the rater similarity for each
item. Specifically, it measured whether or not the raters saw the same behaviors, not just
whether or not the overall sums were similar. The analysis was simply a percentage of
the number of items that were rated the same way, and the two raters agreed on 96% Of
the behaviors. In order to account for within-rater variance, the role negotiation scores
were standardized within raters. In summary, role negotiation was a 4-item, standardized
measure based on the coded behaviors from the first and second production sessions.

Performance. Task performance was measured by the number of final units
completed in the second production session. Final units were chosen as the performance
measure, because they were the primary concern of the “organizations” in the study.
Each participant in the dyad received the same score. Similar tO role negotiation,
performance is a dyadic-level variable that was disaggregated to the individual level for

these analyses. The performance measure was collected by the experimenter at the end of

67
each production session by counting the number of final units, components and

unfinished parts completed before breaking apart the components. The range of the
performance score was zero to seven final units.

Task Satisfaction. Task satisfaction was measured by a nine-item scale created
for this study. It included items such as “I feel fairly satisfied with the kind of work I
did.” The items were rated on a seven-point Likert scale ranging from Strongly Disagree
(1) to Strongly Agree (7). The scale had a coefficient alpha of .8601.

Job Change. Job change was a self-report measure Of role innovation that was
developed for this study. The scale was used as a check for the role innovation coding
scheme and the two were significantly correlated (r = .28, p < .05). The 10 item scale
included statements such as “I changed the order of my procedure”. The items were rated
on a seven-point Likert scale ranging from Strongly Disagree (1) to Strongly Agree (7).

The scale had a coefficient alpha of .8752.

RESULTS

Data Analysis

The psychometric properties of the scales and constructs were assessed before
testing the conceptual model. The means, standard deviations, reliabilities and
intercorrelations for the scales are presented in Table 1. The hypotheses in this study
were primarily tested using hierarchical regression analysis (Cohen & Cohen, 1983).
Variables were entered into the analyses according to their order in the model (see Figure
1). The individual difference variables (individualism/ collectivism (IC), mastery and
performance learning orientation) were entered into the equation in the first step, since
they are believed to be stable characteristics. The manipulations (climate for innovation
and goal focus) were entered in the second step, and the other variables were entered in
the order of the model. The following analyses show the effects of the different
predictors and manipulations on the different dependent variables. An alpha level of .05
was used in the text description for all statistical tests. However, a few marginally
significant results are discussed in greater detail with an alpha level of .10. The
regression tables reflect three alpha levels (.01***, .05", .10") in order to give a more
complete picture of the results of this study. The regression results did not support most
of the hypotheses, so additional analyses were done in order tO examine the relationships
from alternative perspectives. Some of the hypotheses were re-examined by considering
the change of the variable over time by using repeated measures multivariate analysis of
variance with covariates (RM-MANCOVA). Table 10 appears at the end of the Results

section and provides a summary of the hypotheses and their outcomes.

68

69

._mcowamp 5 8258—2 EBB-tog 8.32
.83.? voafiapqfim 08 cosmcowo: 29. EB cougoafi 2cm :o. v a u .1. .3. v a u ..

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

@- 8; 8.. 8.. 8. a. a. 8.. 8. 8. am
men an... a... 88. an...“ a. a. 3.... 88 8... :32
883 82.8 53 88.8 888 8.8 53 88.8 88.8 588888
3. :8. 8.- 8. :3: .2. 8. 8. :8. :3. sea-.2
83 82.8 88.8 88.8 83 8.3 8:8 888
2 8.- 8. .:. 8. :.- 8. 8. 2. 0058.888

883 8.8 88.8 88.8 88.8 88.8 833 889,85
8. :8. 8.- 8.- .2. 8. 8.- 8. Box.”
53 883 883 8.3 883 8.3 53882
8. 8. 8.- 8. .3. 8. 8. 23:
88: A83 A83 883 83
ea. 09. .1 _. 15—. ac. aee_. x23 .0
83 8:8 82.8 8”:
a 8. 8.- 8. 8. 88580 .m
8...: 885 888 8:82.:—
s. 8. 8.- 8. E «38:04.
83 88.8 $352.8
8. 8. .2. 223858
823 =25:on
8. 8. 85.58.08 .N
:o_§=oto
2.. been: ._
2 8 w A o m e m m _

 

 

 

 

 

 

 

 

 

 

838:5, ESE no.“ macaw—080285 v5 mozzfiazom 8:28:60 Empeflm £582 ”m 2an

 

70
Horizontal Member Exchange (HMX). Hierarchical regression was used to

examine how the individual differences, climate for innovation and goal focus influenced
the quality of the relationship (HMX) between partners. The first step of the regression
entered IC and the two learning orientation variables. The second step entered the
climate for innovation condition and the goal focus condition. Table 4 presents the
results of the analysis.

Results indicated that the individual differences accounted for a significant
amount of the variance in HMX (R2 = .06, F = 4.82, p < .05). The beta weights show that
IC and mastery orientation account for most of the variance and performance orientation
is not significant. The beta weight indicated that IC was significantly and positively
related to HMX ([3 = .17 at step 1, p < .05). Since a high IC score indicated a more
collectivist outlook, this supported Hypothesis 10 which stated that participants with a
collectivist outlook would have higher quality relationships than participants with

individualist outlooks. Mastery learning orientation was also positively related to HMX

(B = .13 at step 1, p < .05). This provided support for Hypothesis 12 which stated that
mastery learning orientation would be positively related to HMX. Performance
orientation did not account for any variance in HMX (B at Step 1 = .07, p > .05). Though
the step was significant, only IC and mastery orientation had significant beta weights.
This indicates that performance orientation was not significantly related to HMX.

The second step Of the regression entered climate for innovation and goal focus.
Hypothesis 4 predicted that participants in organizations with climates for innovation

would have higher HMX scores than participants in traditional organizations. Hypothesis

71
5 stated that participants with group goals would have better HMX scores than

participants with individual goals. However, the second step of the regression did not
account for any additional variance in HMX (A R2 = .01, A F = 1.71 , p > .05). The
climate beta weight ([3 at Step 2 = .10, p > .05) and the goal focus beta weight (0 at Step 2
= -.07, p > .05) were both non-significant for the step. Thus, neither hypothesis four or
five was supported. Overall, the five variables in the regression accounted for seven

percent Of the variance in HMX (R2 = .07, F = 3.59, p < .05).

Table 4: Hierarchical Regression Analysis Results for HMX

 

 

Step Predictors [3 at Step Final [3 (If R1 (Adj. R’) A R2

1 Individ./Collectivism .l7* * * .16* "‘ "‘ 3 .O6* * * .06“ * *
Mastery Orientation .13 * * . 14* * (.05*"‘*)
Performance Orientation .07 .07

2 Climate for Innovation .10 .10 5 .07* * * .01
Goal Focus -.07 -.07 (.05** *)

*p S .10, ** p S .05; ***p S .01;N = 228

Role Negotiation. Hierarchical regression was used to examine how the
individual differences, climate for innovation, goal focus and HMX influenced role
negotiation between partners. The first step of the regression entered IC and the two
learning orientation variables. The second step entered the climate for innovation
condition and the goal focus condition, and the third step entered HMX. Table 5 presents

the results of the analysis.

72
The proposed model did not explain any of the variance in role negotiation.

Results indicated that individualism/ collectivism was the only variable that accounted for
a significant amount of the variance in role negotiation, and the step was not significant
(R2 = .02, F = 1.63, p > .05). The beta weight indicated that IC was positively related to
role negotiation (I3 = .14 at step 1, p < .05). NO specific hypothesis were made
concerning this relationship. The other individual differences variables were not
significantly related to role negotiation. Neither mastery orientation (D at Step 1 = .01, p
> .05) nor performance orientation ([3 at Step 1 = .02, p > .05) had significant beta
weights for this step. NO specific hypotheses were made concerning these relationships.
Step two entered the climate and goal manipulations and did not account for any
additional variance in role negotiation (AR2 = .00, A F = .07, p > .05). This indicated that
climate for innovation ([3 at Step 2 = .01, p > .05) and goal focus (B at Step 2 = -.02, p >
.05) did not affect the amount of role negotiation. The third step entered HMX and
hypothesis six predicted that there would be a positive relationship between HMX scores
and the amount Of role negotiation. Step three of the regression showed no support for
Hypothesis six (A R2 = .00, A F = .02, p > .05) and the beta weight for HMX was non-
significant ([3 at Step 3 = -.01, p > .05). Overall, the variables in the regression accounted

for two percent of the variance in role negotiation (R2 = .02, F = .83, p > .05).

73
Table 5: Hierarchical Regression Analysis Results for Role Negotiation

 

 

Step Predictors p at Step Final [3 (If R1 (Adj. R’) A R2

1 lndividJCollectivism .14“ .14" 3 .02 .02
Mastery Orientation .01 .02 (.01)
Performance Orientation .02 .02

2 Climate for Innovation .01 .01 5 .02 .00
Goal Focus -.02 -.02 (.00)

3 HMX -.01 -.01 6 .02 .00

(.00)

”pg .10, ** p S .05; ***pS .01;N =228

Role Innovation. Hierarchical regression was used to examine how the individual
differences, climate for innovation, goal focus, HMX, and role negotiation influenced
role innovation. The first step of the regression entered individualism/ collectivism and
the two learning orientation variables. The second step entered the climate for innovation
condition and the goal focus condition, and the third step entered HMX. The fourth step
added role negotiation and the fifth step added the climate by goal interaction variable.
Table 6 presents the results of the analysis.

Results indicated that the individual difference variables did not account for much
of the variance in role innovation (R2 = .01, F = .82, p > .05). Individualism/ collectivism
( [i at Step 1 = .09, p > .05) was not significantly related to role innovation. This provided
no support for Hypothesis 11 which predicted a positive relationship between a
Collectivist outlook and role innovation. Neither mastery orientation (B at Step 1 = .00, p
> .05) or performance orientation ([5 at Step 1 = -.05, p > .05) had significant beta weights

for this step. This provided no support for Hypothesis 13 which predicted a positive

74
relationship between mastery learning orientation and role innovation. However, the null

relationship supported the expectation that there would be no significant relationship
between performance learning orientation and role innovation.

In the second step, the manipulations did not explain a significant amount of the
variance in role innovation (AR2 = .02, AF = 1.84, p > .05). Hypothesis I predicted that
participants in a climate for innovation would role innovate more than participants in a
traditional climate. The beta weight for climate for innovation provided marginal support
(p < .10 level) for Hypothesis 1 (B at Step 2 = .13, p > .05). The beta for climate in the
final step was stronger (B at Step 4 = .13, p < .05). In step two, goal focus (B at Step 2 = -
.01, p > .05) was not significantly related to role innovation. This provided no support
for Hypothesis 2 that predicted a positive relationship between group goal focus and role
innovation. Step three of the regression showed no relationship between HMX and role
innovation (AR2 = .01, AF = 1.52, p > .05). In the fourth step, role negotiation accounted
for significant variance in role innovation (AR2 = .32, AF = 108.82, p < .05). This

significant positive relationship (B = .57 at step 4, p < .05) supported Hypothesis 7 which
predicted a positive relationship between role innovation and role negotiation. However,
this relationship may be artificially inflated due to measurement bias, since the two
variables were collected using the same tape, coders, and a similar coding scheme. The
fifth step entered a moderated relationship between climate and goal focus which did not
account for any Of the variance in role innovation (AR2 = .00, AF = 1.66, p > .05). This
provided no support for Hypothesis 3 which predicted that climate and goal focus would

interact. Overall, the variables in the re ression accounted for 36% of the variance in role
8

75
innovation (R2 = .36, F = 15.29, p < .05) with most of that variance coming from role

negotiation.

Table 6: Hierarchical Regression Analysis Results for Role Innovation

 

 

Step Predictors p at Step Final [3 df R1 (Adj. R2) A R2
l Individ./Collectivism .09 .02 3 .01 .01
Mastery Orientation .00 .01 (.00)
Performance Orientation -.05 -.05
2 Climate for Innovation .13* .34” 5 .03 .02
Goal Focus -.01 .22 (.01)
3 HMX -.08 -.08 6 .03 .01
(.01)
4 Role Negotiation .57*** .57*** 7 .35*** .32***
L33***)
5 Interaction -.31 -.31 8 36*" .01
(Climate x Goal) (.33***)

*p S .10, ** p S .05; ***ps .01;N =228

Pefionlance. Hierarchical regression was used to examine how the individual
differences, climate for innovation, goal focus, HMX, role negotiation and role
innovation influenced performance. The first step of the regression entered
individualism/ collectivism and the two learning orientation variables. The second step
entered the climate for innovation condition and the goal focus condition, and the third
step entered HMX. The fourth step added role negotiation and the fifth step added role
innovation. The climate by goal interaction variable was entered last in step six. Table 7

presents the results of the analysis.

76
Results indicated that the individual differences in step one did not account for a

significant amount of the variance in performance (R2 = .01, F = .76, p > .05). None of
the three variables, individualism/ collectivism (B at Step 1 = .02, p > .05), mastery
orientation (B at Step 1 = .09, p > .05) or performance orientation (B at Step 1 = .01, p >
.05), had significant beta weights for this step. The second step entered the manipulation
variables which, as a whole, did not explain any of the variance in performance (AR2 =
.01, AF = 1.47, p > .05). Although goal focus showed no significant relationship (B at
Step 2 = .01, p > .05), climate for innovation did have a marginally significant (p < .10)
relationship with performance (B at Step 2 = -.11, p > .05). Step three Of the regression
showed that HMX accounted for a marginal amount (p < .10) Of the variance in
performance (AR2 = .01, AF = 1.34, p > .05; B at Step 3 = .11, p > .05). In the fourth
step, role negotiation did not account for additional variance in performance (AR2 = .00,
AF = 1.11, p > .05). The fifth step showed that role innovation also did not account for
any of the variance in performance (AR2 = .00, AF = .43, p > .05). This provided no
support for Hypothesis 8 which predicted a positive relationship between role innovation
and performance. The sixth step entered a moderated relationship between climate and
goal focus which did not account for any of the variance in performance (AR2 = .00, AF =
.004, p > .05). Overall, the variables in the regression accounted for four percent of the

variance in performance (R2 = .04, F = 1.06, p > .05).

77
Table 7: Hierarchical Regression Analysis Results for Performance

 

 

Step Predictors p at Step Final [3 df Rz (Adj. R’) A R2
l Individ./Collectivism .02 .01 3 .01 .01
Mastery Orientation .09 .08 (.00)
Performance Orientation .01 -.01
2 Climate for Innovation -.11* -.11 5 .02 .01
Goal Focus .01 .04 (.00)
3 HMX .11* .11 6 .04 .01*
(.01)
4 Role Negotiation .07 .10 7 .04 .00
. (.01)
5 Role Innovation -.05 -.05 8 .04 .00
(.01)
6 Interaction -.02 -.02 9 .04 .00
(Climate x Goal) (.00)

*p S .10, ** p _<_ .05; ***p S .01; N = 228

Task Satisfaction. Hierarchical regression was used to examine how the individual
differences, climate for innovation, goal focus, HMX, role negotiation and role
innovation influenced task satisfaction. The first step of the regression entered
individualism/ collectivism and the two learning orientation variables. The second step
entered the climate for innovation condition and the goal focus condition, and the third
step entered HMX. The fourth step added role negotiation and the fifth step added role
innovation. The climate by goal interaction variable was entered last in step six. Table 8
presents the results of the analysis.

As a whole, the individual differences accounted for a significant amount of the
variance (R2 = .11, F = 9.20, p < .05) in task satisfaction with most Of the variance

accounted for by the learning orientations. Results indicated that IC did not have a

78
significant relationship with task satisfaction (B at Step 1 == .04, p > .05). However, both

mastery orientation (B at Step 1 = .23, p < .05) and performance orientation (B at Step 1 =
.21, p < .05) had significant, positive relationships with task satisfaction. The
manipulation variables entered in step two did not account for any Of the variance in task
satisfaction (AR2 = .01, AF = 1.30, p > .05). Neither of the variables, climate (B at Step 2
= .04, p > .05) or goal focus (B at Step 2 = .09, p > .05), had significant relationships with
task satisfaction. Step three of the regression showed that HMX accounted for variance
in task satisfaction (AR2 = 08, AF = 23.13, p < .05) with a positive, significant
relationship (B at Step 3 = .30, p < .05). In the fourth step, role negotiation did not
account for additional variance in task satisfaction (AR2 = .00, AF = .09, p > .05). The
fifth step showed that role innovation also did not account for any of the variance in task
satisfaction (AR2 = .01, AF = 1.50, p > .05). This provided no support for Hypothesis 9
which predicted a positive relationship between role innovation and task satisfaction.

The sixth step entered a moderated relationship between climate and goal focus which did
not account for any of the variance in task satisfaction (AR2 = .00, AF = .02, p > .05).
Overall, the variables in the regression accounted for 21% Of the variance in task
satisfaction (R2 = .21, F = 6.41, p < .05) with most of the variance coming from the

learning orientations and HMX.

79
Table 8: Hierarchical Regression Analysis Results for Task Satisfaction

 

 

Step Predictors B at Step Final B df R‘ (Adj. R2) A R2
l Individ./Collectivism .04 -.01 3 .1 1* * "' .l 1 * “ "‘
Mastery Orientation .23*** .18*** (.10***)
Performance Orientation .21 "' “ * .18“ "' *
2 Climate for Innovation .04 .00 5 .12“ * "‘ .01
Goal Focus .09 .08 (.10* **)
3 HMX .30*** 29*" 6 .20*** .08***
(Isaac)
4 Role Negotiation .02 .07 7 .20* "' * .00
(Isaac)
5 Role Innovation -.09 -.09 8 .21* * * .01
(18***)
6 Interaction .04 .04 9 .21“ * * .00
(Climate x Goal) (.18***)

*pS .10, ** pg .05; ***p .<. .01;N =228

80
Additional analyses.

The regressions used to test the model hypotheses found few significant results.
The data did not support the model, and the variables had zero to minimal relationship
with each other. The model and hypothesis testing was based on scores from the second
production session. However, this analysis strategy did not Offer many useful results, so
additional analyses needed to be considered. Due to the unique design Of this study, there
was an additional opportunity to examine some of the dependent variables over time. It
was possible that some relationships between the variables could exist across time or
when variables interacted.

In order to consider some Of the dependent variables over time, a repeated
measures multivariate analysis of covariance (RM-MANCOVA) was computed. This
analysis examined the dependent variables as a group effect, over time, with the
covariates included. The RM-MANCOVA examined the following four dependent
variables that were measured after session one and after session two: performance, role
innovation, HMX and role negotiation. Climate for innovation and goal focus were
entered as the predictor variables. Mastery and performance learning orientation and
individualism/ collectivism were entered as covariates. The results of the RM-

MANCOVA are presented in Table 9.

Table 9: RM-MANCOVA Results

Source
Covariates'
Performance
Role Innovation
HMX
Role negotiation

Climate x Goal
Performance
Role Innovation

HMX
Role negotiation

Goal
Performance
Role Innovation
HMX

Role negotiation

Climate
Performance
Role Innovation
HMX

Role negotiation

Climate x Goal it Time

Performance
Role Innovation
HMX

Role negotiation

Goal it Time
Performance
Role Innovation
HMX

Role negotiation

3, 221
3, 221
3, 221
3, 221

1,221
1,221
1,221
1,221

1,221
1,221
1,221
1,221

1,221
1,221
1,221
1,221

1,221
1,221
1,221
1,221

1, 224
1, 224
1, 224
1, 224

F.

.59
.88
5.34***
1.36

.002
.44
.81
.95

1.13
.02
.56
.11

3.90"
4.39”
2.76*
.03

.20
.06
1.18
1.59

5.87"
.04
1.75
3.12*

82

Table 9: RM-MANCOVA Results continued...

Source g1_f E
Climate x Time
Performance 1, 224 .93
Role Innovation 1, 224 .20
HMX 1, 224 .002
Role negotiation 1, 224 .001
Time
Performance 1, 224 737.23***
Role Innovation 1, 224 .41
HMX 1, 224 6835*”
Role negotiation 1, 224 .00

aCovariates were mastery and performance learning orientation and individualism/
collectivism.
* pg .10, ** p S .05; *** p_<_ .01

83
The results of the RM-MANCOVA supported the few results found in the original

regression model testing such as the impact of the covariates on HMX (F = 5.34, p < .05).
Climate impacted role innovation (F = 4.39, p < .05) which confirmed hypothesis one.
Participants in the innovative organization role innovated more than participants in the
traditional organization (see Figure 4). This analysis provided more complete evidence

for the marginal relationship found using the regression analysis.

 

Climate Effect on Role Innovation

 

 

 

 

 

 

 

2
3 0.5
m r————
g ‘ .
’5 0 ' Trad ' '
,+ . ; .
i g l, . {
E L+hnov.j ,1
3 ~05 ‘
o
m
-1
Time 1 Time 2

 

Figure 4: Climate Effect on Role Innovation

84
Furthermore, the impact of climate on performance (F = 3.0, p < .05) was shown

as seen in Figure 5. Participants in the traditional organization performed better than
participants in the innovative organization. Although no direct hypothesis was made
concerning climate and performance, it was expected that participants in the innovative
organization would perform better after role innovating more. These relationships were

not supported. Figure 5 also demonstrated that performance increased over time.

 

Climate Effects on Performance

 

‘:._Trad. ; :
l+lnnov. .

Performance Score

 

 

 

 

Trme1 Turne2

 

Figure 5: Climate Effect on Performance

85
The RM-MANCOVA also found a marginal relationship (p < .10) between

climate and HMX (F = 2.76, p < .10). Participants in a climate for innovation had higher
HMX scores than participants in a traditional climate as seen in Figure 6. This provided
marginal support for hypothesis four which was not found with the regression analyses.

Figure 6 also demonstrated that HMX increased over time.

 

Climate Effects on HMX

 

 

 

 

 

 

7
6 .. $
.:

2 5 .1. E
g 4 1+Trad. ;
x " l+lnnov., i
g 3 e. ~——————~

2 -.

1

Time1 T|me2

 

Figure 6: Climate Effect on HMX

86
There were a few additional time-related relationships, such as performance being

affected by a goal by time interaction. As shown in Figure 7, performance differed for
the two goal conditions between the two sessions. In the first session, the participants in
the individual goal condition performed better than the participants in the group goal
condition. However, in the second session, the group goal participants improved. This
relationship was not specifically hypothesized in the study, but it provided valuable
information by indicating that the goal conditions did have different effects on

performance.

 

Goal Effects on Performance

 

 

‘_._ hd. Goal :
1+ Group Goal 3 l

 

denh-UICDV

 

Performance Score

 

 

O

 

Time 1 Time2

 

Figure 7: Goal Effect on Performance

87
Role negotiation also showed a marginal (p < .10) effect for the goal by time

interaction as seen in Figure 8. Participants in the group goal condition did a smaller
amount of role negotiation at time one but a larger amount at time two than the individual
goal participants. This indicated that the group goal participants probably learned the
value of discussing the task as they worked together. This relationship was also not
specifically hypothesized, but it also shows that the goal condition exerts subtle effects on

some variables over time.

 

Goal Effect on Role Negotiation

 

0.8 a
0.6 .:
0.4 ,

0.2 l i . ‘

o- —% .+hd.Goal Q
'02 3 1 1+Group Goall ‘
-o.4 -. *
-0.6 w

‘0.8 +-
-1

 

fl

 

Role Negotiation Score

 

 

 

Time 1 Time 2

 

Figure 8: Goal Effect on Role Negotiation

The strongest results came from the time effect itself. Performance showed a
dramatic time effect (F = 737.23, p < .05) as it increased from time one to time two.
However, this is a predictable result. The first session was essentially a learning trial and

performance was expected to be low. By the second session, participants were expected

88
to have learned the task and be able to perform better. Another result is that HMX

changed over time (F = 68.35, p < .05). This result probably stems from the fact that
people rated each other higher as they got to know one another during the sessions.

The RM-MANCOVA supported the original findings of the model testing, pulled
out some more subtle relationships and demonstrated some predictable but

unhypothesized findings (i.e. performance improved over time).

89

Table 10 Summary of Hypothesis Support

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Hyp Variables in Analysis Expected Direction of Supported?
Results

1 IV: Climate (innov./ trad.) More R1 in innovative org Yes
DV: Role Innovation (RI)

2 IV: Goal (group vs. ind) More R1 in group goal NO
DV: Role Innovation (RI) condition

3 IV: Climate x Goal (Goal x Org) variable will be NO

Interaction sig.: Innov org = more RI
DV: Role Innovation (RI) with group goal. Trad. org =
more RI with Ind. goal.

4 IV: Climate Higher HMX in innovative NO
DV: HMX Score org. condition

5 IV: Goal (group v. ind.) Higher HMX in group goal NO
DV: HMX Score condition

6 IV: HMX Positive rel. between HMX NO
DV: Role Negotiation (RN) and RN

7 IV: Role Negotiation (RN) More R1 for participants who YES
DV: Role Innovation (RI) do more role negotiation

8 IV: Role Innovation (RI) Performance is better for NO
DV: Performance (PF) participants who RI

9 1V: Role Innovation (RI) Task satisfaction is higher for NO
DV: Task Satisfaction participants who R1

10 IV: Collectivism Participants high in YES
DV: HMX collectivism will have higher

HMX scores

11 IV: Collectivism Participants high in NO
DV: Role Innovation (RI) collectivism will do more R1

12 IV: Mastery orientation Participants with high mastery YES
DV: HMX orient. will have higher HMX

13 IV: Mastery Orientation Participants with high mastery NO

 

DV: Role Innovation

 

orient. will have higher HMX

 

 

 

DISCUSSION

This study was designed to investigate the predictors and outcomes of role
innovation. The model presented two organizational variables, climate for innovation
and goal focus, that were expected to directly and indirectly affect role innovation. In
order to better understand the construct of role innovation, the process leading to it was
also investigated. This process included horizontal member exchange (HMX)
relationships which were expected to affect role negotiation which would then affect role
innovation. The model concluded by considering performance and task satisfaction as
two possible outcomes of role innovation. The model also included three individual
difference variables, mastery and performance learning orientation and individualism/
collectivism (IC). This discussion section will first consider the two primary research
questions that were centered around role innovation. Next, the relationships with final
outcomes and individual differences will be considered. Finally, the discussion section
will conclude with the limitations, contributions and future directions for research.
Prim Research uestions

The model was designed to answer two primary research questions. The first
question was: Will the organizational factors of climate for innovation and goal focus
affect role innovation? It was expected that factors in the organizational environment
would encourage or constrain an individual’s tendency to role innovate. Primary and
secondary messages sent by organizational policies, procedures and the working
environment should cue people as to how role innovations would be received.
Participants were expected to use this information when deciding whether or not to role

innovate.
90

91
Climate for innovation was one organizational factor that was expected to impact

role innovation, since people get messages about acceptable behavior from the climate.
In the initial regression analysis, only a marginal relationship was found between climate
for innovation and role innovation. Specifically, hypothesis one which predicted a
positive relationship between climate for innovation and the amount of role innovation
was only marginally supported at the p < .10 level. This suggested that climate for
innovation may be related to role innovation, but the regression analysis could not
confirm the relationship. However, the RM-MANCOVA analysis did find that climate
for innovation was related to role innovation as predicted in hypothesis one (F (1, 221) =
4.39, p < .05). Participants in the innovative organization did role innovate more than
participants in the traditional organization. In the innovative organization, participants
heard stories about a successful employee who innovated and an organizational
description that promoted innovation. As intended, these messages must have indicated
to the participants that innovation was encouraged which allowed the behavior to occur.
The relationship between climate and role innovation was hard to detect, since the RM-
MANCOVA found it but the regression was marginal. This discrepancy probably
occurred because the RM-MANCOVA had more power and data than the regression
analysis.

In addition to climate, goal focus was an organizational factor that was expected
to influence role innovation. Goal focus distinguished between giving individual or
group goals for the interdependent task. It was predicted that participants with group

goals would role innovate more. They had to focus on the group task as well as the

92
individual task which gave them opportunities to role innovate with their own work, the

group work or the interactive roles. It was expected that people with individual goals
would focus more on their own work and not do as many of the group-related role
innovations. None of the analysis results supported these predictions, so no support was
found for hypothesis two that predicted a positive relationship between group goal focus
and role innovation. It is possible that participants concentrated on the obviously
interdependent nature of the task and did not give any more thought to the goal
manipulation. The simplicity of the task may have made all possible role innovations
apparent to both conditions, not just the group goal condition. Since this study was
concerned about goal focus, “do your best” goals were used instead of challenging,
specific goals as recommended by many researchers (Latham & Locke, 1991). This may
also partially explain why no effect was found for the goal manipulation - the
manipulation itself might not have been effective. A better strategy may have been to use
specific and difficult goals with an individual or group focus, because they may have
provided more incentive to pay attention to the goal manipulation. Hypothesis three
predicted an interaction effect of climate and goal focus based on the expectation that
group goals would cause people to conform to the norms established by the climate.
Therefore, participants with group goals in innovative climates were expected to role
innovate more than participants with individual goals. Participants with group goals in
traditional climates were expected to role innovate less than participants with individual
goals. However, the weak effect of the climate and goal manipulations meant that no

interaction was possible and no support was found for hypothesis three.

93
The climate and goal manipulations were also hypothesized to relate to HMX.

Hypothesis four predicted a positive relationship for climate for innovation and HMX
which was marginally supported by the RM-MANCOVA analysis. Although marginal,
the relationship went the expected direction, since participants in the innovative climate
had higher HMX scores. Past research indicated that climate provides workers with clues
concerning appropriate interpersonal and task related behavior (Scott & Bruce, 1994; Farr
& Ford, 1990). The marginal results (and the hard to detect relationship between climate
and role innovation) may be due to the difficulty of manipulating an organizational
climate variable in a lab setting. The participants might not have been completely tuned
in to the organizational climate presented in the manipulation. Instead they might have
been attuned to the climate created by being in an experiment. This experiment climate
might have induced them to strictly follow the directions instead of innovating on the
task. The climate in real organizations is likely to have a bigger impact on HMX
relationships and role innovation than found in this lab study. In real organizations,
climate has been shown to influence intentions and behavior (Schneider & Bowen, 1985).
Hypothesis five which predicted a positive relationship between group goal focus
and HMX was not supported. There was no difference in relationship quality for
participants in the individual goal condition as compared to those in the group goal
condition. It is possible that the goal manipulation was not salient enough due to the
constraints of the “do your best goals.” Furthermore, HMX tended to have a ceiling
effect, because participants were hesitant to give their partners a low score. Participants

may have based the HMX score on non-task related issues such as how much they liked

94
the partner. The experimental setting and nature of the participants may have negatively

affected the HMX and related analyses. This study did contribute to the understanding of
organizational influences on role innovation by showing that climate for innovation did
have an impact, even if no evidence was found conceming the effect of goal focus on role
innovation. The lack of results for the goal manipulation may have been due to
weaknesses in the study design (see Limitations), and goal focus should not be quickly
dismissed as a potential influence.

The second research question considered in this study was: How do HMX and
role negotiation contribute to the role innovation process in interdependent workgroups?
By considering the process, it was hoped that the negotiation patterns and interaction of
the dyad could explain why some participants role innovated more than others. It seemed
reasonable that certain types of interaction patterns would lead to more or less role
innovation, since the task was interdependent. The model predicted that horizontal
member exchange (HMX) and role negotiation would precede role innovation and help to
predict it. However, this model as a whole was not supported and the main significant
relationship found was confounded by measurement problems.

HMX reflected the quality of the relationship within the dyad and was expected to
establish the interaction pattern between the partners. If the partners liked working
together, they were expected to negotiate more and eventually feel less constrained to role
innovate. In this study, HMX measured the quality of the relationship between the
partners, and role negotiation was supposed to provide a behavioral measure of the

outcome of that quality. Hypothesis six predicted a positive relationship between HMX

95
and the amount of role negotiation, however this was not supported. None of the

variance in role negotiation was explained by this model. The simplicity of this task
might have negated the need for role negotiation. Many times the role innovations were
done without discussing them first. If the partner had to change to accommodate the role
innovation, the change was often made quickly and without discussion. Most of the
conversation between partners centered around classes and tests and was completely
unrelated to the task. HMX was based on non-task interaction, so it is not surprising that
it did not relate to the task—oriented role negotiation.

Role negotiation was not related to HMX, however, a strong positive relationship
was found between role negotiation and role innovation which supported hypothesis
seven. This was expected since people who role negotiate with their partner are more
likely to feel comfortable about proceeding with a role innovation. Although this
hypothesis was supported, the relationship was probably artificially inflated by the
similar measurement of the two variables. This potential confound was discussed in
more detail in the limitations section.

Through the RM-MANCOVA, another un-hypothesized relationship was found
for role negotiation. A marginal effect (p > .10) was found for the goal by time
interaction on role negotiation. Participants in the group goal condition did a smaller
amount of role negotiation at time one but a larger amount at time two than the individual
goal participants. The group goal participants probably learned the value of discussing
the task as they worked together. This finding does indicate that there were some subtle

relationships concerning the goal focus manipulation. The results of the study provided

96
limited support for the two primary research questions. For question one, climate for

innovation was shown to be an organizational factor that affected role innovation.
However, no relationship was found between goal focus and role innovation. For
question two, the model did not contribute much to the understanding of role innovation.
Final Outcomes

In addition to manipulations, the model of role innovation included two final
outcome variables - performance and task satisfaction. Hypothesis 8 predicted that
performance would improve as more role innovation occurred. The task was designed
awkwardly in order to initiate corrective role innovations, therefore role innovation
should have allowed better performance. However, this hypothesis was not supported.
According to the regression analyses, none of the variables predicted performance (except
for marginal effects of climate and HMX). As a whole, the six variables predicted zero
percent of the variance in performance.

However, the RM-MANCOVA analysis showed an un-hypothesized relationship
between climate for innovation and performance. The results indicated that participants
in the traditional organization performed better than participants in the innovative
organization. Although this relationship was not explicitly hypothesized, the found
relationship was contrary to expectations. It was expected that participants in the
innovative climates would role innovate more, develop better procedures, build more
products and therefore perform better. However, the results indicated the opposite, and
the traditional climate eventually lead to better performance. Since climate was shown to

impact role innovation, participants in the innovative climates might have spent time role

97
innovating at the cost of immediate production time. The effect of the climate

manipulation and advantage of the role innovation might have been seen in later
production sessions after the role innovations were implemented and made routine.

Goal focus also had an un-hypothesized goal by time relationship with
performance that was found by the RM-MANCOVA. In the first session, the participants
in the individual goal condition performed better than the participants in the group goal
condition. However, in the second session, the group goal participants performed better.
This performance relationship more closely followed the expectations of the study. At
time one the individual goal participants performed better which may be due to higher
levels of personal responsibility and less social loafing. At time two, some of the
individual goal participants had figured out that they were being rewarded only for their
individual components and started focusing on the individual tasks. Since the
performance measure was based on the number of completed final units, participants in
the individual condition ultimately did worse on performance.

This study found some interesting relationships, but not the primary relationship
in the model between role innovation and performance. The simplicity of the task may
have allowed adequate performance regardless of role innovation. The session tapes
seemed to indicate that motivation predicted performance better than role innovation.
Participants who discussed the performance reward and agreed to try for it performed
very well regardless of innovative behavior. Role innovation might need to be studied

using a more complex task or re-conceptualized to include a motivational aspect.

98
Since the problem is usually easy to solve, role innovation might actually rely

more on motivation than on innovation. Some researchers (Staw & Boettger, 1990) even
believe that role “innovation” is somewhat of a misnomer, since the appropriate behavior
is rather obvious and does not always require much creativity. Farr & Ford (1990)
recognized that the innovative aspect of role innovation must be considered on a
continuous scale. Since role innovation solutions are often obvious, most individuals
should be capable of initiating them (Farr & Ford, 1990). Other researchers would add
that all creativity and innovation result more from hard work than from creative genius
(Staw, 1990). These researchers indicate that role innovation could be considered from a
motivational perspective instead of an innovation perspective. From a motivational
perspectives, variables such as job commitment, job involvement, ambition and
proactiveness may have an impact on role innovation. If a worker is committed to his or
her job, wants the job to improve and wants to make a good impression, he or she should
be more likely to invest the time and energy to role innovate when necessary. Less
motivated people should be content with the status quo and not seek change.

In addition to performance, hypothesis nine predicted a positive relationship
between role innovation and task satisfaction. Since the task was designed to be
awkward, it was expected that participants would be more satisfied after making some
changes. However, role innovation did not predict task satisfaction and the hypothesis
was not supported. Motivation and interest appeared to affect task satisfaction as well as
performance. Some of the participants considered working with Tinker Toys to be a

treat, while others considered the whole experiment as a chore. Compared to

99
performance, the model as a whole did a better job of predicting task satisfaction. The six

variables accounted for 18% of the variance in task satisfaction. This variance is
primarily accounted for by mastery orientation, performance orientation and HMX. None
of these relationships were hypothesized in the study.
Individual differences

Further analysis considered individual difference variables related to role
innovation. Both collectivism and mastery orientation were expected to predict role
innovation. Hypothesis 11 predicted that participants high on collectivism would role
innovate more, since the task was interdependent. They were expected to relate better to
their partner, negotiate more and feel more comfortable role innovating. However, no
support was found for hypothesis 11. This lack of results may have been due to
experimental constraints. It is possible that the behavior of the participants was
constrained by being in an experiment and not affected by general tendencies such as
collectivism. Furthermore, collectivism was measured by the participant’s preference for
working in groups which may have been negated by the forced group situation of the
experiment. Furthermore, hypothesis 13 predicted a positive relationship between
mastery orientation and role innovation, but this relationship was also not supported. It is
possible that the task did not offer enough challenge and interest to engage different
responses between participants who were high and low on mastery.

Some predictions were also made about other variables in the model such as
relationships with HMX which had several positive relationships with the individual

difference variables. Hypothesis 12 predicted a positive relationship between mastery

100
learning orientation and HMX which was supported. Participants with higher mastery

learning orientation were more likely to report higher quality relationships with their
partners. The higher mastery orientation probably led the participants to enjoy the
challenge of the new task and cooperate with the partner on the interdependent aspects of
the task. As expected, performance orientation was not related to HMX. Performance
oriented people probably sought to do well on the task, but doing well on an
interdependent task may or may not have required establishing a good relationship. If the
participant saw the relationship as an important part of performing well, then HMX
should have increased. However, people with high performance orientations might also
have seen the relationship as tangential to performance and not worried about it. This
dichotomy of approaches for performance oriented peOple probably canceled out any
significant relationship.

IC also related to HMX as predicted in hypothesis 10, since collectivism was
positively and significantly related to HMX. Participants who were high in collectivism
reported higher quality relationships with their partner. The collectivism measure asked
about the individual’s satisfaction with working in groups. It followed logically that
participants who enjoyed working in groups would develop better relationships with their
partner. The three individual difference variables explained five percent of the variance
in HMX.

The results of this study were disappointing. Few of the hypotheses were
supported and the primary research questions were not adequately answered. The poor

results may be due to the limitations of the study as discussed in the next section, or they

101
may reflect a conceptual or operational problem. Role innovation has always been a

challenging concept to study. By its very nature, innovative behavior is a hard concept to
predict or place boundaries around. Staw & Boettger (1990) found a surprisingly low
base rate of task revision in their research. They also found that conformity to control
mechanisms such as goal setting and norms restricted role innovation even more. The
manipulations in the study may unwittingly have decreased the already hard to detect
behavior. Future research will have to try to distinguish between study effects and actual
role innovation.
Limitations

The lack of results may have been due to some basic limitations of the research
that included the measures, coding, manipulations, participants and aggregation of the
variables. The most obvious problem centered on the measures used in the study. Role
innovation was the primary variable in the study and it was hard to operationalize, code
and transform into a robust scale. The role innovation measure was based on the task
design and refined after the pilot study, and it consisted of 15 behaviors that were
potential role innovations for the given task. Operationally, these behaviors placed
limitations on what could be considered as role innovation. This did not allow for role
innovations that were outside of the expectations of the experimenter. For instance, one
possible role innovation occurred if the participant chose not to fill out the required work
orders, since the work orders slowed down production time. However, several dyads did
not fill out the work orders during the production, but one partner would do all of the

paperwork when he or she had a lull. This was an innovative split of the workshare,

102
however, it did not count as a role innovation according to the original coding scheme.

Coding was also a difficult aspect of the role innovation variable. Although the behaviors
on the coding sheet were very specific, it was still sometimes questionable as to whether
the behavior had occurred or not. Furthermore, the participants were often inconsistent in
their behavior which opened questions concerning how a behavior should be marked.
Participants might do a role innovation once and then revert back to the original behavior,
and sometimes the role innovation appeared to be an accident.

It was also difficult to turn the coding scheme into a scale with strong
psychometric pr0perties. The original conceptualization gave each participant a point for
every one of the fifteen role innovations that was performed, and role innovation was
planned be a count of the total number of behaviors. Treating role innovation as a count
did not allow for any psychometric tests to determine whether or not the measure was
robust. The measure was then re-conceptualized as a scale, and tests were conducted to
check for internal consistency. These tests indicated that only five of the behaviors hung
together as a coherent scale, so these five behaviors became the final scale. This switch
from a count to a reduced scale indicated that the variable was not well-constructed at the
start. The extraneous ten behaviors just complicated the coding. The role negotiation
variable was developed and coded in a similar fashion as the role innovation variable,
which created an artificially high relationship between the two variables.

The best solution for the coding issues might be to use a more global coding
scheme. The identified behaviors could be used to guide a more general rating of a

question such as, “Worker #1 role innovated a lot, a little or none at all”. An appropriate

103
global measure could be developed that was less behaviorally oriented. In a global

measure, the participants would get more credit for intentional and repeated role
innovations. A behaviorally anchored rating scale (BARS) could be used to guide the
global ratings. Studies have shown that BARS can improve reliability (Maas, 1965) and
accuracy (Vance, Kuhnert, & Farr, 1978) over other types of ratings. A new rating
approach might help future research.

Along with the coding scheme, the coders had some problems in this study. They
did not receive enough training which created some uncertainty concerning the
consistency of the coding. One coder in particular had a complete misunderstanding
about the coding which was not discovered until most of the coding was completed,
therefore most of her coding was rejected. This event and a basic time constraint resulted
in very few tapes being double coded which made it difficult to assess inter-rater
reliability. Some of the differences between coders were minimized by standardizing the
scores within raters before combining them for the final analyses.

The climate and goal manipulations also had some problems. Climate is always
hard to manipulate in a laboratory session, but the pilot study did indicate that
participants in the two conditions perceived the organizations as being significantly
different. One analysis found a significant relationship for climate and role innovation
whereas the other analysis found a marginal relationship. Based on these findings, it
should be reasonable to assume that the relationship exists but is weakened by the

difficulty in studying climate in the laboratory. As mentioned earlier, participants might

104
have been responding to the climate of being in an experiment as well as the climate

presented in the study.

In the pilot study, the goal focus manipulation provided participants with difficult,
specific goals, but this had the side effect of stopping production if the participants
reached the goal before the session was over. In order to keep production going for the
whole session, the goal manipulation in the actual study was made generic (“do your
best’), which seemed to be adequate since the primary purpose was to make an individual
or group focus. However, the study found no results for the goal manipulation. The goal
manipulation might not have been salient enough and might not have created the pressure
created by trying to reach a specific, difficult goal. It is possible that this manipulation
would have worked better if the goals had been specific and difficult (Latham & Locke,
1992) as well as individual or group focused.

The participants used in the study might also have affected the results. They were
mostly first year student in their first semester of college and showed far more reluctance
about “breaking the rules” than the more advanced pilot study participants did. The
participants frequently talked about how they should follow all of the rules, since they
were in an experiment. In a real-world situation, workers would be comfortable in their
work environment and be better able to gauge what role innovations might be
appropriate.

Another problem was that several of the variables (performance and role
negotiation) were disaggregated from the dyad level to the individual level. In essence,

this created a multi-level model that consisted of individual and dyadic level variables.

105
The fact that these two variables were constrained by the dyad was not accounted for in

the study.
Contributions

Although many of the hypotheses were not supported, this study did make several
important contributions. The first contribution was that it did find a relationship between
climate for innovation and role innovation even with the manipulation difficulties. This
provides a basis for considering other issues such as how to foster role innovation in
employees if desired. Furthermore, the study helped develop the HMX construct and
demonstrated the relationship between HMX and other variables such as mastery
orientation and task satisfaction. Second, the study provided a strong, well-bounded
definition of role innovation. Past researchers had relied on poorly specified definitions
that often did not capture the multi-dimensional aspects of role innovation. The study
definition highlighted the task-focus and social elements of role innovation, was based on
the role, job and innovation literature, and provided the boundaries for the concept.

A third contribution was the attempt to study role innovation in a laboratory
setting, since most earlier research was either conceptual or survey-based. The laboratory
provided greater control over the manipulations. However, the disappointing results of
this study may indicate that it is not appr0priate to study role innovation in a contrived
setting. Fourth, this study tried to predict and identify role innovative behaviors within a
specific context by developing an appropriate coding scheme. This involved identifying
the specific behaviors that were role innovations. This was one of the first attempts to

break the behavior down into observable segments which could be quantified. This

106
coding scheme was necessary for the study, but might also have prohibited the notice of

other role innovations. A fifth contribution was that this study considered the process
involved in role innovation instead of simply defining it as an outcome. An attempt was
made to link role innovation to HMX and role negotiation as well as final outcomes. This
placed role innovation into a complete context and exposed some of the complexities of
looking at the whole process.

Perhaps the most important contribution of this study is that it suggested some
conceptual changes to the variables and their definitions. Although the role innovation
definition provided good structure for the concept, the poor results and subjective
analysis of the session tapes indicate that it may be lacking a motivational element. As
discussed earlier, role innovation may rely partially on the worker’s motivation to change
and improve the job. The definition of role innovation could be reconceptualized as a
multi-faceted construct that includes task-focus, a social element and a motivational
element. The motivational elements might be captured through measures of job
commitment, job involvement, proactiveness and more.

However, an alternative approach would be to accept the current definition of role
innovation and investigate motivation as a predictor or instigator of it. The definition
currently presents role innovation as the behavior of the worker when changing a task.
The behavioral definition could remain as is, and motivation could act as a predictor.
Motivation is hinted at in the current definition since it is a “proactive process”, but it

might have a stronger impact as a predictor. By considering motivation as a predictor and

107
not a basic part of the definition, it would allow future researchers to investigate many

different motivational predictors that might affect role innovation.

The study also suggested a conceptual change to the performance variable. In this
study, the definition of performance was constrained by the task design. The simulation
involved companies that were trying to produce and ship final products, therefore
performance was measured by the number of final products produced. This definition
was based on the mission statements of the simulation organizations. However,
performance can be a complicated, multi-dimensional construct. One issue concerned
what the performance variable was actually measuring. In this study, performance might
simply have been measuring working speed, speed of information acquisition (i.e. reading
speed for the manual), or manual dexterity. It is important to determine whether the
performance variable is measuring the intended skill, trait or behavior. The multi-
dimensional nature of performance also indicated that it could include other dimensions
such as a measure of quality. Due to the age and poor condition of the Tinker Toys, this
study was unable to measure quality of the product. Future research could consider more
complete measures of performance. This study presented a definition of role innovation
and stimulated some important conceptual ideas and directions for future research.

Future Research Directions

In future studies, researchers should try to improve on the limitations of this
study. For instance, new ways to operationalize role innovation must be developed.
Researchers must find an objective means for determining when a role innovation has

occurred by answering questions such as: Does role innovation occur when behavior

108
deviates from a standard, when an individual perceives that he or she has done something

new, or when co-workers perceive a behavior as different? Research on role innovation
will not proceed until questions such as this can be satisfactorily answered and built into a
research plan. As role innovation is being defined, researchers should consider the
motivational aspects of it that were discussed earlier. It may be useful in the fiiture to
take a motivational perspective to role innovation instead of trying to fit it within the
innovation research framework.

The limitations of this laboratory study indicate that role innovation should
probably be studied in natural field settings. Variables such as climate that are likely to
influence role innovation do not work well in a lab setting. Furthermore, when
considering the role innovation process, it might be more appropriate to consider it from a
motivational perspective. Since role innovations are usually small-scale changes, they
are often apparent to many people. Motivation may be the reason why some people
choose to make changes and others do not. Finally, if role innovation is studied between
dyads or teams, then the level of analysis should reflect that. The interdependent nature
of most tasks makes it difficult to identify individual contributions and initiative.

In conclusion, this study demonstrated a relationship between climate and role
innovation, but failed to support many of the other central hypotheses. It indicated that
the approach to studying role innovation needs to be re-conceptualized. The difficulties
in the study highlight the problems inherent in studying novel response behaviors such as
role innovation. Many of the non-significant relationships in this study can be attributed
to operational and design problems and should not automatically be used to claim that

role innovation is not affected by the variables in the model. Large amounts of anecdotal

109
evidence exists in support of role innovation. Most people who have worked in a factory

can cite examples of workers who have redesigned their processes or bypassed safety
measures (not recommended) in order to work faster or better. Early researchers have
related stories about places like Hawthorne where all sorts of unofficial, time-saving
devices and tools were found in the workbenches after the plant closed. These provide
anecdotal evidence that role innovation occurs in organizations. Employee-initiated
change may become critical in today’s quickly changing work environment, and role
innovation is the most basic type of change. It falls to the organizational researchers to

find a way to study and eventually encourage this behavior.

APPENDICES

APPENDIX A:
ORGANIZATION DESCRIPTIONS

110

New Directions, Inc.

Welcome to New Directions, Inc. We are proud to have you as part of the New
Directions team. We have prepared the following orientation materials in order to
introduce you to the company.

What products does New Directions, Inc. produce?

New Directions, Inc. is a middle-sized company that makes communication satellites.
Our satellites are helping to change the world. They allow people on opposite parts of the
globe to have instantaneous communication with phones and with new video-
conferencing. Other features include:

24-hour remote modem hook-ups regardless of location
Location positioning for ships on any ocean

Real-time heat sensor feedback analysis of military targets
High-tech weather watch forecasting.

At New Directions, Inc., we connect the world with the future.

 

111

How are these products produced?

As part of a fast-paced, competitive industry, we focus on creating newer and better
products, services and techniques for our business. We are a high-tech company that puts
as much attention into assembling a superior part as we do into designing it. We strive
every day to improve our product and our company. In order to beat our competition, we
are always striving to find new and better ways to do things.

Everyone in the organization shares this competitive spirit. Our employees want to make
New Directions the best in its field. Since our product is so important, we work very hard
to assure high quality. Quality of the product is everyone’s responsibility, from the
workers who assemble the satellites to the company President.

What type of employees do well at New Directions, Inc. ?

New Directions, Inc. recognizes the importance of having good employees in order to
achieve high quality. We only hire employees who are bright, creative and motivated.
Everyone has a contribution to make to their job and to the organization. At New
Directions, we expect our employees to make a contribution to the organization, not just
to show up and punch the clock. We hire employees who are willing to work in the New
Directions spirit.

 

New Directions, Inc. Employee of the Month

New Directions, Inc. is proud to announce the employee of the month - Lee Peterson.

Lee was selected as employee of the month due to contributions made in the Inventory
department. Lee was trained to use the original inventory tracking system. The old
system used paper-based tracking forms and handwritten log sheets to determine where to
send products. Lee began to realize that the inventory tracking could be done in a better
way, and developed a computerized tracking system for inventory. This system uses
electronic tags and the system can immediately find any product and tell the status of that
product. This new system has improved delivery time and accuracy by almost 55% over
the old system. Lee saw a need at New Directions and stepped forward to offer a
solution. Lee Peterson represents the spirit of New Directions, Inc.

112

Company Vision

We will strive to develop and produce superior, high-quality satellites that always exceed
the expectations of our customers. We will work to become the leader in our industry by

developing new products and envisioning new ways to use them. By hiring ambitious and
creative employees, we will ensure that workers in all divisions will strive to find new and
better ways to design, build and market our product. We strive to take our company, our

employees and our customers in new directions.

 

113

Johnson & Sons, Inc.

Welcome to Johnson & Sons, Inc. We are proud to have you as part of the Johnson &
Sons team. We have prepared the following orientation materials in order to introduce
you to the company.

What products does Johnson & Sons, Inc. produce?

Johnson & Sons, Inc. is a middle-sized company that makes telegraph machines. Our
telegraph machines are helping to preserve a traditional form of communication. They
allow people to communicate through Morse Code. Our telegraph machines allow people
in remote areas to communicate. Other uses for the telegraph machines include:

0 Military communication

0 Badge training for Boy Scouts.

At Johnson & Sons, Inc., we help to preserve the past.

 

114

How are these products produced?

As part of an established industry, we focus on producing our products the same way that
we always have. We are a traditional company that is devoted to assembling a product
that meets all of our standards and requirements. Since our product is so important, we
work very hard to maintain our traditional high quality. Quality of the product is
everyone’s responsibility, from the workers who assemble the telegraph parts to the
company president. We strive every day for consistency and reliability. Everyone at
Johnson and Sons shares this respect for tradition and helps maintain our ways of doing
business.

What type of employees do well at Johnson & Sons, Inc. ?

Johnson & Sons, Inc. recognizes the importance of having good employees in order to
achieve high quality. We only hire employees who are prompt, dependable and honest.
At Johnson and Sons, we expect our employees to do things the Johnson & Sons way.
We have employees who understand and appreciate our traditions.

 

Johnson & Sons, Inc. Employee of the Month

Johnson and Sons, Inc. is proud to announce the employee of the month - Lee Peterson.
Lee was selected as employee of the month due to contributions made in the Inventory
department. Lee was trained to use the inventory tracking system. The system uses
paper-based tracking forms and handwritten log sheets to determine where to send
products. Lee contributes to the department by typing up the log sheets so that they are
more readable. Lee has been a faithful employee at Johnson and Sons for 10 years. Lee
is always on time for work and never misses a day. Lee was also elected historian for the
company and keeps scrapbooks that show the history of the company. Lee is the type of
committed and dependable employee that this company relies upon. Lee Peterson
represents the spirit of Johnson & Sons, Inc.

115

Company Vision ,

We will strive to develop reliable telegraph machines and parts that always meet the
expectations of our customers. We will work to maintain our place in the industry by
producing products with the traditional Johnson & Sons quality for the customers that
we have served for years. By hiring dependable employees, we Will ensure that workers
in all divisions will continue the traditions of the company. We strive to maintain the
traditions and customer satisfaction that are the hallmark of our company. .

 

 

APPENDIX B:
GOAL MANIPULATION

116

Manufacturing Division

In the Manufacturing Division, our job is to construct the satellite. The station where you
will work is responsible for assembling the outside casing of the satellite. You and your
partner will work together. Each of you have been assigned a different component to
build. After building your own component, you will work with your partner to build a
third component. Finally, you and your partner will construct the final unit. The final
unit will be shipped to our customers. New Directions, Inc. depends on the production of
the final units. The final units must be completed on time and have high quality in order
to satisfy our customers.

Goals

Each employee is expected to build the individual components and then use them to
construct the final units with his/ her partner. The individual components and the final
units must all be completed. In order to meet our production demands, all employees
must work hard to prepare the satellite casings for our customers.

Goals - Worker #1

Your goal is to produce as many completed units of Component #1 in one production
session that you can. Your performance on the job will be measured by how many units
of Component #1 you complete.

Goals - Worker #2

Your goal is to produce as many completed units of Component #2 in one production
session that you can. Your performance on the job will be measured by how many units
of Component #2 you complete.

One production session will last for 20 minutes. Use the standard operating procedures
and pictures to do your work.

 

APPENDIX C:
STANDARD OPERATING PROCEDURE

118

Job Details & Responsibilities

This page will give you some more information about how to do your job. Please read it
carefully.

1. Job Guidelines

1. All workers are expected to construct the components assigned to them using the
standard operating procedures for your Worker #.

2. All workers are expected to work with their partner to construct the final unit. You
will build your components and your partner will build his or her components and
then together you will build the third component and the final unit. Start the
process over again.

3. Pictures of each of the components and the final unit are included to demonstrate
what the products look like.

4. The standard operating procedures include different instructions for Worker #1 and
Worker #2. Please make sure that you are completing the instructions assigned to
you.

5. You are free to talk and interact with your partner during the production sessions.
However, the experimenter cannot answer many questions.

6. You will have the manuals available to you during the production sessions.

11. Job Responsibilities

In addition to constructing your assigned component, there are a few special tasks
assigned to each worker. The other worker will use a work order to request that you do
the task. You are expected to coordinate these responsibilities as well as continue
constructing your components. The responsibilities are as follows:

Worker 1: Worker 1 is responsible for inserting all fins onto rods. As work orders come
in from other workers, Worker 1 will attach the fin to the rods (provided by the other
worker) and pass them and the work order back to the partner.

Worker 2: Worker 2 is responsible for capping all rods. As work orders come in from
other workers, Worker 2 will attach a close-ended cap to the rods (provided by the other
worker) and pass them and the work order back to the other worker.

 

119

Standard Operating Procedure
Worker 1

Use the following instructions to complete your part of the unit.

Component # l

1.

10.

11.

12.

Remove the following parts from inventory:
0 8 blue rods (R-l)
o 4 wheels (S-6)
0 4 regular spools (S-1)
0 4 orange rods (R-3)

Insert a blue rod (R-l) into the hole of a wheel (S-6). Repeat process three more
times.
Put a regular spool (S-l) onto the empty end of each of the four blue rods (R-l)
from step #4.
Insert a blue rod (R-l) into the other side of the hole on the regular spool (S-l).
Fill out a Work Order (Form 1) that requests Worker #2 to cap the four orange rods
(R-3) with close-ended caps (C-l). Give the work order and four orange rods (R-3)
to Worker #2.
Receive the four capped orange rods (R-3) and the Work Order. Place a check on
the last line (completion line) of the Work Order and place it in file box.
Remove the following parts from inventory:

0 8 red rods (R-4)

o 4 regular spools (S-l)

Insert one of the capped, orange rods (R-3) into a side hole of a regular spool (S-l).
Repeat three more times putting one rod on each spool.

Take the spool and rod combination from step 4 and connect it to the spool
combination from Step 8. Repeat for all four rods. Confirm that the capped, orange
rods (R-3) are pointing out diagonally as seen in the picture for Component #1.
Connect the four rod structures from Step 9 together with 4 red rods (R-4) as seen in
the diagram.

Insert a red rod (R-4) from the top hole of the regular spool (S-I). Repeat for all
four spools.
Set Component #1 aside and start working on Component #3.

120

Component # 3 (Shared Assembly with Worker # 2)

1.

PS”

9°>’.O‘E"

Remove the following parts from inventory:
0 1 red rod (R-4)

2 green rods (R-2)

1 yellow fin (F-2)

1 fitting (F-3)

1 joint (J-l)

1 spool (S-l)

Place the fitting (F-3) over one end of one of the green rods. Make sure that the end
of the green rod extends beyond the connector.

Attach the fin (F-2) to the end of the green rod from step #2 near the fitting.

Attach the two green rods to the center hole of the spool. The green rod with the fin
should be on top.

Put the joint (J-l) on the end of the red rod (R-4).

Connect the red rod to a side hole of the spool (S-l).

Pass the partially completed component to Worker #2.

Wait for Worker #2 then begin working on Final Assembly.

Final Assembly (Shared Assembly with Worker # 2)

Work with Worker #2 in order to combine the components into the final unit as
seen in the picture.

Connect the four spools (S-l) from component 2 to the upright red rods from
component 1.

Place component 3 on component 2. The joints should fit into the red rods on
component 2. The lower green rod of component 3 should go through the center
hole of the spools in component 2.

Put the final assembly in the Completed Unit area.

Return to the beginning of the procedure. Begin working on another unit.

121

Standard Operating Procedure
Worker 2

Component # 2

1.

10.

11.

12.

13.

14.

15.

16.

Remove the following parts from inventory:
0 2 green rods (R-2)

6 orange rods (R-3)

2 five-hole spools (S-5)

1 close-ended cap (C-1)

1 open-ended cap (02)

Insert a green rod (R-2) in one of the outer holes in the base of the spool (S-5).
Insert the second green rod (R-2) in the hole across from the first rod.

Insert the other end of the green rods (R-2) into another five-holed spool (S-5).
Insert four orange rods (R-3) into every hole on the side of the first five-hole spool
(S-S).

Insert an orange rod (R-3) into one of the holes on the side of the second five-hole
spool (S-S).

Place the open-ended cap (02) over the other end of the rod from step #5. Make
sure that the end of the rod sticks through the open-ended cap.

Place the close-ended cap (C-l) over the end of the orange rod (R-3) from step #6.
Fill out a Work Order (Form 1) that requests Worker #1 to place a fin on an orange
rod (R-3). Give the work order and the orange rod (R-3) to Worker #1.

Receive the orange rod (R-3) with the fin (F -2) attached and the Work Order. Place
a check on the last line (completion line) of the Work Order and place it in file box.
Insert the orange rod (R-3) from step #9 into the opposite hole in the spool from the
rod from step #7.

Remove the following parts from inventory:

6 blue rods (R-l)

4 red rods (R-4)

4 regular spools (S-1)

1 five-hole spool (S-S)

Insert two of the blue rods (R-l) in the top of the first spool from step #10 (one with
the cap and fin). Place the blue rods (R-l) in the same holes as the green rods
below.

Place the remaining five-hole spool (S-S) onto the free ends of the two blue rods (R-
1).

Insert 4 blue rods (R—l) into the side holes of the last five-hole spool (S-S) from step
#13.

Place a regular spool (S-l) on the end of each blue rod from step #15. Insert a red
rod out of the top of each of the four regular spools.

Set aside component #2.

 

122

Component # 3 (Shared Assembly with Worker # 1)

1.

b.)

Remove the following parts from inventory:
0 3 red rods (R-4)
o 3joints(J-1)
Receive the partially completed component #3 from Worker #1.
Insert a joint (J -1) over the end of one of the red rods (R-4).
Insert the other end of the red rod (R-4) into the regular spool (S-l) across from the
other red rod.
Repeat steps #3 and #4 for the other two red rods (R-4) and joint (J-l). Place the
red rods at 90 degree angles to each other around the edge of the spool.
Set aside Component #3 then begin working on Final Assembly.

Final Assembly (Shared Assembly with Worker # 1)

Work with Worker #1 in order to combine the components into the final unit as
seen in the picture.

Connect the four spools (S-l) from component 2 to the upright red rods from
component 1.

Place component 3 on component 2. The joints should fit into the red rods on
component 2. The lower green rod of component 3 should go through the center
hole of the spools in component 2.

Put the final assembly in the Completed Unit area.

Return to the beginning of the procedure. Begin working on another unit.

APPENDIX D:
PROGRESS REPORTS

123

Progress Report
Your goal for the last production session was to complete as many of the final units as

you could. Your performance is measured by how many final units you complete.

Indicate how many final units that you and your partner completed during the production

session:

 

124

Progress Report
Worker #1

Your goal for the last production session was to complete as many units of Component #1
as you could. Your performance is measured by how many assigned components you

complete.

Indicate how many units of Component #1 you completed during the production session:

 

Progress Report
Worker #2

Your goal for the last production session was to complete as many units of Component #2
as you could. Your performance is measured by how many assigned components you

complete.

Indicate how many units of Component #2 you completed during the production session:

 

APPENDIX E:
CHANGE MEMOS

 

 

125
Memo

To: All Workers
From: Corporate Headquarters
Re: Change in Schedule

Our largest client has just informed us that they need a rush job on a large special order.
With our current production rate, we will not be able to fill this order, so we will have to
worker faster and harder. It is critical for our customer service policy and our financial
well-being that we meet this demand. Remember the vision of New Directions, Inc.:

Company Vision

We will strive to develop and produce superior, high-quality satellites that always exceed
the expectations of our customers. We will work to become the leader in our industry by
developing new products and envisioning new ways to use them. By hiring ambitious and
creative employees, we will ensure that workers m all divisions will strive to f nd new and
better ways to design, build and market our product. We strive to take our company, our
employees and our customers in new directions.

This sudden increase in production means that all of our workers will need to work as
hard as they possibly can. Everyone will be expected to do a lot more work in the same
amount of time.

All workers will have new, tougher production goals. You are still expected to complete
your assigned components as well as the final units.

Goals

Your goal is to work as hard as possible to produce as many completed final units in this
production session with your partner that the two of you are capable of producing . Your
performance on the job will be measured by how many final units you and your partner
assemble.

One production session will last for 20 minutes. Use the standard operating procedures
and pictures to do your work.

 

 

126
Memo

To: All Workers
From: Corporate Headquarters
Re: Change in Schedule

Our largest client has just informed us that they need a rush job on a large special order.
With our current production rate, we will not be able to fill this order, so we will have to
worker faster and harder. It is critical for our customer service policy and our financial
well-being that we meet this demand. Remember the vision of Johnson & Sons.

Company Vision

We will strive to develop reliable telegraph machines and parts that always meet the
expectations of our customers. We will work to maintain our place in the industry by
producing products with the traditional Johnson & Sons quality for the customers that
we have served for years. By hiring dependable employees, we will ensure that workers
in all divisions will continue the traditions of the company. We strive to maintain the
traditions and customer satisfaction that are the hallmark of our company.

This sudden increase in production means that all of our workers will need to work as
hard as they possibly can. Everyone will be expected to do a lot more work in the same
amount of time. All workers will have new, tougher production goals. You are still
expected to complete your assigned components as well as the final units.

New Goals - Worker #1

Your goal is to work as hard as possible to produce as many completed units of
Component #1 in this production session that you are capable of producing. Your
performance on the job will be measured by how many units of Component #1 you
complete.

New Goals - Worker #2

Your goal is to work as hard as possible to produce as many completed units of
Component #2 in this production session that you are capable of producing. Your
performance on the job will be measured by how many units of Component #2 you
complete.

One production session will last for 20 minutes. Use the standard operating procedures
and pictures to do your work.

APPENDIX F:
SCALES

127

Background

Please answer the following background questions on your answer sheets.

1. What is your gender?
1 = female
2 = male

2. What is your age?
1 = less than 18 years
2 =18 -19 years
3 = 20 - 21 years
4 = 22-23 years
5 = greater than 23 years

3. What is your overall grade point average?

l=O-.9
2=1.0tol.9
3=2.0-2.9
4=3.0-3.9
5=4.0

4. What is your academic class?

1 = First-year
2 = Sophomore
3 = Junior

4 = Senior

5 = Fifth-year

5. How well do you know your partner?
1 = Not at all
2 = Barely know
3 = Know somewhat well
4 = Know very well

 

128
Working Style
People approach work in different ways. Think about how you approach your work.

Using the scale below, indicate how much you agree with the following statements.
Write your answers on your answer sheet.

 

1. Strongly Disagree
Disagree ‘
Neutral

Agree

SAP?!"

Strongly Agree

 

fl
0

I do my best when I’m working on a fairly difficult task.

When I have difficulty solving a problem, I enjoy trying different approaches to see

which one will work.

I try hard to improve on my past performance.

The opportunity to do challenging work is important to me.

The opportunity to extend the range of my abilities is important to me.

The opportunity to learn new things is important to me.

I prefer to work on tasks that force me to learn new things.

When I fail to complete a difficult task, I plan to try harder the next time I work on

it.

9. The things I enjoy the most are the things I do the best.

10. I feel smart when I do something without making any mistakes.

11. I prefer to do things that I can do well rather than things I do poorly.

12. I like to be fairly confident that I can successfully perform a task before I attempt it.

13. I am happiest at work when I perform tasks on which I know that I won’t make any
errors.

14. I feel smart when I can do something better than most other people.

15. The opinions others have about how well I can do certain things are important to
me.

16. I like to work on tasks that I have done well on in the past.

N

”N999.“

 

129

Team Attitudes Scale

Sometimes it may be best when people put the interests of groups ahead of personal
desires. At other times it may be best for people to concentrate on what is best for them
personally. Using the scale below, indicate how much you agree with the following
statements. Write your answers on your answer sheet.

 

 

1. Strongly Disagree

2. Disagree

3. Slightly Disagree

4. Neither Agree nor Disagree
5. Slightly Agree

6. Agree

7. Strongly Agree

 

Only those who depend on themselves get ahead in life.

To be superior a person must stand alone.

If you want something done right, you’ve got to do it yourself.

What happens to me is my own doing.

In the long run the only person you can count on is yourself.

Winning is everything.

I feel that winning is important in both work and games.

Success is the most important thing in life.

It annoys me when other people perform better than I do.

Doing your best isn’t enough; it is important to win.

I prefer to work with others in a group rather than working alone.

Given the choice, I would rather do a job where I can work alone rather than doing

a job where I have to work with others in a group.

13. Working with a group is better than working alone.

14. People should be made aware that if they are going to be part of a group then they
are sometimes going to have to do things they don’t want to do.

15. People who belong to a group should realize that they are not always going to get
what they personally want.

16. People in a group should realize that they sometimes are going to have to make
sacrifices for the sake of the group as a whole.

17. People in a group should be willing to make sacrifices for the sake of the group’s
well-being.

18. A group is more productive when its members do what they want to do rather than
what the group wants them to do.

19. A group is most efficient when its members do what they think is best rather than
doing what the group wants them to do.

20. A group is more productive when its members follow their own interests and

concerns.

.°°>‘F7‘S".‘>E*’.NI“

u—su—au—Ao
.Nr‘SD‘

 

130

Task Confidence Scale #1

Using the scale below, indicate how much you agree with the following statements.
Write your answers on your answer sheet.

 

 

1. Strongly Disagree

2. Disagree 0

3. Slightly Disagree

4. Neither Agree nor Disagree
5. Slightly Agree 1

6. Agree

7. Strongly Agree

 

1. I am confident that I can perform the task given to me.
I can meet the challenges of my role in this task.
I am certain that I can manage the requirements of my position for this task.

I believe I will fare well in this task if the workload is increased.

9‘95”!"

I am confident that I can cope with my role if the simulation becomes more
complex.

6. I am certain I can cope with task components competing for my time.

 

 

131

Goal Commitment Scale #1

Using the scale below, indicate how much you agree with the following statements.
Write your answers on your answer sheet.

 

1. Strongly Disagree

2. Disagree

3. Slightly Disagree

4. Neither Agree nor Disagree
5. Slightly Agree

6. Agree

7. Strongly Agree

 

 

1. It is hard to take this task goal seriously.

It is unrealistic for me to expect to reach this task goal.
Quite frankly, I do not care if I achieve this task goal or not.
I am strongly committed to pursuing this task goal.

It would not take much to make me abandon this task goal.

9‘99?!”

1 think this task goal is a good goal to shoot for.

132

Task Attitude Scale

Using the scale below, indicate how much you agree with the following statements.
Write your answers on your answer sheet.

 

 

1. Strongly Disagree

2. Disagree

3. Slightly Disagree

4. Neither Agree nor Disagree
5. Slightly Agree

6. Agree

7. Strongly Agree

 

bud

50909999935)

~ H
H O
o o

I feel that this task is designed as well as it could be.

I wish I had a better procedure for working with my partner.

My job runs as quickly as it should.

I wish I had a better understanding of what is expected of me.

I would prefer to use different parts to build a component.

I find the task confusing.

I find that every time I start to work on the task I run into obstacles.
There are a lot of petty and arbitrary rules associated with this job.
The demands made of me with this task are unreasonable.

I do not like many of the tasks I have to do.

I often feel frustrated while doing this task.

 

I33

Partner Interaction Scale #1

Some people work well as teammates and others do not work well together. Think about
your interaction with your partner on this task. Using the scale below, indicate how much
you agree with the following statements. Write your answers on your answer sheet.

 

 

1. Strongly Disagree

2. Disagree

3. Slightly Disagree '

4. Neither Agree nor Disagree

5. Slightly Agree

6. Agree

7. Strongly Agree

 

Ni"

My partner and I would probably get along well as teammates.

My partner would say that he/she and I would probably get along well as
teammates.

I could probably count on my partner to help me out.

My partner would say that he/ she could count on me to help out.

My partner’s ability would probably make him/ her an excellent teammate.
My partner would say that my ability would probably make me an excellent
teammate.

I would probably like my partner as a coworker.

My partner would say that he/ she would probably like me as a coworker.

I could probably trust my partner to stick by me if things got difficult.

My partner would say that he/ she could probably trust me to stick by him/ her if
things got difficult.

My partner would probably do an excellent job on a team.

My partner would say that I would probably do an excellent job on a team.
My partner and I would probably have a good working relationship.

My partner would say that he/ she and I would probably have a good working
relationship.

 

134

Task Satisfaction Scale

Using the scale below, indicate how much you agree with the following statements.
Write your answers on the answer sheet.

 

 

1. Strongly Disagree

2. Disagree

3. Slightly Disagree

4. Neither Agree nor Disagree

M

. Slightly Agree
6. Agree

\1

. Strongly Agree

 

H
o

I enjoyed working on this task.

I feel fairly satisfied with the kind of work I did.

Most peOple will be fairly pleased with this task.

I am satisfied with the chances I had to learn new things in this task.
I worked well with my partner.

I am pleased with the method I used to do this task.

I am satisfied with the freedom I had to make changes.

I am glad that I had a chance to work on this task.

>°9°>IQMPPJN

Overall, I am satisfied with the way I dealt with this task.

 

135

Job Change Scale

Compare how you did the task compared to how the directions in the procedure manual.
Using the scale below, indicate how much you agree with the following statements.
Write your answers on your answer sheet.

 

 

1. Strongly Disagree

2. Disagree

3. Slightly Disagree

4. Neither Agree nor Disagree
5. Slightly Agree

6. Agree

7. Strongly Agree

 

fl
0

I did the job differently than specified in the standard operating procedures.
I followed the procedure precisely as described in the manual.

I put my parts together in an order different from the procedure.

I developed a new method for gathering parts.

My partner and I distributed the work in a new way.

I made my job less complicated.

I developed new rules for doing my job.

I changed the tasks of my job.

>0?°.\’.°‘.‘":‘=‘P’!\’

I did the job differently than specified in the manual.

H
.O

I made the task less confusing.

 

APPENDIX G:
VIDEOTAPE CODING SHEET

10.

136

Guidelines for Video Coding Sheet

You need to devote full attention to the videotapes when you are coding. Some of
the behaviors that you are looking for are small and subtle. It is important that you
catch them. The tapes will have inter-rater reliabilities done on them, so we must
have consistency.

You will watch the tapes to see if the workers do any of the 15 behaviors listed on
the coding sheet.

There is a column for Worker #1 and a column for Worker #2 in each time period.
Each worker will get a separate score for each time period.

If a worker performs a behavior, then he or she receives one point. Place a mark in
the box for that behavior for the worker. Each worker can only get one point for
each behavior in each production session.

Give the point to the worker who initiates the action unless the workers do the
behavior independently.

The third column in each time period is labeled “RN” which stands for role
negotiation. Role negotiation refers to whether or not the partners discussed a
possible change before initiating it. When a role innovative behavior occurs, give
the workers a point for role negotiation IF the new behavior is discussed. (i.e.,
“Why don’t I start capping my own rods?” or “Do you think we should stop filling
out the work forms?”). Do not give the workers a point if the role innovative
behavior is not discussed.

Workers can get role negotiation points even if they do not change their behavior.
If the workers discuss a change, but decide not to do it, they still get a point for role
negotiation. So you must pay attention to all of their task-relevant discussions.

Add up the points for each column and put the results in the Total row at the bottom
of the sheet.

Remember that we have to respect the confidentiality of the participants. If you
know any of the participants, do not discuss their performance. Sometimes
participants will have unusual non-task related conversations, but we have an
obligation not to repeat the conversations.

If you have any specific questions, then write them down and ask me later.

 

137

Video Coding Sheet: Tinker Toy Assembly Task

Subj numbers
Tape number
Coder:

Date:

 

 

 

 

 

Session 1

Session 2

 

Behavior

W2

W1

RN

W2

W1

RN

 

1. Worker moves a part bin(s) to make parts more accessible
for self. Must physically move bin.

 

2. Worker removes parts from inventory piece by piece or in
uncounted clumps. Any other unique part removal/ usage.

 

3. Worker takes or gives needed parts (caps or fins) to partner.

 

4. Worker starts capping own rods or putting on own fins.

 

5. Worker stops filling out work order forms.

 

6. Worker stops using written procedures & uses pictures or
models to guide work.

 

7. Worker works on putting several individual components
together at one time. (i. e. makes enough legs for two Comp
#1 's without making a fit” Comp#1 first)

 

8. Worker begins constructing more individual components
before completing a final unit. (i. e. makes two Comp #1 ’s
without putting the first one into a final unit)

 

9. Worker decides not to make final assembly at all.

 

10. Worker helps partner with the partner’s assignment.

 

11. Worker corrects a mistake made by the partner.

 

12. Worker goes back to correct a mistake he/ she made on
earlier unit (one that had been finished and set aside).

 

13. Worker takes control of task and tells partner what to do.

 

14. Worker completes Component # 3 without help.

 

15. Worker completes Final Unit without help.

 

 

Totals

 

 

 

 

 

 

 

 

LIST OF REFERENCES

LIST OF REFERENCES

Amabile, T. M. (1988). A model of creativity and innovation in organizations. In B. M.
Staw & L. L. Cummings (Eds.), Research in Organizational Behavior, (Vol. 10,
pp. 123-167). Greenwich, CT: JAI Press.

Ames, C. (1992). Classrooms: Goals, structures, and student motivation. Journal of
Educatiorgil Psychology. 8_4, 261-271.

Ames, C. & Archer, J. (1988). Achievement goals in the classroom: Student’s learning

strategies and motivation processes. Journal of Educational Psychology, 8_Q, 260-
267.

Asch, S. E. (1951). Effects of group pressure upon the modification and distortion of
judgments. In H. Guetzkow (Ed.), Groups, Leadership, and Men. New
Brunswick, NJ: Rutgers University Press.

Ashforth, B. E., & Saks, A. M. (1996). Socialization tactics: Longitudinal effects on
newcomer adjustment. Academy of Management Journal, 32 (1), 149-178.

Baumeister, R. F ., & Leary, M. R. (1995). The need to belong: Desire for interpersonal
attachments as a fundamental human motivation. Psychological Bulletin, 11_7,
497-529.

Brett, J. M. (1984). Job transitions and personal and role development. In K. M.
Rowland & G. R. Ferris (Eds), Research in Personnel and Human Resources
Management, (Vol. 2 , pp. 155-185). Greenwich, CT: JAI Press.

Bumingham, C., & West, M. A. (1995). Individual, climate and group interaction

processes as predictors of work team innovation. Small Group Research, 2_6 (1),
106-1 17.

Button, S. B., Mathieu, J. E., & Zajac, D. M. (1996). Goal orientation in organizational
research: A conceptual and empirical foundation. Organizational Behavior and

Humpn Decision Processes. _6_7_ (1), 26-48.

Cohen, J. & Cohen, P. (1983). Applied Multiple Regression/ Correlation Analysis for the
Behavioral Sciences (2nd edition). Lawrence Erlbaum Associates: Hillsdale.

138

 

139

Cook, J. D., Hepworth, S. J., Wall, T. D. & Warr, P. B. (1981). The Experience of Work.
Academic Press: London.

Dansereau, F ., Graen, G., & Haga, W. J. (1975). A vertical dyad linkage approach to
leadership within formal organizations: A longitudinal investigation of the role

making process. Organizational Behavior and Human Performance, 13, (1), 46-
78.

Deci, E. L., & Ryan, R. M. (1980). The empirical exploration of intrinsic motivational
processes. Advances in Experimental and Social Psychology, 1_3_, 39-80.

Duda, J. L., & Nicholls, J. G. (1992). Dimensions of achievement motivation in
schoolwork and sport. Journ_al of Educational Psychology, 8_4, 290-299.

Dweck, C. S. (1986). Motivational processes affecting learning. American Psychologist,
41, 1040-1048.

Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation and
personality. Psychological Review. 9_5, 256 - 273.

Elliot, E. S. & Dweck, C. S. (1988). Goals: An approach to motivation and achievement.
Journal of Personplitv and Social Psychology. 54, 5 - 12.

Ettlie, J. E., & Reza, E. M. (1992). Organizational integration and process innovation.
Academy of Management Journil, 35, (4), 795-827.

Farr, J. L. & Ford, C. M. (1990). Individual innovation. In M. A. West & J. L. Farr
(Eds.), Innovation and Creativitygt Work, (pp. 63-80). Chichester: John Wiley
& Sons Ltd.

Farr, J. L., Hofmann, D. A., & Ringenbach, K. L. (1993). Goal orientation and action
control theory: Implications for industrial and organizational psychology.
International Review of Industrial and Organizational Psychology, 8, 193-232.

 

Fisher, C. D. (1986). Organizational socialization: An integrative review. In K. M.
Rowland & G. R. Ferris (Eds.), Research in Personnel and Human Resources
Management, (V 01. 4 , pp. 101-145). Greenwich, CT: JAI Press.

Galbraith, J. R. (1982). Designing the Innovating Organization. Organizational
Dynamics, m, (3), 5-25.

Graen, G. B., & Scandura, T. A. (1987). Toward a psychology of dyadic organizing. In

B. M. Staw & L. L. Cummings (Eds.), Research in Organizational Behavior,
(Vol. 9 , pp. 175-208). Greenwich, CT: JAI Press.

140

Greenberger, D. B. (1991). Responses to inadequate personal control in organizations.
Jougal of Social Issues, 41, (4) 111-128.

Guzzo, R. A., & Shea, G. P. (1992). Group Performance and Intergroup Relations in
organizations. In M. D. Dunnette & L. Hough (Eds.), Handbook of Industrial

and Organizational Psychology. (2nd ed.), (V 01. 3, pp. 269-313). Palo Alto, CA.:
Consulting Psychologists Press.

Hackman, J. R. (1992). Group influences on individuals in organizations. In M. D.

Dunnette & L. Hough (Eds.), Handbook of Industrial and Organizational
Psychology, (2nd ed), (V 01. 3, pp. 199-267). Palo Alto, CA.: Consulting
Psychologists Press.

Hackman, J. R., & Oldham, G. R. (1980). Work Redesign. Reading, MA: Addison-
Wesley. Chapter 4, pp. 71-97.

Hollander, E. P. (1958). Conformity, status, and idiosyncrasy credit. Psychological
Review , _6_5, 117-127.

Hollenbeck, J. R., Klein, H. J ., O’Leary, A. M., & Wright, P. M. (1989). Investigation of
the construct validity of a self-report measure of goal commitment. Journal of
Applied Psychology. B, 951-956.

Hui, C. H., & Villareal, M. J. (1989). Individualism-Collectivism and psychological
needs. Journal of Cross-Cultural Psychology , 2_0 (3), 310-323.

Hulin, C. L., Roznowski, M., & Hachiya, D. (1985). Alternative opportunities and

withdrawal decisions: Empirical and theoretical discrepancies and an integration.
Psychological Bulletin , fl (2), 233-250.

Ilgen, D. R., & Hollenbeck, J. R. (1991). The structure of work: Job design and roles. In
M. D. Dunnette & L. Hough (Eds.), Handbook of Industrial and Organizational
Psychology, (2nd ed.), (Vol. 2, pp. 165-207). Palo Alto, CA.: Consulting
Psychologists Press.

Jabri, M. M. (1991). The development of conceptually independent subscales in the
measurement of modes of problem solving. Educational and Psychological
Measurement , 5_1, 975-983.

James, L. R., James, L. A., & Ashe, D. K. (1990). The meaning of organizations: The
role of cognition and values. In B. Schneider (Ed.), Organizational Climate and
Culture (pp. 40-84). San Francisco: Jossey-Bass.

 

141

James, L. R., & Jones, A. P. (1974). Organizational climate: A review of theory and
research. Psychological Bulletin , _8_l_ (12), 1096-1112.

Janis, I. L. (1982). Groupthink (2nd ed.). Boston: Houghton Mifflin.

Jones, G. R. (1986). Socialization tactics, self-efficacy, and newcomer’s adjustments to
organizations. Academy of Management Journal , 2_9 , 262-279.

Kanfer, R. (1991). Motivation theory and industrial and organizational psychology. In
M. D. Dunnette & L. Hough (Eds.), Handbook of Industrial and Organizational

Psychology, (2nd ed.), (V 01. 2, pp. 75-170). Palo Alto, CA.: Consulting
Psychologists Press.

Kanter, R. M. (1983). The Change Masters. New York: Simon and Schuster.

Katz, D., & Kahn, R. L. (1978). The Social Psychology of Organizations. New York:
Wiley.

Keman, M. C., Bruning, N. S. & Miller-Guhde, L. M. (1994). Individual and group
performance: Effects of task complexity and information. Human Performance,
1(4), 273-289.

Kidwell, R. E., & Bennett, N. (1993). Employee propensity to withhold effort: A
conceptual model to intersect three avenues of research. Academy of
Management Review , fl, 429-456.

Kieman, M. J. (1993). The new strategic architecture: learning to compete in the twenty-
first century. Academy of Management Executive , 1(1), 7-21.

King, N. (1990). Innovation at work: the research literature. In M. A. West & J. L. Farr
(Eds.), Innogflion rmd Cregtivity flgk, (pp. 15-59). Chichester: John Wiley
& Sons Ltd.

Kirton, M. J. (1976). Adapters and Innovators: A description and measure. Journal of
Applied Psychology, 6, 622-629.

Kozlowski, S. W. J., Gully, S. M., Smith, E. M., Nason, E. R., & Brown, K. G. (1995,
May). Sequenced m_astery training and advanced organizers: Effects on learning,
self-efficacy, performance and generalization. Paper presented at the Tenth

Annual Conference of the Society for Industrial and Organizational Psychology,
Orlando, FL.

142

Kozlowski, S. W. J ., & Hults, B. M. (1987). An exploration of climates for technical
updating and performance. Personnel Psychology , fl, 539-563.

Latham, G. P., & Locke, E. A. (1991). Self-regulation through goal setting.
Organizational Behavior and Human Decision Processes , 59, 212-247.

Locke, E. A., Durham, C. C., Poon, J. M. L. & Weldon, E. (in press). Goal setting,
planning, and performance on work tasks for individuals and groups In S. L.
Friedman & E. K. Scholneck (Eds.), Why, How and When do we Plan? The
Developmental Psycholgy of Planning. Hillsdale, NJ: Lawrence Erlbaum
Associates.

Locke, E. A., & Latham, G. P. (Eds) (1990). A Theory of Goal Setting and Task
Performance. Englewood Cliffs, NJ: Prentice Hall.

Maas, J. B. (1965). Pattern scaled expectation interview: Reliability studies on a new
technique. Joum_al of Applied Psychology, 42, 431-433.

Mitchell, T. R., & Silver, W. S. (1990). Individual and group goals when workers are
interdependent: Effects on task strategies and performance. Journal of Applied
Psychology , _7_5_, 185-193.

Morris, M. H., Avila, R. A., & Allen, J. (1993). Individualism and the modern
corporation: Implications for innovation and entrepreneurship. Journal of
Management , 12 (3), 595-612.

Nason, E. R. (1995). Horizontal Team Member Exchange (HMX): The effects of dyadic

relationship Quality on team processes and outcomes. Unpublished doctoral
dissertation, Michigan State University, East Lansing.

 

Nicholson, N. (1984). A theory of work role transitions. Administrative Science
Quarterly , 22 , 172-191.

O’Leary, A. M., Martocchio, J. J., & Frink, D. D. (1994). A Review of the influence of

group goals on group performance. Academy of Management Journal , 3_7_ (5),
1285-1301.

Schein, E. H. (1971). Occupational socialization in the professions: The case of role
innovation. Journal of Psychiatric Research, _8_, 521-530.

Schneider, B. (1985). Organizational Behavior. Annual Review of Psychology, 3p, 573-
611.

143

Schneider, B., & Bowen, D. E. (1985). Employee and customer perceptions of service in
banks: Replication and extension. Journal of Applied Psychology, 19(3), 423-
433.

Scott, S. G., & Bruce, R. A. (1994). Determinants of innovative behavior: A path model

of individual innovation in the workplace. Academy of Management Journal , fl
(3), 580-607.

Seers, A. (1989). Team-member exchange quality: A new construct for role-making
research. Organizational Behavior and Human Decision Processes , fl, 118-135.

Shamir, B. (1990). Calculations, values, and identities: The sources of collectivistic work
motivation. Human Relations. fl, 313-332.

Spector, P. E. (1975). Relationships of organizational frustration with reported
behavioral reactions of employees. Journal of A lied Ps cholo , _6_0, 635-637.

Staw, B. M. (1990). An evolutionary approach to creativity and innovation. In M. A.
West & J. L. Farr (Eds.), Innovation and Creativity at Work, (pp. 287-308).
Chichester: John Wiley & Sons Ltd.

Staw, B. M., & Boettger, R. D. (1990). Task revision: A neglected form of work
performance. Academy of Management Journal , 3; (3), 534-559.

Van De Ven, A. H. (1986). Central problems in the management of innovation.
Management Science. _3_2, (5), 590-607.

Vance, R. J ., Kuhnert, K. W., & Farr, J. L. (1978). Interview judgments: Using external
criteria to compare behavioral and graphic scale ratings. Organizational Behavior
and Human Performance 22, 279-294.

 

Van Dyne, L., Cummings, L. L., & Mclean Parks, J. (1995). Extra-role behaviors: In
pursuit of construct and definitional clarity (A bridge over muddied waters). In B.

M. Staw & L. L. Cummings (Eds.), Research in Organizational Behavior, (V 01.
17 , pp. 215-285). Greenwich, CT: JAI Press.

Van Maanen, J. (1978). People processing: Strategies of organizational socialization.
Organizational Dymics. _7_ (1), 19-36.

Van Maanen, J ., & Schein E. H. (1979). Toward a theory of organizational socialization.
In B. M. Staw (Ed), Research in Organizational Behavior, (V 01. l , pp. 209-264).
Greenwich, CT: JAI Press.

144

Wagner, J. A. (1995). Studies of Individualism-Collectivism: Effects on cooperation in
groups. Academy of Management Journal , 3_8_ (1), 152-172.

Wagner, J. A. & Moch, M. K. (1986). Individualism-Collectivism: Concept and
Measure. Group & Organization Studies , 1_1_ (3), 280-303.

Weingart, L. R. (1992). Impact of group goals, task component complexity, effort, and
planning on group performance. Journal of Applied Psychology , 21(5), 682-
693.

Weingart, L. R. & Weldon, E. (1991). Processes that mediate the relationship between a
group goal and group member performance. Human Performance , _4 , 33-54.

Weldon, E. & Weingart, L. R. (1988, August). A theopy of group goals and group
performance. Paper presented at the meeting of the Academy of Management,
Anaheim, CA.

West, M. A. (1987a). A measure of role innovation at work. British Journal of Social
Psychology , 2_6 , 83-85.

 

West, M. A. (1987b). Role Innovation in the world of work. British Journal of Social
Psychology , 2p , 305-315.

West, M. A. (1990). The social psychology of innovation in groups. In M. A. West & J.
L. Farr (Eds.), Innovation and Crgtivity at Work, (pp. 309-333). Chichester:
John Wiley & Sons Ltd.

West, M. A. & Rushton, R. (1989). Mismatches in the work-role transition. Journal of
Occupational Psychology , Q , 271-286.