 

 

 

   

.. 3-1. 0‘

 
 
 

 

 

 

 

 

This is to certify that the
thesis entitled

EFFECTS OF REWARD INTERDEPENDENCE ON TEAM
PERFORMANCE:
A PROCESS APPROACH

presented by

Chi Cong Dang

has been accepted towards fulﬁllment
of the requirements for the

Masters of Art degree in Pslchology

 

72% M%

“Major Professor’ ﬁnature
July 18, 2004

Date

MSU is an Aﬁirmative Action/Equal Opportunity Institution

 

_-—-—.—-.—-—._-_-_._-_ t __

 

 

 

 

LIBRARY
Michigan State
University

 

 

 

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/DateDue.p65-p.15

 

EF

 

EFFECTS OF REWARD INTERDEPENDENCE ON TEAM PERFORMANCE:
A PROCESS APPROACH

By

Chi Cong Dang

A THESIS

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

MASTERS OF ART
Department of Psychology

2004

ABSTRACT

EFFECTS OF REWARD INTERDEPENDENCE ON TEAM PERFORMANCE:
A PROCESS APPROACH

By
Chi Cong Dang

A lab study of 80 3-person teams was conducted to examine the process in which
positive reward interdependence inﬂuenced team performance of interdependent tasks.
At the individual level, positive reward interdependence was found to have a positive
effect on members’ cooperative orientation and a negative effect on their competitive
orientation. In turn, cooperative orientation had a positive inﬂuence on individual
teamwork effort, and competitive orientation had a positive effect on individual taskwork
effort. At the team level, positive reward interdependence was found to have a positive
effect on teamwork effort and a negative effect on taskwork effort. Teamwork effort and
taskwork effort subsequently had independent positive inﬂuences on team performance.
The effect of taskwork effort was, however, moderated by the level of task

interdependence. Implications of the ﬁndings for how to reward teams were discussed.

ACKNOWLEDGEMENTS

This work was completed with the support and assistance of many individuals. I
am very grateful for their encouragements and contributions over the years. I would like
to thank Dr. Sheldon Zedeck for leading me to the ﬁeld of the Industrial/Organizational
Psychology. My deep gratitude goes to Dr. Daniel Ilgen, my thesis committee chair, who
guiding me through the entire process. I greatly appreciate his caring encouragement,
thoughtful advice, and hard work. I would also like to thank Drs. Kevin Ford and Steve
Kozlowski, my other thesis committee members, for providing many helpful feedbacks
and comments. Many thanks go to Kelley Baas, Jennifer Olson, Jeffrey Pink, and
Michael Truman, my undergraduate assistants, who helped with the data collection.

I would also like to express my gratitude to Michigan Statement University and
its Department of Psychology for the education and ﬁnancial support that I have received
for the past three years. I would like to thank my parents and siblings for their ﬁ'equent
encouragements and supports. Finally, my deep gratitude goes to Maria, my wife, for
her deliberate sacriﬁce, unconditional love, and unwavering belief that I should pursue

my calling and will succeed.

iii

 

i"
.2..." ' .L-‘i____" "

 

LIST
LIST

INTR‘

MET

 

TABLE OF CONTENTS

LIST OF TABLES .......................................................................... vi

LIST OF FIGURES ........................................................................ vii

INTRODUCTION .......................................................................... l
Salience of Team Rewards ......................................................... l
A Brief Review of Team Reward Literature ..................................... 2
Proposed Model of Team Reward Effects ........................................ 8
Multilevel Consideration ............................................................ 11
Task and Task Interdependence .................................................... 13
C00perative and Competitive Orientations ....................................... 15
Teamwork and Taskwork Effort ................................................... l7
Determinants of Individual Teamwork and Taskwork .......................... 21
Determinants of Team-Level Teamwork Effort and Taskwork Effort ....... 26

Effects of Teamwork Effort and Taskwork Effort on Team Performance. .. 28

METHODS .................................................................................... 31
Training of Team Members ......................................................... 31
Task .................................................................................... 32
Task Interdependence ............................................................... 35
Reward Structure Manipulations .................................................. 35
Manipulation Checks ............................................................... 36
Procedure ............................................................................. 36

iv

 

Fr .
i i" ni- 'I‘IL
“HUI-3.1-

RESI

DISC

 

APP!

BIBI

Measures .............................................................................. 38

Cooperative and Competitive Orientations .............................. 38

Individual-Level Task Effort .............................................. 39

Individual-Level Teamwork Effort ....................................... 39

Teamwork Effort ............................................................ 39

Taskwork Effort ............................................................ , 40

Team Performance ......................................................... 40

RESULTS .................................................................................... 41

Analysis Strategies .................................................................. 41

Individual-Level Analyses ......................................................... 41

Team-Level Analyses ............................................................... 53

DISCUSSION ................................................................................ 61

Theoretical Contributions ........................................................... 61

Practical Implications ............................................................... 64

Limitations ........................................................................... 65

New Research Directions ........................................................... 67

APPENDICES ............................................................................... 70
BIBLIOGRAPHY ........................................................................... 1 12

 

 

Table
Table
Table i
Table
Table

Table
Table

Table

 

LIST OF TABLES

Table 1. Summary Statistics for Individual-Level Variables .......................... 42
Table 2. HLM Results of Effects on Individual-Level Teamwork Effort. . . . . . . . 47
Table 3. HLM Results of Effects on Individual-Level Taskwork Effort ............ 51
Table 4. Summary Statistics for Team-Level Variables ............................... 53
Table 5. Effects of Task and Reward Interdependence on Team Performance. . 55

Table 6. The Effects of Task and Reward Interdependence

on Teamwork and Taskwork .................................................... 57
Table 7. The Effects of Teamwork Effort and Taskwork Effort

on Team Performance ............................................................ 58
Table 8. Moderating Effects of Task Interdependence ................................. 59

vi

LIST OF FIGURES

Figure 1. Effects of Reward Interdependence on Team Performance:
A Process Approach ............................................................. 10

Figure 2. Portion of the Web Page Assigned to the Greek Letter Specialist. . . . . 33
Figure 3: Individual-Level Model ......................................................... 43

Figure 4: Team-Level Model ............................................................... 54

vii

INTRODUCTION

Salience of Team Rewards

Teams are increasingly organized as primary work units within organizations
(Guzzo & Dickson, 1996). One survey by Fortune magazine estimated that half of
America's largest companies were experimenting with teams (Dumaine, 1990). Another
survey by Lawler, Mohrman, and Ledford (1992) reported 51 percents of Fortune 1,000
companies had more than 20 percents of their employees in teams. The number
increased to 68 percents in a follow-up survey (Lawler, Mohrrnan, & Ledford, 1995).

As the usage of teams as work units becomes more widespread, organizations ﬁnd
that individually based reward systems do not necessarily support a team-based approach
to organizing work (Zingheim & Schuster, 1997). As a result, more organizations are
moving toward using team-based pay systems, in which a portion of an individual’s pay
is contingent on the performance of his/her team. Lawler et a1. (1995) reported that from
1990 to 1993 the percentage of organizations using team-based pay plans covering at
least 20 percents of the workforce increased from 22 to 31 percents. The percentage
covering over half of the workforce rose from 12 to 17 percents. Gross (1995) reported
in another survey that 51 percents of companies had group-based rewards or were
considering implementing them. The trend for group-based reward is continuing as many
organizations adopt the practice (Zingheim & Schuster, 1997).

As team-based rewards become more prevalent, organizations face the challenge
of determining how to pay individuals in a team-based environment (Guzzo & Dickson,
1996). Under team-based reward system, a team receives a reward or bonus contingent

on its performance. Higher team performance is typically awarded with a higher reward.

The critical issue organizations are facing is how to distribute the team reward among
team members. Two methods of reward distribution are often used. One is to distribute
the reward equally, regardless of their individual contributions. The other is to divide the
reward differentially, contingent on individual contributions. When team reward is
divided differentially, the question then becomes how differentially the reward should be
divided.
A Brief Review of Team Reward Literature

Research literature on team reward methods has relied heavily on the theory of
cooperation and competition (Deutsch, 1949; Deutsch, 1985) and its conceptualizations
of reward interdependence. According to this theory, positive reward interdependence
exists when team members’ rewards are positively correlated. With positive reward
interdependence, team members sink or swim together. An example is the situation in
which members of a basketball team share the prize equally when they win a tournament.
Negative reward interdependence is present when team members’ rewards are negatively
correlated. As one member gains, other members lose in this type of reward distribution.
An example is an arrangement in which only the best salesperson is given the team bonus
when the sales team achieves its performance target. Reward independence exists when
team members’ rewards are only contingent on individual contributions and independent
of other members’ rewards.

The main ﬁnding of team reward research is that for interdependent tasks positive
reward interdependence results in better team performance than reward independence and
negative reward interdependence (Cotton & Cook, 1982; DeMatteo, Eby, & Sundstrom,

1998; Deutsch, 1949; Gordon, Welch, Offringa, & Katz , 2000; Johnson & Johnson,

1989; Johnson, Maruyama, Johnson, Nelson, & Skon, 1981; Miller & Hamblin, 1963;
Slavin, 1982; Stanne, Johnson, & Johnson, 1999; Wageman & Baker, 1997). With a few
exceptions (Gordon et al., 2000; Rosenbaum, Moore, Cotton, Cook, Heiser, Shovar, &
Gray, 1980; Wageman & Baker, 1997), most studies operationalized positive reward
interdependence as equal division of rewards among team members. Negative reward
interdependence was studied as a winner-take-all or high-differential reward distribution
in which members’ rewards are negatively correlated (winner-take-all is often the case).
Reward independence is operationalized as individual-based distribution (also called
reward independence or individualistic reward distribution) in which members are
rewarded in proportion to their performance independent of the team performance.

Most studies in the team reward literature explained the effect of reward
interdependence through its impact on teamwork processes, deﬁned as behaviors directed
toward the team members (Ilgen & Sheppard, 1999). In a review of hundreds of studies,
Johnson & Johnson (1989) reported that the main reason for the superior performance of
positive reward interdependence is that it promotes helping behaviors, resource and
information exchange, constructive feedback, constructive approach to conﬂicts,
mutually supportive behaviors, acting in a trusting and trustworthy way, and positive
mutual inﬂuence. With a few exceptions, the effect of reward interdependence on
taskwork behaviors, deﬁned as behaviors directed toward the task (Ilgen & Sheppard,
1999), has been ignored.

In our view, the ﬁnding that positive equal reward distribution is more effective
than reward independence and negative reward interdependence is of limited

signiﬁcance. Negative reward interdependence such as winner-take-all distribution is at

the extreme end of the reward distribution spectrum. It is very competitive in nature and
counter-productive in many situations. It is what Mischel (1973) called the strong
situation. In the strong situation, individuals likely exhibit uniform behaviors in order to
cope the demanding nature of the situation. Under a winner-take-all reward system, team
members are likely extremely competitive and destructive to one another, leading to great
decrement in the team performance of an interdependent task. Individualistic reward
distribution is also highly competitive, though to a lesser extent. It was known as
ineffective for rewarding teams early on in many organizations (Zingheim & Schuster,
1997; Hackman, 1998). With the ﬁnding that equal reward distribution is more effective
than these two types of highly competitive reward distributions, the question of how to
distribute team reward largely remains. The ﬁnding only indicates that equal reward
distribution is superior to reward systems that focus only on individual performance. It
does not necessarily say that equal reward distribution is the most effective and
organizations should use it to reward team members.

For many interdependent tasks, team performance is likely inﬂuenced by both
teamwork and taskwork behaviors (Salas, Dickinson, Converse, & T annenbaum, 1992).
It follows that an effective reward distribution should not only promote teamwork
behaviors but also stimulates individual task performance. Because a mix reward
system in which a portion of the reward is distributed equally and the rest is distributed
proportionally has cooperative as well as competitive components, four studies (Gordon
et al., 2000; Rosenbaum et al., 1980, Study 2 and 2a; Wageman & Baker, 1997) have

attempted to study its effect.

Two lab studies conducted by Rosenbaum et a1. (1980) examined the effects of
mixed reward structures for a tower-building task. In one study, 60 3-member teams of
college students were assigned to ﬁve reward conditions that varied in the portion of
equally distributed team rewards (100%, 80%, 50%, 20%, and 0%, respectively) and the
portion of proportionally-distributed team rewards (0%, 20%, 50%, 80%, and 100%,
respectively). The results showed that the equal reward distribution (100% of team
reward was distributed equally) led to higher team performance than all other reward
distributions, which did not differ from one another. The second study was designed the
same way as the ﬁrst except that the competitive element of the mixed reward
distribution was changed from proportional distribution to the distribution in which the
highest performer of the team was rewarded proportionally for his/her contribution while
other team members were not rewarded. The results indicated that equal reward
distribution outperformed the 50-50 mixed reward distribution and the extremely
competitive 0-100 reward distribution. However, it did not differ from the 80-20 and 20-
80 mixed reward distributions.

In Gordon et a1. (2000)’s study, 32 3-member teams of mostly college students
worked under mixed and equal reward distributions on a simulation game in which team
members interacted to manipulate garbage trucks to collect trash in a city. In the equal
reward condition, the team reward was distributed equally. In the mixed reward
condition, 50 percent of the team reward was distributed equally, and the rest was divided
proportionally. The results showed that the reward conditions did not result in different

levels of team performance.

Wageman and Baker (1997) examined 56 2-members teams’ performance on an
interdependent task of correcting technical writing syntax errors, using a 3 (reward
interdependence) X 3 (task interdependence) design. Reward interdependence varied
from high (an equal distribution) to medium (a mixed reward distribution in which half of
the team reward was distributed proportionally and the rest was divided equally) to low (a
proportional distribution). The task was designed so that a member of the team had to
depend on the other’s knowledge and assistance to complete his/her assigned team
subtask. Task interdependence was manipulated by varying the portion of errors a team
member could correct by his/her own knowledge from 50 percent (high interdependence)
to 67 percent (medium interdependence) and to 100 percent (zero interdependence). The
results showed that reward and task interdependence interacted to inﬂuence team
performance. For the high-interdependent task, equal reward distribution was most
effective. For the moderate interdependence task, the mixed reward distribution was
most effective. For the independent task, the proportional distribution was most
effective.

If different mixed-reward systems can be considered as representing different
levels of positive reward interdependence, the results of the four studies reviewed above
suggested that the effect of positive reward interdependence on team performance is
rather complex. Rosenbaum et al.’s (1980) ﬁrst study showed that team performance at
the very high level of positive reward interdependence (100-0 mix) is higher than those of
lower levels of positive reward interdependence, which did not differ statistically.
Operationalizing positive reward interdependence differently, their second study

indicated that team performance changed in a non-linear way across different levels of

positive reward interdependence. It was high at the very high level of positive reward
interdependence (100-0 mix), and it was still high at the high level of positive reward
interdependence (80-20 mix). However, it dropped at the moderate level of positive
reward interdependence (50-50 mix). It then went back to high at the low level of
positive reward interdependence (20-80 mix), and it dropped again at the very low level
of positive reward interdependence (0-100 mix). Gordon’s (2000) indicated that team
performance at the very high (100-0 mix) and moderate (SO-50 mix) levels of positive
reward interdependence did not differ. Wageman and Baker’s (1997) study added a new
complexity to the relationship of positive reward interdependence and team performance
by considering the levels of task interdependence. For high interdependence task, team
performance decreased as positive reward interdependence changed from very high (100-
0 mix) to moderate (50-50 mix) to very low (0-100 mix). For low interdependence task,
team performance, however, increased when positive reward interdependence changed
from very high level (100-0 mix) to moderate level (SO-50 mix). It decreased when
positive reward interdependence changed from moderate level to very low level (100-0
mix).

Taken together, these studies suggest that the effect of positive reward
interdependence on team performance likely depends on the operationalization of
positive reward interdependence and the level of task interdependence. Although this
conclusion is based on a small number of studies, it is critical to ask why the effect of
positive reward interdependence on team performance varies.

To date, our knowledge of the process in which positive reward interdependence

affect team performance have been very limited. Wageman & Baker (1997) assumed that

equal reward distribution facilitates teamwork activities on the one hand, but it induces
free-riding among team members on the other. These assumptions, however, were not
tested in their study. Rosenbaum et al. (1980) and Gordon (2000) attempted to explain
the effect of positive reward interdependence through teamwork behaviors. Like most
studies in team reward research literature, their approach was to manipulate the reward
distribution and then measured performance and teamwork behaviors at individual and/or
team levels. The relationships of reward interdependence to team performance and
teamwork behaviors were subsequently analyzed by comparing the aggregated means at
the contextual (reward) level using AN OVA or linear regressions. Such an approach had
a number of limitations. In term of the content, it has largely ignored the effect of
positive reward interdependence on taskwork behaviors as important mediators. Because
team performance of an interdependent task is likely inﬂuenced by both teamwork and
taskwork behaviors (Salas, Dickinson, Converse, & Tannenbaum, 1992), failure to
examine taskwork behaviors is a critical limitation in our view. In term of data analysis,
the reward-level analysis treats team-level and individual-level variances as measurement
errors that should be ignored. We think this is also a serious limitation because it is
important to know how individuals and teams react differently under the inﬂuence of the
reward system. By ignoring individual and team differences, researchers missed the
opportunity to examine other factors that may interact with positive reward
interdependence to inﬂuence team and/or individual-level behaviors.
Proposed Model of Team Reward Effects

Given the salience of the question of how to reward team members and the

limited knowledge of team reward effect, the purpose of the current study is to examine

the mediating processes through which positive reward interdependence inﬂuences team
performance on interdependent tasks at the multiple levels of analysis. The model
explaining the process is depicted in Figure 1. To facilitate the reading of the model,
contextual factors are depicted by the boxes with broken borders; the individual-level
factors are represented by the boxes with regular borders; and the team-level factors are
described by the boxes with bolded borders. According to the model in Figure 1, positive
reward interdependence inﬂuences team performance through psychological and
behavioral mechanisms. Speciﬁcally, positive reward interdependence affects team
members’ competitive and cooperative orientations. Member’s orientations and task
interdependence inﬂuence team members’ expressed behaviors conceptualized as
individual teamwork effort and taskwork effort. The individual-level behaviors are
assumed to combine to give rise to the team-level teamwork effort and taskwork effort,
which are also affected by task interdependence. Together, teamwork effort and
taskwork effort inﬂuence team performance, but their effects are moderated by task
interdependence. (From this point on, please note that team-level teamwork effort and
taskwork effort would be simply referred as teamwork effort and taskwork effort,
respectively; the analogous constructs at the individual level are referred as individual
teamwork effort and individual taskwork effort).

The constructs in the model and their relationships, other than positive reward
interdependence, are further elaborated in the remaining sections of the Introduction.

Speciﬁcally, the constructs are going to be deﬁned. The relevant literature is reviewed.

5803me 3305 < Hoogcuoﬁom 83... no 665953985 Emaom no maoobm n. 8%.".

 

 

 

 

 

 

 

ooggowhom
v.5. ease

 

 

 

     

 

 

 

 

 

 

 

couﬁcoto
03:89:00

 

 

/.

 

 

ﬂobm
€ch3...
team
6195.3...
.3239:
8:353?an u
vv.3. ..
team
\ 6:03:80...
k $5365

 

 

 

couﬁcoco
3:82.000

 

 

 

 

 

team
foaﬁaoh

 

 

005208283
@833.

10

 

 

 

Because team behavior and function is a multilevel phenomenon (Kozlowski & Bell,
2003), we think it is important to ﬁrst clarify the level of analysis in order to facilitate the
discussions of the constructs that follow.

Multilevel Con_sideration

Team characteristics and processes must be regarded as a multilevel phenomenon
(Kozlowski & Bell, 2003). Teams are composed of team members. They do not think,
feel, or behave; team members do (Kozlowski & Bell, 2003). Teams live in a context
and are inﬂuenced by the context’s characteristics. Within such a context, team members
interact and inﬂuence one another. Their interactions and mutual inﬂuence give rise to
emergent properties of the team and can change the characteristics of the surrounding
context. Because team characteristics and processes exist at individual and team levels
within a context, it is critical that team research must have an explicit multilevel
consideration. The level consideration suggests that the constructs, their measurements,
and data analysis need to be aligned so that the conclusions about the constructs’
relationships can be properly inferred (Kozlowski & Bell, 2003).

In our model, positive reward interdependence and task interdependence are
contextual factors. Positive reward interdependence is expected to inﬂuence team
members’ psychological states, conceptualized as cooperative and competitive
orientations. These psychological states and task interdependence were expected to
inﬂuence team member behaviors as reﬂected in individual teamwork and taskwork
effort constructs. The linkage between reward and task interdependence, team members’

psychological states, and behaviors therefore could properly be analyzed at the individual

11

level nested within teams. HLM is an appropriate data analytic technique (Bliese, 2000;
Hoffman, Grifﬁn, & Gavin 2000; Bryk & Raudenbush, 1992).

As indicated in the model, individual teamwork effort and taskwork effort are
assumed to give rise to teamwork and taskwork effort at the team level. Because the
process of deriving team-level behaviors from team members’ behaviors is largely guided
by appropriate logics--not so much by empirical evidence--the relationship between
individual-level teamwork effort and team-level teamwork effort as well as the
relationship between individual-level taskwork effort and team-level taskwork effort
would not be tested.

As indicated in the model, individual teamwork effort and taskwork effort are
assumed to give rise to teamwork effort and taskwork effort at the team level. Because
the process of deriving team-level behaviors from team members’ behaviors is largely
guided by appropriate theories-mot so much by empirical evidence--the relationship
between individual-level teamwork effort and team-level teamwork effort as well as the
relationship between individual-level taskwork effort and team-level taskwork effort
would not be tested.

At the team level, positive reward interdependence and task interdependence were
hypothesized to inﬂuence teamwork effort and taskwork effort. In turn, teamwork effort
and taskwork effort were expected to inﬂuence team task performance, but their effects
were moderated by task interdependence. Because reward and task interdependence are
contextual factors and because teamwork effort, taskwork effort, and team task
performance are team-level constructs, analyses of the relationships among these

constructs could properly done at the team level with multiple regressions.

12

Task and Task Interdependence

In team studies, it is important to specify the nature of the task to establish the
necessary boundary conditions (Kozlowski & Bell, 2003). The task we addressed in this
work has two distinguishing characteristics. First, the task can be broken down to
relative independent components, which require specialized expertise to complete. Team
members have specialized roles based on their expertise and are assigned the components
that ﬁt their expertise. Second, team members do not have the full knowledge and skills
to complete all of their own component. They work on their own on the subcomponents
that they have the knowledge for but need assistance from other team members for the
components that they do not.

This form of task is common in many product planning and deve10pment teams.
In order to speed the product to the market, the team task is broken down to relatively
independent components so that they can be worked on in parallel in the interest of time.
Components are assigned to team members who have the relevant expertise. Because a
component often is too complex for one member to handle, he/she often needs assistance
on the parts that he/she does not have the expertise. The components are brought together
at regular intervals and conﬁgured into the team product. The team evaluates the
product’s progress and makes adjustment to the teamwork process. The cycle is repeated
until the product meets certain criteria or the deadline they have to present it to the
customers.

One important dimension of a task is task interdependence. Task interdependence
is often deﬁned as the extent to which team members must exchange information,

resource, and expertise or actually work together to achieve their desired performance

13

(Cummings, 1978; Thompson, 1967; Van Der Vegt, Eman, Van De Vliert; 2001;
Wageman & Baker, 1997). Many researchers (Bell & Kozlowski, 2002; Tesluk,
Mathieu, Zaccaro, & Marks, 1998; Thompson, 1967) classiﬁed tasks by examining the
workﬂow through which team members’ inputs are combined to yield the team output.
Tasks are frequently classiﬁed into four groups with increasingly complex workﬂows:
pooled, sequential, reciprocal, and intensive interdependence (Bell & Kozlowski, 2002;
Tesluk, Mathieu, Zaccaro, & Marks, 1998; Thompson, 1967). Pooled interdependence is
a task in which subtasks are performed separately in any order. Sequential
interdependence is a task in which subtasks are completed in pre-speciﬁed sequence.
Reciprocal interdependence is different from sequential in that the reversals in the order
of the workﬂow between any two team members are often required to complete the
whole task. Intensive interdependence is the workﬂow in which team members have to
interact with one another to get the task done.

Other researchers classiﬁed tasks by the manner in which team members’ inputs
are combined to yield the team output (Steiner, 1975). An additive task is a task in which
members’ inputs are added to form the team output. A disjunctive task is a task in which
the input of the best team member is considered the team output. A conjunctive task is a
task in which the input of the worst team member is evaluated as the team output. A
complex task is the task that members’ inputs are combined in a linear (regression
weighting) or nonlinear way to form the team output.

In the current study, we deﬁne task interdependence as the degree to which team
members must share materials, information, or expertise in order to achieve valued group

outcome. Similar to Wageman and Baker (1997), this view of task interdependence

14

perceives task interdependence as the extent to which members have to depend on others
for information, resource, and expertise for their task performance. Task interdependence
increases when the task requires team members to depend more on one another’s
material, information, or expertise to complete the team task (Wageman & Baker, 1997).
This conceptualization of task interdependence ﬁts the intensive workﬂow described in
Thompson’s classiﬁcation and the complex task described in Steiner’s typology. It views
tasks in terms of team members’ resource interdependence. Further, it recognizes that the
workﬂow and behavioral interactions within small teams tend to be complex and shifting
and may not ﬁt any pattern. With this view, task interdependence is a structural feature of
work that constrains the level of behavioral interactions among team members. The
behavioral interactions within team themselves are unspeciﬁed, and they likely change
over time within teams and can be different between teams.
Cooperative and Competitive Orientations

According to theory of cooperation and competition (Deutsch, 1949, 1985;
Johnson & Johnson, 1985), positive reward interdependence leads to cooperative
orientation, whereas negative reward interdependence results in competitive orientation.
Cooperative and competitive orientation are deﬁned as situation-induced psychological
states reﬂecting how team members perceive their outcomes are linked and how they
likely respond to each other’ behaviors (Deustch, 1985). Team members with a
cooperative orientation perceive that their outcomes are positively linked, believing they
swim or sink together. Effective actions by one member are expected to enhance other
members’ success while ineffective ones hinder it. They celebrate each other’s success

and commiserate each other’s failure. Team members are cooperative and

15

psychologically open to the others’ inﬂuences. On the other hands, team members with
a competitive orientation perceive that their outcomes are negatively linked, believing
one member’ 3 win means others’ losses. Effective actions by one member are expected
to hinder other members’ success while ineffective ones enhance it. They are distressed
and at others’ success and rejoice at other’s failure. Team members are competitive and
psychologically close to the others’ inﬂuence.

Although Deustch considered cooperative and competitive orientations as two
ends of a continuum (Deustch 1949, 1985), empirical evidence indicates that they are two
separate dimensions. Based on Deustch’s theory of cooperation and competition, the
social interdependence scales (Johnson & Norem-Hebeisen, 1979) were developed to
measure students’ cooperative and competitive attitudes. Factor analyses of the data
gathered from more than 6,000 students from kindergarten through college showed that
cooperative and competitive attitudes were two orthogonal factors (Johnson & Norem-
Hebeisen, 1979). Subsequent studies showed that the cooperative and competitive
attitudes were unrelated (Onwueqbuzie & Daley, 1998; Bailey, Onwueqbuzie & Daley,
2000; Jones, Slate, & Marini, 1995).

While cooperative and competitive attitudes are less situation speciﬁc than
Deustch’s constructs of cooperative and competitive orientations, evidence gathered by
more situation-speciﬁc measures also showed supports that cooperative and competitive
orientations are two separate dimensions. Vegt, Emans, and Evert (1999) developed a
measure of perception of goal interdependence, which we believe is a narrower measure

of social orientation, to study the effect of goal interdependence on team performance.

16

Factor analysis of the data gathered with their measure resulted in two factors: negative
and positive perceptions of goal interdependence.

As a result, in the present study we treated social orientation as two separate
constructs: cooperative orientation and competitive orientation. We deﬁne cooperative
orientation as a psychological state that predisposes an individual to cooperate and act
mainly for the interest of his/her team. Competitive orientation refers to a psychological
state that predisposes an individual to strive and act mainly to promote his/her own self-
interest. According to our conceptualization, the competitive and cooperative
orientations are relatively independent. In a given situation, one person can experience a
high level of cooperative orientation and a high level of competitive orientation. Another
person can be low on both dimensions. Another individual can experience a high level of
cooperative orientation but a low level of competitive orientation. Still, another person
can have a low level of cooperative orientation and a high level of competitive
orientation. Consistent with theory of cooperation and competition, we predicted that

Hypothesis 1: Positive reward interdependence will have a positive effect on

cooperative orientation.

Hypothesis 2: Positive reward interdependence will have a negative effect on

competitive orientation.
lgafmwoi and Taskwork Effort

According to Bowers, Braun, and Morgan (1997) and Morgan, Salas, and
Glickrnan (1994), team behaviors can be classiﬁed in two mutually exclusive and
exhaustive categories of activities: teamwork and taskwork. Taskwork refers to a team

member’s activities directed toward the tasks and tools—the non-interpersonal

17

component of the task. It includes the behaviors traditionally associated with individual
task performance. However, in an interdependent team, members must work together to
perform the team task. Teamwork refers to interpersonal interactions among team
members that are necessary for team goal setting, planning, exchanging information,
coordinating actions, feedback, and work load balancing. Because of the interdependent
nature of team tasks, team members often juggle the demands of teamwork with the
needs to get their own task work done.

filamworlgiffort

Many taxonomies of teamwork have been proposed (Fleishman & Zaccaro, 1992;
Hackman, 1987; Hackman & Morris, 1975; Marks, Mathieu, & Zaccaro, 2001; Prince &
Salas, 1993; Weldon, Jehn, & Pradhan, 1991). For example, Hackman (1987) suggested
six general categories of teamwork behaviors of an effective team. According to this
classiﬁcation, effective teams (1) minimize losses due to coordination and motivation (2)
create a sense of team spirit, (3) consider the competency of individual team members,
(4) encourage individual members to learn from one another, (5) maximize the efﬁciency
of performance strategies, and (6) foster creative planning. Although these behavioral
categories are descriptive, they do not provide detailed information concerning teamwork
activities (Bowers et al., 1997). Based on their work in the TADMU S project, Smith-
J entsch, Johnston, and Payne (1997) identiﬁed four dimensions of teamwork behaviors:
information exchange, communication, supporting behavior, and team initiative. They
further speciﬁed dimensions at a more speciﬁc level of detail. For example, the
information exchange dimension includes three speciﬁc behaviors: (1) seeking

information from all available sources (2) passing information to the appropriate persons

18

before being asked and (3) providing overall situation updates. Although different
teamwork taxonomies vary in content, speciﬁcity, and depth, they include behaviors and
groups of behaviors that fall in three broad categories: coordination, cooperation, and
communication (Kozlowski & Bell, 2003; Klimoski and Mohammed, 1994).

Empirical evidence shows that teamwork dimensions used in the team literature
are highly correlated. For example, Marks, Sabella, Burke, and Zaccaro (2002) examined
two important teamwork processes, coordination and backup behaviors, in a lab study.
The correlation of these two dimensions was .65. In another lab study, Mathieu, Heffner,
Goodwin, Salas, and Cannon-Bowers (2000) measured three dimensions of team process:
coordination, cooperation, and communication. The mean correlations of the three
dimensions over three measurement times were .49, .71, and .59, respectively. Factor
analyses of the teamwork data in this study consistently showed the existence of a single
underlying dimension for each of three measurements. As a result, the three dimensions
were combined to form a single team process variable. Barrick, Stewart, Neubert, and
Mount (1998) found similar results in a ﬁeld study. The correlation of communication
and team conﬂict was -.89. The correlation between communication and workload
sharing was .67, and the correlation between workload sharing and team conﬂict was .78.

Because teamwork dimensions are highly correlated, we chose to treat teamwork
as a unidimensional construct. Our purpose is to examine how different reward systems
affect teamwork process as a whole. This treatment, we believe, is justiﬁed in light of the
evidence of the high correlations of different teamwork dimensions.

Teamwork is often treated as a team-level construct (Ilgen & Sheppard, 2000). It

emerges as a result of team members’ interactions. The elements that give rise to

19

teamwork are team members’ teamwork behaviors. Therefore, teamwork can exist at
both team and individual levels. In this study, we deﬁne teamwork effort as the
evaluation of the intensity of teamwork behaviors of a team against objective or
subjective standards. Similarly, individual teamwork effort is the judgment of the
intensity of a team member’s teamwork behaviors against objective or subjective
standards. Given the task we considered in this work, the team behaviors include the
sharing of materials, information, and/or expertise by team members. The standard refers
to the amount of material, information, and/or skills that members share. Although this
variable can be considered as a team outcome variable, we treated it as process variable
in this study because the acts of sharing material, information, and expertise contribute to
team performance.

Taskwork Effort

 

For an interdependent task, a team needs both teamwork and taskwork to
accomplish its task (Bowers et al., 1997; Morgan et a1, 1994). Although taskwork is
critical for task performance, it has been largely neglected in team research and especially
in team reward literature. From a motivational perspective, taskwork can be
distinguished conceptually by what the team chooses to do, how hard it works, and how
long it persists. These three dimensions are parallel to direction, intensity, and
persistence components of motivational outcomes (Kanfer, 1992). The direction
dimension of taskwork refers to the types of task activities in which team members
engage. The intensity dimension denotes how hard a team works at taskwork activities.
The persistence dimension reﬂects a pattern of behaviors that a team exhibits over a long

period of time (Kanfer, 1992). Because the intensity dimension refers to how hard a team

20

works and is likely be highly inﬂuenced by reward structures, it is the taskwork
dimension of interest in our study. The direction and persistence of taskwork behaviors
will be controlled in our study as teams are required to work for a prescribed period of
time and their behavioral options are restricted.

We deﬁned taskwork effort (the intensity dimension) as the evaluation of the
intensity of task-relevant behaviors. It is typically referred as on-task behaviors and
includes behaviors such as the usage of material, knowledge, and skill to work on the
task. It does not include off-task behavior such as daydreaming or irrelevant web surﬁng.
The standard of evaluation refers to the frequency or amount of such behaviors for a
speciﬁc period of time. Taskwork effort is different from task performance, which is
often used a proxy measure of task effort in many studies (Kanfer, 1992). Task
performance is a measure of successful taskwork behaviors, whereas taskwork effort is a
measure of task-relevant behaviors, which include both successful and failing attempts to
do the tasks. Similar to teamwork, taskwork is often treated as a team-level construct
(Ilgen & Sheppard, 2000). Taskwork effort at the team level emerges as a result of team
members’ task effort. Therefore, taskwork effort also exists at the individual level and
can be deﬁned as the evaluation of the intensity of a member’s task-relevant behaviors
according to a subjective or objective standard.

Determinagts of Individual Teamworlgand Tas_kworl_<

Theory of cooperation and competition (Deutsch, 1949; 1985) predicts positive
reward interdependence leads to cooperative orientation and better teamwork processes
than negative reward interdependence. Research evidence has largely supported these

predictions. Johnson and Johnson (1989) reviewed hundreds of studies in a meta-

21

analysis. They found that positive reward interdependence led to more effective
teamwork processes. The effect was attributed to the cooperative orientation of team
members. It should be noted that most of team reward research examined individual-
level data or aggregated data. They should be therefore considered as empirical evidence
to evaluate individual-level relationships. Within the narrow range of positive reward
interdependence, Rosenbaum et al. (1980) found that reward distributions that were
approximately equal among team members led to more effective teamwork processes
than different levels of mixed reward distributions. Therefore, we expected to replicate
that cooperative orientation had a positive effect on individual teamwork effort and that
positive reward interdependence had a positive effect on individual teamwork effort.
With a lower level of positive reward interdependence, reward distribution is
likely more differential. Some team members receive higher levels of reward, whereas
others get lower levels of reward. A high level of reward in terms of money is likely
associated with higher valence than a low level of reward because of the instrumental and
symbolic values of money (Stajkovic & Luthans, 2001). Money is instrumental in
obtaining other goods and services that can fulﬁll physiological and psychological needs.
Symbolically, because it is often associated with status in organizations, it can generate
social comparisons about one’s relative social standing (Stajkovic & Luthans, 2001).
Thus, obtaining a high level of reward likely enhances the one’s self-interest, whereas
getting a low level of reward likely demotes one’s self-interest. In a performing team,
because more individual taskwork effort likely leads to better individual performance and

a higher level of associated reward, we expected that competitive orientation had a

22

positive effect on individual taskwork effort and positive reward interdependence had a
negative effect on individual taskwork effort.

Although these expectations are plausible, little research evidence exists to
evaluate their validities. With a few exceptions, team reward research has not
investigated the linkage of reward interdependence and taskwork effort for
interdependent tasks. Our review of the literature showed that only three studies have
examined the effect of reward interdependence on taskwork effort (Konovsky &
Podsakoff, 1993; Newcomb, Brady, & Hartup, 1979; Rosenbaum et a1, 1980). Using a
tower-building task in which team members worked together to build a single tower out
of wood blocks, Newcomb, Brady, and Hartup (1979) found that competitive reward
distribution, whether operationalized as proportional distribution or winner-take-all, was
associated with a higher number of handled blocks than equal reward structure among 88
two-person groups of preschool students. Similarly, Konovsky and Podsakoff (1993)
found proportional distribution of team reward resulted in a higher number of blocks
handled than equal distribution for 64 three-person teams of college students working on
the same tower-building task.

Also using the tower-building task, Rosenbaum et al.’s (1980) Study I examined
the performance of 36 three-person groups under equal reward structure and two
competitive reward structures. One of competitive reward condition was operationalized
as proportional distribution, and the other was operationalized as the distribution in which
the high performer of the team was rewarded proportional to his performance while the
other two team members were not rewarded. The results showed that both competitive

reward conditions resulted in more numbers of block handled than the equal reward

23

condition, but the effects were only marginally signiﬁcant (p < .08). The marginal
signiﬁcance of the effect happened probably because the study did not have enough
power due to the small number of teams in each reward condition.

Because the number of blocks handled in the tower-building task can be a
reasonable indicator of task effort, the results of the three studies (N ewcomb, Brady, &
Hartup, 1979; Rosenbaum et a1, 1980; Konovsky & Podsakoff, 1993) showed that
competitive reward leads to a higher level of task effort. However, the tower building
task used by all three studies has been criticized as biased towards encouraging taskwork
effort (Chertkoff & Mesch, 1997). In this task, the higher the tower, the higher the
probability that it will fall. As a result, low performing members under the competitive
reward structure can beneﬁt by adding more blocks. If the tower still stands, their
performance improves. If the tower falls, it is more beneﬁcial for them because all team
members have to start from scratch. Because of this bias, it is very plausible that team
members in the competitive reward structure are motivated to handle more blocks as the
tower falls more frequently. Team members in cooperative reward structure, on the other
hand, are motivated to keep their tower tall and therefore handle the blocks with care,
resulting in a smaller number of blocks handled.

Team reward research on independent tasks also suggests that competitive reward
distribution leads to higher level of taskwork effort than equal reward distribution.
Reviews of team reward studies by several researchers (Chertkoff & Mesch, 1997;
DeMatteo et al, 1998; Miller & Hamblin, 1969) showed that competitive reward
distribution has been found to result in more effective performance for independent tasks.

Because teamwork can be considered not important for this type of task, the superior

24

performance can reasonably be attributed to the higher level of taskwork effort resulted
from competitive reward system.

In sum, we made the following predictions:

Hypothesis 3: C00perative orientation will have a positive effect on individual

teamwork effort.

Hypothesis 4: Competitive orientation will have a positive effect on individual

taskwork effort.

Hypothesis 5: Positive reward interdependence will have a positive effect on

individual teamwork effort.

Hypothesis 6: Positive reward interdependence will have a negative effect on

individual taskwork effort.

At the individual level, positive reward interdependence was expected to have a
positive effect on individual teamwork effort. The effect could be explained by the
mediating effect of cooperative orientation. However, we did not attempt to predict
whether the mediation was full or partial. Similarly, our hypotheses suggested that
positive reward interdependence had a negative effect on individual taskwork effort. The
effect could be explained by the mediating effect of competitive orientation. However,
we did not attempt to make any prediction regarding the nature of the mediation.

In the model in Figure 1, task interdependence is construed as the degree to which
team members need to share materials, information, or expertise in order to achieve
valued group outcome. It is a structural feature of work that constrains the level of
behavioral interactions among team members. With high task interdependence, team

members need to depend more on one another for material, information, or expertise to

25

complete the team task. A high level of task interdependence demands team members to
engage in more teamwork behaviors. More social interactions likely reduce the amount
of time members can work on the task, likely leading to a lower level of individual
taskwork effort. Therefore, we expected that,

Hypothesis 7: Task interdependence will have a positive relationship with

individual teamwork effort.

Hypothesis 8: Task interdependence will have a negative relationship with

individual taskwork effort.

Determinants of Team-Level Teamwork Effort and Taikwork Effort

The model in Figure I predicted that higher individual-level teamwork effort and
taskwork effort give rise to higher team-level teamwork effort and taskwork effort. As
researchers become more aware of multilevel issues (Klein & Kozlowski, 2002), the
logics for how individual behaviors are combined to form team-level behaviors needs to
be explicated.

The team task in this study requires that a team member works on a relative
independent component for which he/she has the expertise. He/she can work
independently on the subcomponents for which he/she has the knowledge, but needs
assistance from other team members for the subcomponents he/she does not have the
expertise. When the minimum amount of individual reward is high enough to be
desirable, any level of positive reward interdependence likely elicits a reasonable
minimum amount of teamwork. Moreover, because the task components are relative
independent and team members have the expertise for their assigned components, the

lack of teamwork behaviors by one member is not likely to have the devastating effect on

26

team-level teamwork behavior. A high level of teamwork behavior by one team member
is not likely to have the overwhelming impact on the team-level teamwork behavior,
either. It is more likely that the lack of teamwork by one member is compensated by
more teamwork by others. The total amount of team member teamwork is likely what
determines team performance. Thus, we assumed that the sum of individual-level
teamwork effort captured the majority of the variance of team-level teamwork effort.
Because positive reward interdependence were hypothesized to have a positive effect on
individual teamwork effort, we expected that

Hypothesis 9: Positive reward interdependence will have a positive effect on

teamwork effort at the team level.

Similarly, when the minimum amount of individual reward is high enough to be
desirable, any level of positive reward interdependence likely elicits a reasonable
minimum amount of taskwork. Moreover, because the task components are relatively
independent and team members have expertise for their assigned components, low
taskwork effort by one team member unlikely have a catastrophic effect on the team-level
taskwork effort. A high level of effort by one team member will not likely have an
monumental effect on team-level taskwork effort, either. It is more likely that the lack of
effort by one team member is compensated by more effort exerted by other members.
The total amount of individual effort is likely what determines team performance. Thus,
we assumed that the sum of individual-level taskwork effort captured the majority of the
variance of team-level taskwork effort. Because positive reward interdependence were

hypothesized to have a negative effect on individual taskwork effort, we expected that

27

Hypothesis 10: Positive reward interdependence will have a negative effect on

taskwork effort at the team level.

For the present work, task interdependence is construed as the structural feature of
work that constrains team members’ interactions. High task interdependence requires
team members to interact more often to accomplish the task. More interactions, however,
likely lead to higher level of process loss (Steiner, 1975) and result in a lesser amount of
time for performing the task. Therefore, we expected

Hypothesis 11: Task interdependence will have a positive effect ’on teamwork

effort.

Hypothesis 12: Task interdependence will have a negative effect on taskwork

effort.

Effects of TeamworlLEffort acnd Tas_kwork Effort on Team Performm

The current team reward literature has largely focused on teamwork processes to
explain the differential effects of reward structures on team task performance, which we
deﬁned as the quantity of products that a team produces over a speciﬁc period of time
and meet certain subjective or objective quality criteria. Many team models (cf., Cohen &
Bailey, 1997; Gladstein, 1984; Guzzo & Dickson, 1996) proposed teamwork processes as
the main factors that mediate the relationship between reward structures (team input) and
team task performance (team output). Review after review of team reward research
(Deustch, 1985; Johnson & Johnson, 1989; Chertkoff & Mesch, 1997; DeMatteo et al.,
1997) showed that teamwork had positive effect on team performance. For example,
Johnson and Johnson (1989) concluded in a meta-analysis that cooperative reward system

was superior because it encouraged helping, information and resource exchange,

28

constructive feedbacks, constructive approach to conﬂict resolution, and mutually
inﬂuencing behaviors. Evidence from team training research also showed the same
pattern of relationships (cf., Cannon-Bowers & Salas, 1998; Marks et a1, 2002; Mathieu
et al., 2000).

Because taskwork is an important dimension of team behaviors (Bowers et al.,
1997; Jensth-Smith et al., 1998; Morgan et al., 1994), it is likely to have a signiﬁcant
effect on team task performance. In their models of team effectiveness, Hackman and
Morris (1975) and Hackman (1983) proposed that taskwork effort is positively related to
team performance. Although past team reward research did not directly examine the
relationship between taskwork effort and team task performance, research on team goal
setting showed that higher team goals led to increased taskwork effort, which explained
the higher level of team performance (Weingart & Weldon, 1991; Weingart, 1992;
Weldon et al., 1991). Therefore, we predicted that

Hypothesis l3: Teamwork effort will have a positive effect on team performance.

Hypothesis 14: Taskwork effort will have a positive effect on team performance.

Because task interdependence is a structural feature that speciﬁes the amount of
interactions needed to accomplish the team task, it likely moderates the effects of
teamwork and taskwork on team performance. Empirically, Wageman and Baker (1997)
found that task interdependence moderated the effect of reward interdependence on team
performance. Several reviews reported that equal reward distribution resulted in better
performance for interdependent task but inferior performance for independent tasks when
comparing with proportional reward distribution. These ﬁnding suggest that perhaps

team performance is more inﬂuenced by teamwork effort and less affected by taskwork

29

effort for high interdependent task. For low interdependent task, probably taskwork has a
stronger effect and teamwork is less inﬂuential. Therefore, we expected that
Hypothesis 15: Task interdependence will moderate the relationship between
teamwork effort and team performance.
Hypothesis 16: Task interdependence will moderate the relationship between

taskwork effort and team performance.

30

METHODS

Participants were 111 male (42%) and 129 female (58%) undergraduate students
enrolled in an introductory psychology courses at a large Mid-westem university. The
ethnic makeup of the sample was: 76% Caucasian American, 10% African American,
4% Hispanic American, 1% Indian American, and 9% Asian American. Their ages were
from 18 to 24 years (M =19.5, SD =1.4). The subjects selected the study to fulﬁll their
research participation requirements and were randomly assigned to 3—person teams. The
teams were randomly assigned to four different experimental conditions (2 reward
interdependence X 2 task interdependence). Two teams were eliminated from the study
because a member of one team left the experiment due to illness and the experimenter
failed to record the other team’s work properly, leaving a total of 80 teams in the study.
ﬂaii'ng of Team Members

Before working as a team, team members went through training to develop the
necessary skills and knowledge for the team task. During the training, each team
member leamed how to program in HTML a unique, restricted set of symbols.
Speciﬁcally, the ﬁrst team member learned how to code all the Greek letters and became
“Greek Letter Specialist”. The second team member was trained to know how to program
all the Latin letters and became “Latin Letter Specialist”. The third team member learned
how to write all the special signs and became “Sign Specialist.”

The team member learned his/her part by following a training tutorial (See
Appendix C) and an accompanying coding manual (See Appendix D) on the computer

screen. The training tutorial provided step-by-step instructions on how to write the

31

underlying codes for the symbols in a text editor (Notepad) and then display the symbols
on Internet Explorer. The coding manual contained a list of codes for 50 different
symbols. To program a symbol, the team member would type in a unique set of ﬁve
keystrokes including #, &, and three numbers. Team members were told that they did
not have to memorize the speciﬁc codes for the symbols and that they could refer to the
coding manual at any time during both the training and the subsequent team task
sessions. A trainer was available to answer any questions that team members had during
the training. Team members were allowed up to 20 minutes for the training, but most
finished it within 15 minutes.

Ta_sk

The task required a 3-person team to ﬁx 132 errors on a 3-page website. Each
member of the team was assigned the responsibility of ﬁxing one of the three pages for
which he/she had the expertise. The underlying codes for the page were contained in an
HTML ﬁle that the team member had opened on his/her computer screen prior to the start
of team task (See Figure 2 for a portion of the ﬁrst page).

Similar to Figure 2, each of the three web pages had the same structure and
contained 44 errors. In the page, an error appeared as a pair of question marks (??). The
text preceding the question marks described which speciﬁc symbols should be
programmed. For example, one of the errors was “Greek small zeta and Greek small
alpha: ??” To ﬁx it, the team member would need to replace the two question marks with
the codes “&#950 &#945” in the HTML ﬁle, which would be displayed as “C a” on

Internet Explorer.

32

 

 

<html>

<head><title>This page is authored by the Greek Letter
Specialist<ltitle><lhead>

<body>

This page is authored by the Greek Letter Specialist.
<br>

<br>

Two symbols can be combined to form a unit. Below are 44 units:
<br>

<br>

greek small kappa and capital U circumﬂex: ??
<br>

greek capital tau and inverted exclamation sign: ??
<br>

greek capital delta and greek small zeta: ??

<br>

greek capital beta and greek capital zeta: ??

<br>

greek small omicron and capital A grave: ??

<br>

<br>

greek small zeta and greek capital kappa: ??

<br>

greek capital alpha and regular currency sign: ??
<br>

greek small rho and small u grave: ??

<br>

oooooooooooooooooooooooooooooooooooooooooooooooooooooooooo

</body>
</html>

 

Figure 2: Portion of the Web Page Assigned to the Greek Letter Specialist

The page was designed to have a mix of Greek letters, Latin letters, and special
signs. As a result, the team member could ﬁx some of the errors independently using the
codes he/she had in the coding manual. However, the member needed to rely on the
knowledge of other team members as well as his/her own to ﬁx the rest. For example,
Greek Letter Specialist could ﬁx all the errors that contained only Greek letter symbols.
He/she needed to ask Latin Letter Specialist for the Latin letter codes to ﬁx errors that

contained Latin letter symbols and Sign Specialist for special sign codes to ﬁx errors that

33

 

had special sign symbols. The maximum amount of errors that the team member could
ﬁx independently varied depending on the task interdependence condition in which
his/her team was assigned.

It should be noted team members could not copy and paste the entire codes for a
symbol because the coding manual was set up as a read-only acrobat ﬁle. The
experimenter explicitly told a few subjects who attempted to use the Acrobat Find
function to locate the symbols in the coding manual that the search ﬁmction was not
permitted. These controlling actions ensured that each symbol would take approximately
the same amount of effort to program by (1) visually searching the coding manual or
asking other team members for the correct codes and (2) typing them in manually.

For this task, each member of the team was assigned the responsibility of ﬁxing
one of the three pages for which he/she had the expertise. They were told that they
needed to work together as a team to ﬁx the errors in their web pages. They might not
know how to ﬁx all the errors on their pages. For the errors they did not know how to
ﬁx, one of their team members had the knowledge. If they wanted to ﬁx all the errors in
their pages, they needed to work with one another. However, team members could work
in any way they wanted, except they were not allowed to physically type the codes for
their teammates. Team members had a total of 30 minutes to work on the task (Forty-
four errors were the maximum amount of errors that a team member could ﬁx in 30
minutes in a pilot study) with a short break after the ﬁrst 15 minutes during which they
were given feedbacks of their individual and team performance.

Team members worked in a common room, and each used a personal computer.

The computer stations were set up so that the team members were in close proximity and

34

could easily talk to one another face to face. However, they were away far enough so
that they could not see one another’s work from their own workstations. The computers
were not connected when team members worked on the task. As a result, team members
could not share knowledge by emailing. Because of these restrictions, all team members
shared their knowledge with one another verbally.
TﬁkLInterdependence

The task was designed so that in the low task interdependence condition, 22 errors
(50%) in the web page could be corrected by the team member’s own knowledge. The
team member needed to rely on one teammate to ﬁx the 11 errors (25%) and another
teammate to ﬁx the other 11 errors (25%). In the high task interdependence condition,
the subject could ﬁx 14 errors (33%) independently, and he/she needed to the assistance
of one teammate to ﬁx 15 errors (33%) and the other teammate to ﬁx the other 15 errors
(33%).
Reward Structure Manipulations

In the present study, a team would win a cash bonus of $90 if the team’s
performance was in the top 25 percent of all performing teams. In other words, a team
had a 25% chance of getting the team reward. Two reward conditions were examined. In
the high positive reward interdependence condition, team members were told that the
team reward would be distributed equally among team members. Each team member
would get $30 if the team won the bonus. In the low positive team reward
interdependence, team members were told that the team reward would be divided
differentially. The top performer of the team would receive $40; the middle performer,

$30; and the bottom performer, $20.

35

Manipulation Checﬁ

Team members were asked to respond to ﬁve questions (See Appendix H) by
rating on a scale from 1 (strongly disagree) to 5 (strongly agree). The ﬁrst question
referred to their knowledge of what constituted a correct ﬁx. The second and third
questions regarded the team members’ understanding of how individual and team
performance were assessed. The fourth and ﬁfth questions asked the subjects how teams
and individuals would be rewarded.

Procedure

After being greeted and signing voluntary participation agreements, the subjects
answered a questionnaire (See Appendix G) regarding their demographics (gender, age,
ethnicity, and GPA). They then learned how to build simple web pages. Once they
ﬁnished the training phase, the experimenter asked team members to gather in the center
of the room and introduce themselves. To help the team members to become more
comfortable with one another, team members were encouraged to chat about their major,
classes, and college experiences for ﬁve minutes.

After their conversations, the team members were asked to work on a trial team
task. The purpose of the trial task was to allow the team members to become familiar
with the team task. The trial task was the same as the team task with two exceptions.
The ﬁrst is that each web page of the website had only 10 errors instead of 44. The
second is that 4 of the errors could be ﬁxed independently, and 6 errors required other
members’ assistance. For the trial task, team members were told to work together to ﬁx
all the errors in the web pages as a team. The experimenters introduced to the team

members individually as having the expertise in different areas and assigned them to

36

work on the page for which they were the experts. They were told that they might not
know how to ﬁx all the errors on their pages. For the errors they did not know how to
ﬁx, one of their team members had the knowledge. If they wanted to ﬁx all the errors in
their pages, they need to work with one another. However, team members were then told
that they could work in any way they wanted, except they were not allowed to physically
type the codes for their teammates. Though there was no time restriction for this trial
task, most team members took less than 15 minutes to complete their assignments.

Once the subjects ﬁnished the trial task, they were then asked to read the
instruction on how to work on the next team task (See Appendix B). After they ﬁnished
reading, the experimenter showed the full website that the team members needed to ﬁx
and repeated the same instruction verbally. The instructions described the team tasks,
team members’ roles, what was considered as a correct ﬁx, the measurements of
individual and team performance, and the contingencies of performance and reward at the
team and individual levels. It also explained that team members needed to work together
as a team to ﬁx the errors in their web pages.

Subjects were then asked if they understood how they would be rewarded. The
second survey (See Appendix H) was then administered to measure the subjects’
understanding of the reward contingencies and their cooperative and competitive
orientations. Following the survey, team members worked on the team task. Team
members had a total of 30 minutes to work on the task. After the ﬁrst 15 minutes, they
were asked to stop their work so that the experimenters could upload and assemble the
pages together. Once the pages were uploaded, team members could see the outputs of

other members and the entire team. The experimenter counted the number of correct ﬁxes

37

by individual members and the entire team and reported them to the team verbally. Team
members then had another 15 minutes to complete the task.
Measures

Cooperative and Competitive Orientat_ign_s

The cooperative and competitive orientation measures, adapted from Johnson &
Norem-Hebeisen (1979), were used to assess individual cooperative and competitive
orientations. Because the original measure was developed to gauge students’ social
orientation in the classroom, references to the class, students, and learning were changed
to the team, team members, and performance, respectively. The measure contained 15
items (See Appendix H), and each item was rated on a scale from 1 to 7. Seven items in
the measure tapped cooperative orientation. An example item was, “It is a good idea for
team members to help one another perform.” The other eight items were assumed to
measure competitive orientation. An example item was, “I like to compete with other
team members to see who can do the best work.” The higher the score on each scale, the
more cooperative or competitive the respondents considered themselves to be. Alphas
were high in previous studies, ranging from .84 to .94 for cooperative scale and from .85
to .88 for competitive scale. Factor analysis of the data in the present study yielded a
two-factor solution. Item 1 of the competitive orientation scale, however, had high
loadings (larger than .4) on both factors and was therefore eliminated. The ﬁnal scales
contained 7-item cooperative orientation scale and 7-item competitive orientation scale.
The alpha coefﬁcient for cooperative orientation was .92, and the alpha coefﬁcient for

competitive orientation was .86.

38

Individual-Level Task Effort

Individual—level taskwork effort was measured objectively. The number of
symbols, not errors, a subject attempted to ﬁx in her assigned page was used as the
indicator of individual-level taskwork effort. The symbols were the building blocks of
the task, and each required 5 keystrokes (#, &, and 3 numerical digits) to program.
Because the # and & were identical for all symbols and could be COpied and pasted, we
decided to count the symbol as a unit of taskwork effort instead of the keystroke itself. It
turned out that only four subjects attempted to copy and pasted # and & in their work, the
measure of effort in terms of symbols was considered an equivalent measure of
keystrokes.

IndividuiLevel Teamwork Effort

Individual-level teamwork effort was also measured objectively. The piece of
information (the ﬁve-di git code for a symbol) each team member shared with the rest of
the team was recorded and assessed for its accuracy by the experimenter, who used an
observation check list (See Appendix F). Because the amount of shared inaccurate
information was very close to zero (only one subject shared two pieces of inaccurate
information), we used the amount of shared accurate information as indicator of
teamwork effort. The correlation between this measure and the measure of shared
information obtained by examining the objective outputs was very high, at .95.

Eamme Effort

Team-level teamwork effort was measured objectively. The total number of
pieces of information (the ﬁve-digit code for a symbol) team members shared with one

another was recorded by the experimenter and used as the indicator of teamwork effort.

39

The correlation between this measure and the measure of shared information obtained by
examining the team’s ﬁnal outputs was very high, at .96.

Taskwork Effort

Team-level taskwork effort was measured objectively. A rater examined the team
ﬁnal output for the number of symbols the team attempted to ﬁx in the website. The total
number of attempts was the indicator of team—level taskwork effort.

Team Performance

Team performance was measured objectively. A rater examined each team ’8
ﬁnal output for the number of errors that have been ﬁxed. The total number of errors the

entire team ﬁxed was the measure of team performance.

40

RESULTS

Manipulation checks showed that subjects clearly understood how their
performance was measured and the contingency between performance and reward. On
the scale from 1 (strongly disagree) to 5 (strongly agree), the means of the subjects’
understanding of what constituted a correct ﬁx was 4.84 (SD = .37). The means of their
understanding of how team and individual performances were evaluated was 4.89 (SD =
.31) and 4.83 (SD = .38), respectively. The means of their understanding of how team
and individual were rewarded was 4.95 (SD = .21) and 4.92 (SD = .26), respectively.
Aﬂrlysis Stgrtegies

Because the model in Figure 1 described the relationships among the constructs at
both the team and individual levels, two different approaches were used to analyze the
data. HLM was used to analyze the effects of reward and task interdependence on
individual-level endogenous variables (individual-level analyses). Multiple regressions
were applied to assess the effects of reward and task interdependence on team-level
endogenous variables (team-level analyses).

Individual-Level Analyses

The means, standard deviation, ranges, and zero-ordered correlations for all
variables for individual-level analyses were reported on Table 1. The relationships of
reward interdependence with cooperation orientations and competitive orientations were
both signiﬁcant with respective correlation coefﬁcients of 0.16 and —0. 14. Hypotheses 1

and 2 were therefore supported.

41

Sol... modva.

 

 

 

 

o>==cano c>=muoaooo

 

 

Iumdé tnmd :26 .w r .o .m to- ta F .o- 31: RN wKN cocmEuotcn.
:Cd :vmd 3.0 :«N? .m ﬁo - wQAN 8!? an. 5 x8223...
mod :36 towd : Fwd AWN vwd vmdv {03:58.
vod .36. mod- 3.3 was mme 5.6395 02582.00
.96 mod- mvﬂm 9d Qwv co_.m.ccto 9:92.80
86 Epsom
xwm...
, €022me xuoBEmoh acumEcto coszcto Emsom xwmh 3ch am zoos. m._m<.m<

 

 

 

. 2e;

42

At the individual level (See Figure 3), the mediation model of the effects of reward and
interdependence on teamwork effort via cooperative orientation was tested ﬁrst. The
same set of tests was repeated to examine the effect of reward interdependence on the

second dependent variable, taskwork effort.

 

 

5 Individual

 

 

 

 

 

 

 

 

 

 

+ H5 , Teamwork
Cooperatrve Effort
Orientation _—P
/ +H3 I... I'll-III..-
:IIIIIIIIIIIIIIII 5 Task
: Reward : : Interdependence
: Interdependence ' :
'I-IIIIIIIIIIIIIIU. .". ".....'...
\ C t't' + H4
om e 1 we
Orieritation —-> Individual
_ H6 Taskwork

 

 

 

 

Effort

 

 

 

 

Figure 3: Individual-Level Model

Because individual-level teamwork effort and taskwork effort are nested within
teams, the appropriate statistical technique is multilevel random coefﬁcient analysis
(Bryk & Raudenbush, 1992). Multilevel random coefﬁcient analysis (Bryk &
Raudenbush, 1992) can be considered a simultaneous two-stage procedure. In Stage 1,
intercepts and slopes are estimated for each team at the individual level. The overall
statistical signiﬁcances of individual-level relationships are considered “pooled or
averaged” parameter estimates from all teams.

In Stage 2, Stage 1 slopes and intercepts are used as the criteria to estimate the
effects of team-level variables. If a team-level variable (reward interdependence) is

related to the variability of the intercept, it is inferred to have a main effect on the

43

dependent variable. If a team-level variable is related to the variability of the slope, it is
considered to have a moderating effect.

Five random coefﬁcient models were estimated to test the hypotheses regarding
each dependent variable. The ﬁrst model (AN OVA Model) was a null model that had no
predictors at either level. It was used to partition the variance in each dependent variable
into between-team and within-team components. The proportion of between-group
variance to the total variance was calculated and used as an intraclass correlation as in a
one-way analysis of variance (Bryk & Raudenbush, 1992).

The second model (random coefﬁcient model) added the mediator (e.g.,
cooperative orientation) as an individual-level predictor, but it does not have any the
team-level predictor. The pooled value of the within-team parameter estimates provided
the assessment of the averaged effect of the mediator on the dependable variable at the
individual level across teams. The signiﬁcant test of the parameter indicated that the
mediator had an effect on the dependent variable. The effect size was determined by
evaluating the differences between the variance components of this model and those of
the null model.

The third model (intercept-as-outcomes model with task interdependence as the
team-level predictor) included task interdependence as the team-level predictor (e. g.,
cooperative orientation) but did not have the individual-level predictor. The between-
team parameter estimate provided as assessment of the effect of the task interdependence
on the team-means of the dependent variable. The signiﬁcant test of the parameter

indicated that task interdependence had an effect on the dependent variable. The effect

44

size was determined by evaluating the differences between the between-team variance of
this model and that of the null model.

The fourth model (intercept-as-outcomes model with task and reward
interdependence as the team-level predictors) added reward interdependence to the third
model as a team-level predictor. The team-level parameter estimate provided the
assessment of the independent effect of the reward interdependence on the team-means of
the dependent variable. The signiﬁcant test of the parameter estimate indicated that
reward interdependence had incremental effect on the dependent variable, over and above
that of the task interdependence. The incremental effect size of reward interdependence
was determined by evaluating the differences between the between-team variance of this
model and that of the third model.

The ﬁfth model (intercepts-as-outcomes model with social orientation as the
individual-level predictor and task interdependence as the team-level predictor) added the
task interdependence as a team-level predictor to the second model. It were used a
compared model to examine the direct effect of reward interdependence in the sixth
model.

The sixth model (intercepts-as-outcomes model with social orientation as the
individual-level predictor, task and reward interdependence as the team-level predictors)
added reward interdependence as individual-level predictors to the ﬁfth model. The
parameter estimate of reward interdependence and the deviation score provided the
assessments of the reward’s direct effect when the effects of task interdependence and

social orientation were controlled.

45

The HLM analyses of antecedents of teamwork behaviors were shown in Table 2.
In Model 1 (ANOVA model with no predictor), the level-2 intercept of level-1 intercepts
was signiﬁcantly different from zero (G00 = 12.94, t= 26, p < .01), meaning that the
grand mean of teamwork effort across teams was 12.94. The variance of level-1
intercepts (between-team variance) was 16.3 and signiﬁcantly different from zero (U 0 =
16.3, X2(79) = 421.27, p < .01), showing that teamwork effort varied between teams.
The within-team variance was 11.28. Accordingly, ICC was .59 (16.3 /{16.3+11.28}),
indicating between-team variance made up 58% of total teamwork variance.

In Model 2 (random coefﬁcient model with cooperative orientation as the
individual-level predictor), the level-2 intercept of level-1 slopes (the pool slope) was
signiﬁcantly different from zero (G10 = .32, t = 2.31, p < .01), indicating that cooperative
orientation, on average, likely led to higher level of teamwork effort. The change in
deviation score was 13.39 (1393.39 — 1380) and signiﬁcant, indicating model 2 ﬁts the
data better than model 1. As a result, hypothesis 3 was therefore supported. The within-
team residual variance was reduced to 8.7. The between-team variance was decreased to
15.02. Therefore, the cooperative orientation accounted for .20 ((11.28 — 9.07}/11.28) of
within-team teamwork variance, .08 ({16.3 — 15.02}/16.3) of between-team teamwork

variance, and .12 (.20*.41 + .08*.59) of total teamwork variance.

46

 

 

80.0.8... E57800. 05 an 85230925 9030. new
0.08 9.0 .2060... E>0_-_m:n_>_nc_ 05 mm conﬂccto 02.92.08 5.3 Enos. moE00.=0-mm-...ooEE. no Enos.
8.0.8:. E>o_-Emo. 8 8505820....
in. new 5.0605 E>o_-_m=n_>_nc_ 9.. m0 cozmﬁotc 02.99.08 2...... Enos. mcho.:0.mm-...00E.c_ ”m Enos.
90.0.3.0 E>c_-Emm. 05 mm eccenccacnuoE. Ease. ncm xmm. 5.3 Enos. mcho.:0.wm-.ac0.EE_ ”v Enos.

5.0.8:.

E>0_-Emc. 9.. we woconceaanE. .30. 5.3 Enos. mcE00.:0.wm-...ooE.c_ ”m Enos.
8.0.00.0 E>c_-_m:n_>_nc_ 0... mm cozﬂccnc 02.0.8000 5s) Enos. Eotm. Eoncmm UN Enos.

Enos. <>OZ< ”F .085.
5. v a: .8. v a .

.mcwcchEa E... c. 5.6% 90 wosE> .. $5.058“. mEn no~_n.mnc0.m§ 20 man> _.<

 

     
 

 

 

 

 

 

 

 

5.0 58E
8... 00.
A m w v m m EEEEmn. nm.0E_.mm.
8.58 mm. :2 8.22 $.82 89 8.82 28m 8:950
k 2. k 2. HI... 8. . F B... a m .26.. .8222.
2 2 2 .u
8.9. m3: 3...: scenario
twmﬁo .Lmd fund 5 .305 coszoto 9:89.000
t 8. R E 2. 2 .u
9.03 2.25 3. In 8.48 «3% R. 5.. 988-30
2.3.8 2.3.2 :8: row? :89 :2: o: .. 38.25
«octw Eon—.3.
Aomdrmme ENNVEQN 8530822... Ease”.
3.8.63 swarmed sesame remind 85888.25 07.3
Cm.m...mm.o Anvdvtnmd 38.8: med 8.55:0 9:92.000
swerve... 893.;on .mv.m.tmm.m Esra... .E.8...8.~. an. :39 E928
poem 85".
o .25: Elena... e .28: n .25: «as: F aulowi some;

 

 

 

 

 

rem {asemﬂ _m>3-_m8_>_uc_ 8 985 .0 938m 5::

”N 293

47

It should be noted that the variance of level-1 intercepts was signiﬁcantly different from
zero (U0 = 15.31, X2 (79) = 368.39, p < .01), showing that teamwork effort was still
different between teams after cooperative orientation was controlled. The variance of
level-1 slopes of cooperative orientation was also signiﬁcantly different from zero (U 1 =
.23, X2 (79) = 118.36, p < .01), indicating that the effect of cooperative orientation on
teamwork effort varied between teams.

In Model 3 (means-as-outcomes model with task interdependence as the team-
level predictor), the level-2 slope of task interdependence that predicted level-l intercepts
was signiﬁcantly different from zero (G01 = 2.24, t = 2.24, p < .05), indicating that task
interdependence had an effect of team-level teamwork effort. The change in deviation
score was 5.13 (1393.39 —— 1388.26) and signiﬁcant, indicating model 3 ﬁts the data better
than model 1. As a result, hypothesis 7 was supported. The variance of level-1 intercepts
was decreased to 15.26. Therefore, task interdependence accounted for .06 ({16.3 -
15.26}/16.3) of between-team teamwork variance and .04 (.06*.59 + 0*.41) of total
teamwork variance.

In Model 4 (means-as-outcomes model with both task and reward
interdependence as the team—level predictors), the level-2 slope of task interdependence
that predicted level-1 intercepts was still signiﬁcantly different from zero (G01 = 2.24, t
= 2.39, p < .05). The level-2 slope of reward interdependence that predicted level-1
intercepts was also signiﬁcantly different from zero (G02 = 2.08, t = 2.23, p < .05). This
indicated that reward interdependence had a positive effect on teamwork effort after the
effect of task interdependence was controlled. The change in deviation score was 8.27

(1388.26 — 1379.99) and signiﬁcant, indicating model 4 ﬁts the data better than model 3.

48

Hypothesis 5 was therefore supported. The variance of level-1 intercept was reduced to
14.38. Therefore, task and reward interdependence accounted for .12 ({16.3 —-

14.38}/ 16.3) of between-teamwork variance or .07 of total teamwork variance (.12*.59 +
0*.41). In other words, reward interdependence explained an additional .06 of between-
team teamwork variance and .03 of total teamwork variance over and above the effect of
task interdependence.

In Model 6 (intercepts-as-outcomes model with cooperative orientation as the
individual-level predictor and task and reward interdependence as the team-level
predictors) the level—2 slope of reward interdependence that predicted level-1 intercepts
was signiﬁcantly different from zero (G02 = 1.99, t = 2.20, p < .05). The change in
deviation score was 3.99 (1371.24 — 1367.33) and signiﬁcant, indicating model 6 ﬁts the
data better than model 5. This indicated that reward interdependence had an incremental
effect on teamwork effort over and above the effects of task interdependence and
cooperative orientation. Thus, the effect of reward interdependence on individual
teamwork effort is partially mediated by cooperative orientation.

The HLM analyses of antecedents of taskwork behaviors were shown in Table 3.
In Model 1 (ANOVA model with no predictor), the level-2 intercept of level-1 intercepts
was signiﬁcantly different from zero (G00 = 61.67, t = 58.5, p < .01), indicating that the
grand mean of taskwork effort across teams was 61.67. The variance of level-1 intercepts
(between-team variance) was 44.53 and signiﬁcantly different from zero (U0 = 44.53,
X2(79) = 155.73, p < .01), showing that taskwork effort varied between teams. The

within-team taskwork variance was 137.54. Accordingly, ICC was .245

49

(44.53/ {44.53+l37.54}), indicating between-team variance made up 24.5 % of total
taskwork effort variance.

In Model 2 (random coefﬁcient model with competitive orientation as the
individual-level predictor), the level-2 intercept of level-1 slopes (the pool slope) was
signiﬁcantly different from zero (G10=.42, t = 3.50, p < .01), indicating that competitive
orientation, on average, likely led to higher level of taskwork effort. The change in
deviation score was 10.40 (1914.03 — 1901.89) and statistically signiﬁcant, indicating
model 2 ﬁts the data better than model 1. As a result, hypothesis 4 was supported. The
within-team residual variance was reduced to 124.47. The between-team taskwork
variance was decreased to 38.73. Therefore, competitive orientation accounted for .10
((137.54 — l24.47}/137.54) ofwithin-team taskwork variance, .13 ({44.53 —
38.73}/44.53) of between-team taskwork variance, and .11 (.10*.74 + .13*.26) of total
taskwork variance. We noted that the variance of level-1 intercepts was signiﬁcantly
different from zero (U0 = 38.73, X2 (79) = 117.05, p < .01), indicating that taskwork
effort was still different between teams after competitive orientation was controlled. The
variance of level-1 slopes of competitive orientation was not signiﬁcantly different from
zero (U 1 = 0.18, X2 (79) = 85.86, p > .01), showing that the effect of competitive
orientation on taskwork effort did not vary between teams.

In Model 3 (means-as-outcomes model with task interdependence as the team-
level predictor), the level-2 slope of task interdependence that predicted level-l intercepts
was signiﬁcantly different from zero (G01 = -4.01, t = -1.94, p < .05), indicating that task

interdependence had a negative effect of team—level taskwork effort.

50

0.0.0605 E>0_-E00. 0.... 00 00:0nc000n.0.c.

E030. n00 0000200020.... x00. nc0 .2050... E>0_-_0=n_>_nc_ 0... 00 830.028 0>....00E8 5.3 Enos. 00885000....00EE. no Enos.
00.060... _0>0_-E00. 00 00c0nc0a0n.0.:_ x00. n00 .0.0_n0.a E>0_-_0...n.>_nc_ 00 :0..0.c0..0 0>....0aE8 5.3 Enos. 00E00....0-00-...00.0.c_ Um Enos.
0.0.060... E>0_-c..00. 0c. 00 00c0n:0a0n.0.c. E030. n00 0.00. ...s> Enos. m0E00.:0-w0-...00.0.£ .v Enos.

5.0.00.0 E>0.-E00. 0... 00 00830820.... 0.00. 5.3 Enos. 006005900200825 ”m Enos.
.9060... _0>0_-.0:n_>_nc_ 0:. 00 8.0.5.8 0>_._.00E8 5.3 Enos. .00tw Eoncmm .m Enos.
Enos. <>Oz< .. Enos.

8.. v & .c0oEcmE 2.00.900. m 5. v a... .8. v n. .

000055.00 0.... 0. E5020 0.0 00:_0> .- 0.8.0508 0.00 n0~_n.0nc0.mc: 0.0 m0:_0> __<

 

 

 

 

 

 

 

 

 

 

 

80.0 3:52.00
.20... 00.
k 0 m e m m 22:22. 8.05.7.0
5.8.: 0N8... $.82 Nun-8. 8. .8. 8.32 280 8:250
:8. 40.040. 8.5. 8.5. :10. 8.5. m .20. 08.22.
a. 2 a. .o
8.8 3.8 8.2. 22.8-30
02. 0:. two S 08.0 8.0.8.5 2.....8500
t E R w. a. a. .u
2.8. 8.0: 8.8. 2.03 3.: «.32 22.3-50
.8. 5 L . .8 :88 :5. 5. 5.2.8 :33. o: .. .822...
nowtw Eoucwm
.80-... R0- .03-ramm- 88280225 2033.
a... wrote- .4... 3.2.... .0o.~-.. 5.x .3. 1.5.? 8822.825 0.me
80.8.08... 5.8... 5.... Emerge... 8228.5 02.2880
Ceszmto m>_.m.®QooO
83:200.: 3.002.058 33:20.0. Smyrna... .00. 573.. .0 3.02.8. 5 E2250
.800 as...
Ion-mo: n .08... v .28.... a Blvd: N .23.... was; 032.;

 

 

 

22.0 c.8325 Eon-.2223 no 2800.0 2.33. 2.:
m 2%.

51

The change in deviation score was 6.81 (1914.03 — 1907.22) and signiﬁcant, indicating
model 3 ﬁts the data better than model 1. As a result, hypothesis 8 was supported. The
variance of level-1 intercepts was decreased to 41.57. Therefore, the effect size of task
interdependence on between-team teamwork variance was .07 ({44.53 — 41 .57}/44.53) of
between-team taskwork variance and .02 (.07*.26 + 0*.74) of total taskwork variance.

In Model 4 (means—as-outcomes model with both task and reward
interdependence as the team-level predictors), the level-2 slope of task interdependence
that predicts level-1 intercepts was still signiﬁcantly different from zero (G01 = -4.01, t =
-2.06, p < .05). The level-2 slope of reward interdependence that predicted level-1
intercepts was also signiﬁcant different from zero (G02 = -5.99, t = -3.06, p < .01). This
indicated that reward interdependence had a negative effect on taskwork effort after the
effect of task interdependence was controlled. The change in deviation score was 13.73
(1907.22 — 1893.49) and signiﬁcant, indicating model 4 ﬁts the data better than model 3.
Hypothesis 6 was therefore supported. The variance of level-1 intercept was reduced to
33.38. Therefore, task and reward interdependence accounted for .20 ( {41.57 —
33.38}/41.57) of between-team taskwork variance and .05 (.20*.25 + 0*75) of total
taskwork variance. In other words, reward interdependence explained an additional .13
of between-team taskwork variance and .03 of total taskwork variance over and above the
effect of task interdependence.

In Model 6 (intercepts—as-outcomes model with task and reward interdependence
as the team-level predictors and competitive orientation as the individual-level predictor)
the level-2 slope of reward interdependence that predicted level-1 intercepts was

signiﬁcantly different from zero (G02 = -5.21, t = -2.66, p < .01). The change in

52

deviation score was 7.45 (1894.26 — 1886.81) and signiﬁcant, indicating model 6 ﬁts the
data better than model 5. This indicated that reward interdependence had an incremental
effect on taskwork effort aﬁer the effects of task interdependence and competitive
orientation was controlled. Thus, the effect of reward interdependence on individual
teamwork effort was partially mediated by competitive orientation.

Team-Level Analyses

The means, standard deviation, ranges, and zero-ordered correlations for all
variables for team-level analyses were reported on Table 4.

Table 4
Summary Statistics for Team-Level Variables

 

[VARIABLE MEAM SD RANGE TASK REWARD TEAMWORKTASKWORK

 

ASK

[REWARD 0.00

TEAMWORK 38.96 13.65 12-75 0.23' 0.22“
TASKWORK 185.86 28.39 125-244 -O.25* -0.32“ 0.39“

TEAMPERF 83.43 14.42 56-115 -0.28* -0.24* 0.58“ 0.86“

 

 

 

 

* p < 0.05 ** p <0.01.

At the team level (See Figure 4), a series of multiple regressions were run to
assess the effects of task and reward interdependence on team performance via the
mediating effects of teamwork effort and taskwork effort. We tested the relationships in
the model through ﬁve sets of hierarchical regressions.

In Set 1, the effects of the predictor variables (task and reward interdependence)
on the criterion variable (team performance) were assessed. Hierarchical regressions were
used to evaluate the incremental effect of the predictors. Team performance was

regressed on task interdependence in model 1. Reward interdependence and the

53

interaction term of reward and task interdependence were then added in model 2 and 3,
respectively. R2 in model 1 provided an assessment of the task interdependence effect on
team performance. R2 difference between model 2 and 1 provided the incremental effect
of reward interdependence over and above that of the task interdependence. R2
difference between model 3 and 2 allowed the examination of the interaction effect of

task and reward interdependence.

+H9

 

 

Teamwork
Effort

+H13

      

 

   

 

 

 

 

 

 

 

H11+
H15
:IIIIIIIIIIIIIII: E Task 5 TearnTaSk
3 Reward 3 3 Interdependence 3 Performance
3 Interdependence 3 3 3
Illa-IIIIIIIIIIIII """"" """
H1 +H14
H12-
V
Taskwork
- H10 Effort

 

Figure 4: Team-Level Model

In set 2, the effects of the predictor variables (task and reward interdependence)
on each of the two mediating variables (teamwork effort and taskwork effort) were
assessed. Hierarchical regressions were used to evaluate the incremental effect of the
predictors. The mediating variable was regressed on task interdependence in model 1 and
then on both task and reward interdependence in model 2. R2 in model 1 provided an

assessment of task interdependence effect on team performance. R2 difference between

54

model 2 and 1 provided the incremental effect of reward interdependence over and above
that of task interdependence.

In Set 3, the effects of the mediators on team performance were assessed.
Hierarchical regressions were used to evaluate the incremental effects of the mediators.
Team performance was regressed on teamwork effort in model 1, on taskwork effort in
model 2, and then on both teamwork effort and taskwork effort in model 3. R2 difference
between model 3 and 2 provided the incremental effect of teamwork effort over and
above that of the taskwork effort. R2 difference between model 3 and 1 provided the
incremental effect of taskwork effort over and above that of the teamwork effort.

In Set 4, hierarchical regressions were used to assess the effects of task
interdependence on the relationships of team performance to each of the mediators.

Team performance was regressed on task interdependence and the mediator in model 1.
The interaction term of task interdependence and the mediator were added in model 2. R2
difference between model 2 and 1 provided the assessment of the incremental effect of

the interactions on team performance over and above the main effects.

 

 

 

 

Table 5
Effects of Taskand Reward Interdependence on Team Performance
Varlable (N=80) Model 1 Model 2 Model 3

ask Interdependence -8.12**(-2.68) -8.12**(-2.68) -16.90 (-1 .76)
eward Interdependence -6.77**(-2.23) -15.55 (-1.62)
ask X Reward 5.85 (.96)
odel R Square 0.08" 0.136" 0.147“
han e In R Square 0056* 0.011

 

 

 

All values are unstandardized beta coefﬁcients. T values are shown in the parentheses.
* p < 0.05 “ p < 0.01

55

In Set 1 (See Table 5), task interdependence had a negative effect on team
performance (B1: -8.12, t = -2.68, p < .001). High task interdependence led to lower
level of performance than low task interdependence. The effect size was .008. Reward
interdependence had a negative inﬂuence on task performance over and above task
interdependence (B2 = -6.77, t = —2.23, p < 05). When the effect of task interdependence
was controlled, high reward interdependence resulted in a lower level of performance
than low reward interdependence. The incremental effect size was 0.056 and signiﬁcantly
different from zero. The interaction effect of reward and task interdependence, however,
was not signiﬁcant (B3 = 5.85, t = .96, p > .05).

The Effects of Tas_k and Rem Interdependence on Teamwork_and Taskwork Effort

In Set 2 (See Table 6), when teamwork effort was the criterion, task
interdependence had a positive effect (B1 = 6.12, t = 2.45, p < .05). High task
interdependence led to a higher level of teamwork effort than low task interdependence.
Hypothesis 11 was supported. The effect size was .05. Reward interdependence had a
negative incremental effect over and above task interdependence (BZ = 6.13, t = 2.09, p <
.05). High reward interdependence resulted in higher level of teamwork effort than low
reward interdependence over and above the effect of task interdependence. Hypothesis 9
was supported. The incremental effect size was .05 and signiﬁcantly different from zero.
When both task interdependence and taskwork effort were partialed out, the effect of

reward interdependence was still signiﬁcant (B2 = 11.32, t = 4.53, p < .01).

56

 

5. v a: .8. v a .

68055.3 05 5 $505 0.6 moams .r .mEo_oEooo Son 88285.2: 2m wo:_m> =<
£395 “.0 you £5 5 28:9 “coucoaou 05 m_ o_nm:m> “consumes mm 8%: 0.8:? on... --

 

 

 

 

 

 

 

 

 

 

:8. .2. :Nlm. .8. a;
:8. zero :88 :03 .2. .8. 23cm a .822
-- -- -- 89:8. €258»
808 tom. F -- -- -- {oi—hue...
A2.m-v..8.vw- 83:0: 7 893:8. : 293.88 8568822... Essay.
8.1.8. a- A8.N-v.w.2- idebé- 5.3.8.9 89998 8.8.9.8 85.65828... xmﬂ
n .822 N .822 F .822 n Blue: N .23: r .285
{cap fozelaﬂ 8qu 28:;

 

 

 

 

 

 

fog—m3. USN 610388;. :0 3595 EB:— EmBoM Ea xmﬂ. mo ﬂoobm 2:.

o 2an

57

When taskwork effort was the criterion, task interdependence had a negative
effect (B1 = -13.8, t = -2.24, p < .05). High task interdependence led to a lower level of
taskwork effort than low task interdependence. Hypothesis 12 is supported. The effect
size is .06. Reward interdependence had a negative incremental effect over and above
task interdependence (BZ = -17.70, t = -3.01, p < .05). High reward interdependence
resulted in lower level of taskwork effort than low reward interdependence alter the
effect of task interdependence was controlled. Hypothesis 10 was supported. The
incremental effect size was .10 and signiﬁcantly different from zero. When both task
interdependence and taskwork effort were partialed out, the effect of reward
interdependence was still signiﬁcant (BZ = -24.93, t = -5.93, p < .

The Effects of Teamworit Effort and Taskwork Effort on Team PerformaM

 

 

 

 

 

 

Table 7
The Effects of Teamwork Effort and Taskwork Effort on Tegn
Performance

Variable (N=80) Model 1 Model 2 Model 3
Teamwork 0.62“ (6.37) 0.31” (5.34)
Taskwork M” (“-3) 0.38" (13.70)
erodeI R Square 0.34“ 0.74“ 0.81“

 

 

 

All values are unstandardized coefﬁcients. T values are shown in the parentheses.
* p < .05. “p < .01

In Set 3 (See Table 7), teamwork effort had a positive effect on team performance
(B1 = .62, t = 6.37, p < 01). Higher level of teamwork effort led to higher level of team
performance. The effect size was .34. Taskwork effort also had a positive effect on

team performance (B1 = .44, t = 14.8, p < 01). Higher level of taskwork effort led to

58

higher level of team performance. The effect size was .74. When team teamwork and
taskwork efforts were entered as predictors of team performance, they both had
independent effects. The effect size was .81. As a result, teamwork effort has
incremental positive effect over and above taskwork effort. The incremental effect size
was .07 and signiﬁcantly different from zero. Similarly, taskwork effort had a
incremental positive effect over and above teamwork effort. The incremental effect size
was .47 and signiﬁcantly different from zero. Thus, hypotheses l3 and 14 were
supported.

Moderating Effects of Task Interdependence on MediaQr-Performance Relationships

 

 

 

 

 

 

Table 8

Moderating Effects of Task Interdependence

I Variable (N=80) Team Performance Team Performance

Model 1 Model 2 Model 1 Model 2

ask Interdependence -12.55** (-5.42) -16.93 (-2.39) -2.22 (-1.30) 17.23 (1.52)
eward Interdependence
eamwork .72“ (8.47) .56* (.27)

Taskwork .43“ (14.09) .59“ (6.07)

Task X Teamwork .11 (.65)

Task X Taskwork -0.10" (-1.74)
odel R Square 0.52“ 0.527“ .743“ .753"

Ehange In R Square 0.003 0.018

 

 

 

 

All values are unstandardized beta coefﬁcients. T values are shown in the parentheses.
' p < 0.05; "p < .01; p < 0.01
a p = 0.087
In Set 4 (See Table 8), the effect of the interaction of task interdependence
and teamwork effort on team performance was not signiﬁcant (B3 = .11, t = .65, p =
.515). The effect of teamwork effort on team performance did not vary by the level
of task interdependence. Hypothesis 15 was not supported. The effect of the

interaction of task interdependence and taskwork effort on team performance was

59

marginally signiﬁcant (B3 = -.10, t = -1.74, p = .087). The effect of taskwork effort

on team performance was marginally different between tasks.

60

DISCUSSION
TheoretiMontﬁbutions

The goal of this study was to explicitly examine the process in which positive
reward interdependence affected team performance of interdependent tasks. As we
expected, at the individual level positive reward interdependence has a positive effect on
cooperative orientation and a negative effect on competitive orientation. When the effect
of task interdependence was controlled, positive reward interdependence had a positive
effect on teamwork effort and a negative effect on taskwork effort. The effects reward
interdependence on teamwork effort and taskwork effort could be explained in part
through its effects on cooperative and competitive orientations, respectively. At the
team-level, reward interdependence had a positive effect on teamwork effort and a
negative effect on taskwork effort when the effect of task interdependence was
controlled. Teamwork effort and taskwork effort in turn determined team performance.
The effect of reward interdependence on team performance could mostly be explained
through its effects on teamwork effort and taskwork effort.

The present study found that reward interdependence had a negative effect on
taskwork effort with an effect size of .16 after the effect of task interdependence is
controlled. Taskwork effort in turn had a positive inﬂuence on team performance over
and above teamwork effort, with the large incremental effect size of .47. This ﬁnding is
important because most studies in team reward literature have ignored the effect of team
reward on this inﬂuential variable. It is also consistent with a few studies (Newcomb,
Brady, & Hartup, 1979; Rosenbaum et a1, 1980; Konovsky & Podsakoff, 1993) that have

examined the effect of reward interdependence on taskwork effort. However, the present

61

study differs in two respects. One is that unlike the previous studies that examined the
effect of a broader range of reward interdependence (positive reward interdependence,
negative reward interdependence, and reward independence), the current study examined
a narrow range (high and low positive reward interdependence). The other is that the task
used in this study does address the bias inherent that the tower-building task.

Hypothesis 16 (Task interdependence moderates the effect of taskwork efforts on
team performance) was only marginally supported because the p value for the interaction
term was greater than .05 but smaller than .10. This marginal support occurred perhaps
because of three reasons. First, the task interdependence manipulation was rather weak.
The difference between the low- and high- interdependence tasks was the need to share or
not to share 7 pieces of information that were needed to ﬁx 44 errors. Second, the sample
size was rather small, a total of 80 teams, for this study. Third, the interaction effect is
hard to detect in most studies in general. As a result, many researchers recommend that
the cutoff point for p value should be increased to .10 from .05. Because of these three
reasons, we decided that hypothesis 16 was partially supported and concluded that task
interdependence interacted with taskwork effort to inﬂuence team performance, instead
of rejecting the hypothesis because of the p value of .087.

Unexpectedly, Hypothesis 15 was rejected. Although teamwork effort had a
positive effect on team performance, its effect did not vary by the level of task
interdependence. Because the p value of the interaction term was quite high, it unlikely
occurred because of the small sample size. This result changed our view of the effect of
teamwork effort. It showed that though teamwork has a positive effect on team

performance, its effect did not vary for tasks that had moderate to high level of

62

interdependence. This interpretation is reasonable in hindsight. For our task, receiving
one piece of information from another team member allowed the receiver to ﬁx one error.
Because the receiver could readily ﬁx the error when having sufﬁcient information, the
ratio of shared pieces of information and additional ﬁxes likely stayed 1:1 and
unchanged. The more teamwork likely led to higher team performance. This effect
occurred whether the task was high or low on task interdependence. Therefore, the
interaction effect was not supported.

Previous work had also investigated the effect of positive reward interdependence
on team performance. The results have been mixed. One possible explanation is the
effect depends on the operationalization of positive reward interdependence. The other
reason is that the effects probably vary by the level of task interdependence (Rausenbaum
et al., 1980; Gordon et al., 2000). The nature of the ﬁxed effect, however, made such
explanations less certain. More importantly, the question why the effect of positive
reward interdependence vary by the operationalization of positive reward
interdependence and task interdependence remains unanswered as previous work has
made prior assumptions (Rosenbaum et al., 1980; Wageman & Baker, 1997) about the
underlying process or inferred the process retroactively (Gordon et al., 2000).

Our study made a contribution by explaining the underlying mechanism. The
result of our study showed that positive reward interdependence had opposite effects on
two behavioral processes, teamwork effort and taskwork effort, which in turn had
positive effects on team performance. High positive reward interdependence likely led to
higher level of teamwork effort but lower level of taskwork effort, whereas low positive

reward interdependence likely resulted in lower level of teamwork effort but a higher

63

level of taskwork effort. The effects of reward interdependence on teamwork effort and
taskwork effort could be explained in part through its effects on team members’
cooperative and competitive orientations, respectively. The effect of taskwork effort
likely varied by the level of task interdependence of the tasks.

Practical Implications

 

If the results of the present work are replicated in future studies, they have great
implications for the practice of how to reward team members. According to our literature
review, positive reward interdependence is more effective than negative reward
interdependence or zero reward interdependence. Our study showed that within the
narrow range of positive reward interdependence, higher interdependence did not
necessarily led to higher performance (For our task, it actually led to lower performance).
As Wageman and Baker (1997) suggested, the effect of positive reward interdependence
perhaps was moderated by task interdependence. Given the nature of the task
interdependence facing a team is relatively ﬁxed and unchanged, the important question
organization faces is how to change the level of positive reward interdependence to
motivate high performance. Our study suggested that positive reward interdependence
had opposite effects on teamwork effort and taskwork effort, and its effect could be
explained by team members’ cooperative and competitive orientations. This piece of
knowledge can offer useﬁrl guidance on the direction one should follow in changing the
level of positive reward interdependence to improve team performance. When team
performance is unsatisfactory, one needs to examine whether teamwork effort or
taskwork effort is the main cause. If teamwork effort is, one should make the reward

system less differential and more positively correlated. On the other hand, if taskwork

64

effort is the main cause of low performance, one should make the reward system more
differential and less positive correlated. However, one should proceed with caution
because changing the level of reward interdependence involves trade-offs. Increasing
teamwork behavior tends to decreasing taskwork behavior, and vice versa. When
teamwork and taskwork behaviors are not available to guide reward intervention, one can
assess team members’ cooperative and competitive orientations and use them as
diagnostic information. The same process can be followed to determine the direction of
change in positive reward interdependence.
Limitations

There are several limitations of the study. First, the study was conducted in a lab
setting with psychology undergraduate students using a relatively simple task. As many
lab studies, the external validity of the present is questionable and needs to be tested in
the ﬁeld. Second, the studies used ad-hoc teams with a total lifespan of 1.5 hours, and
members worked on the team task for a total of 45 minutes. This kind of team is
temporary and may not represent typical teams that often share a history and prior
experiences (Kozlowski & Bell, 2003). The results of study were therefore more
reﬂective of the early stage of team development. Third, our study operationalized
positive reward interdependence with only two levels of reward distributions. As such,
the results of the study represented ﬁxed effects, which may not generalize to other levels
of positive reward interdependence.

Although the lab study has limitations, the high level of control of the lab setting
is needed to address the research question raised in the present work. The issue here is

how positive reward interdependence affects team performance. It requires the

65

examinations of team members’ psychological and behavioral reactions in responding to
different reward systems in a team environment. This type of question requires the
manipulation of the reward system and a clear contingency between performance and
reward. The effects of reward interdependence on cooperative and competitive
orientations need to be isolated. Moreover, teamwork effort, taskwork effort, and team
performance need to be distinguished and measured adequately. These requirements
would be very hard to be fulﬁlled in the ﬁeld setting because of low level of control.
Typically, researchers do not have much control over pay manipulations for both teams
and individuals in the ﬁeld. The team performance-reward and individual performance-
reward contingencies are often not clearly established. The cooperative or competitive
orientation of a worker is likely a function of individual differences, reward
interdependence, role interdependence, and goal interdependence embedded in the ﬁeld
setting. As a result, it would be very difﬁcult to isolate the effects of reward
interdependence on cooperative and competitive orientations. Furthermore, in the ﬁeld,
teamwork effort and taskwork effort are not easily distinguishable and captured.
Experimental control is needed to differentiate and measure them. For these reasons, a
lab study is needed because of its high level of control.

Despite the shortcomings of the lab setting, we believe, our present study
addressed the research question reasonably. We manipulated the reward distribution in a
moderate way. Subjects clearly understood the basis on which they would be rewarded.
Their cooperative and competitive orientations were assessed right after the reward
manipulation check and had signiﬁcant group differences. Their teamwork and taskwork

effort were measured rigorously. Therefore, the process through which positive reward

66

interdependence inﬂuenced team performance was adequately captured because of the
psychological realism of the experiment.

Like many team reward research studies, the present work suffered the limitations
of the ﬁxed-effect study by having only two levels of positive reward interdependence.
The results of the studies were ﬁxed and speciﬁc to the particular levels of positive
reward interdependence being studied. However, we believe the ﬁxed effects in our
study are generalizable to other levels of positive reward interdependence because there
is a strong theoretical basis to assume that the relationships of positive reward
interdependence and various criteria are monotonic functions (increasing or decreasing
functions). The hypotheses that positive reward interdependence has a positive effect on
cooperative orientation (Deustch, 1985) and a negative effect on competition are
theoretically sound and plausible. Perhaps, most people would probably agree with such
assertions, and it would be hard to imagine the relationships in the reverse direction. The
same argument can be applied to the relationships of positive reward interdependence and
teamwork effort and taskwork effort. Moreover, we have clear prior evidence that
positive reward interdependence has a positive effect on teamwork behaviors (Johnson &
Johnson, 1989). And we also had some collaborating evidence to suggest that it has a
negative relationship with taskwork effort (Newcomb, Brady, & Hartup, 1979;
Rosenbaum et al, 1980; Konovsky & Podsakoff, 1993).

New Research Directiona

Because the effect of rewards on individual-level teamwork and taskwork were

not fully mediated by cooperative and competitive orientation, future studies should

examine additional psychological states. One of the potential mediators is state anxiety.

67

When positive reward interdependence is low, performance evaluation likely becomes
more salient. Team members may then experience heightened state anxiety. Anxiety
was known to affect individual effort and performance. State anxiety therefore can be
potential mediator and should be examined at both the individual and team levels.

Another potential mediator is the perception of valence of the individual reward.
Past research has clearly established that individuals tend to overestimate their ability and
performance. It follows that they will expect to have higher averaged performance
ranking within teams and receive higher level of rewards on average when team reward is
distributed differentially. Thus, positive reward interdependence may have a negative
relationship with aggregated perception of the valence of individual rewards.

Although individual differences were assumed to contribute to the variance of the
social orientations in conjunction with reward interdependence, their effects were not
examined in the present study. To further expand our knowledge, future research should
examine individual difference factors that can affect social orientation. Two potential
individual factors are collectivism and individualism. By deﬁnition, individualists tend
to accord greater importance to their personal needs than the needs of the groups
(Wagner, 1995). They likely look after themselves and ignore group interests if they
conﬂict with their personal desires. Collectivists, on the other hand, tend to give higher
priority to the interests of their groups than their personal needs (Wagner, 1995). They
likely look out for the interest of the groups to which they belong, even if such actions
require the sacriﬁce of personal interests. Therefore, it is likely that collectivism has a
positive effect on cooperative orientation and teamwork effort, whereas individualism has

a positive effect on competitive orientation and taskwork effort.The task we examined in

68

the present studies is relatively simple. For a more complex task, having high level of
cooperative orientation does not necessarily translate into individual teamwork effort.
Similarly, high level of competitive orientation may not lead to taskwork effort. Future
research may need to examine factors that may moderate the relations of social
orientation and team process behaviors. Perhaps, individual social skills moderate the
relationship between cooperative orientation and teamwork effort. The current study
only examines team performance during a short duration. Future research should
examine the effect of positive reward interdependence over a long term. The effects of
team feedback will probably become relevant over time. It will be informing to know
how social orientation and process behaviors vary over time, and how they will vary

when teams fail or succeed.

69

APPENDICES

70

Appendix A: Html Version of the Task

<html>
<head><title>This page is authored by the Greek Letter Specialist</title></head>

<body>

This page is authored by the Greek Letter Specialist.
<br>

<br>

Two symbols can be combined to form a unit. Below are 44 units:
<br>

<br>

greek small kappa and capital U circumﬂex: ??
<br>

greek capital tau and inverted exclamation sign: ??
<br>

greek capital delta and greek small zeta: ??
<br>

greek capital beta and greek capital zeta: ??
<br>

greek small omicron and capital A grave: ‘2?
<br>

<br>

greek small zeta and greek capital kappa: ??
<br>

greek capital alpha and regular currency sign: ??
<br>

greek small rho and small u grave: ??

<br>

greek capital rho and greek small pi: ??

<br>

greek small sigma and little eth sign: ??

<br>

<br>

greek capital zeta and greek small nu: ??

<br>

greek small iota and capital A circumﬂex: ??
<br>

greek capital upsilon and Japanese yen sign: ??
<br>

greek small delta and small 0 acute: ??

<br>

greek capital omicron and little ae sign: ??
<br>

<br>

71

greek small nu and greek capital chi: ??
<br>

greek capital lambda and capital A ring: ??
<br>

greek small theta and small cent sign: ??
<br>

greek capital nu and greek small omicron: ??
<br>

greek small beta and greek capital rho: ??
<br>

<br>

greek capital theta and greek small theta: ??
<br>

greek small pi and greek capital eta: ??
<br>

greek capital iota and small i circumﬂex: ??
<br>

greek small xi and greek capital tau: ??
<br>

greek capital mu and small function sign: ??
<br>

<br>

greek capital chi and greek small kappa: ??
<br>

greek capital psi and capital A dasis: “.7?
<br>

greek small epsilon and greek capital pi: ??
<br>

greek capital xi and large pound sign: ??
<br>

greek capital phi and greek small iota: ??
<br>

<br>

greek small upsilon and small 0 circumﬂex: ??
<br>

greek capital gamma and greek capital upsilon: ??
<br>

greek small psi and greek capital theta: ??
<br>

greek capital sigma and greek small xi: ??
<br>

greek small tau and small division sign: ??
<br>

<br>

72

greek capital pi and greek small mu: ??

<br>

greek small gamma and greek capital sigma: ??
<br>

greek capital omega and small 0 dasis: ??

<br>

greek small alpha and greek capital iota: ‘2?
<br>

greek capital epsilon and n tilde sign: ??

<br>

<br>

greek capital kappa and greek small lambda: ??
<br>

greek small lambda and capital A acute: ??
<br>

greek capital eta and greek capital phi: ??

<br>

greek small mu and little thorn sign: ??
<fbody>

</html>

73

<html>
<head><title>This page is authored by the Latin Letter Specialist</title></head>

<body>

This page is authored by the Latin Letter Specialist
<br>
<br>

Two symbols can be combined to form a unit. Below are 44 units:
<br>
<br>

capital I grave and left gumet sign: ??
<br>

small u dasis and greek small alpha: ??
<br>

capital Y acute and small u acute: ??
<br>

small a grave and capital Y acute: ??
<br>

capital U acute and inverted question sign: ??
<br>

<br>

capital 0 dasis and small i acute: ??
<br>

small y dasis and and capital U grave: ??
<br>

capital 0 grave and greek capital epsilon: ??
<br>

small c cedilla and capital 0 stroke: ??
<br>

capital E grave and small degree sign: ??
<br>

<br>

small u circumﬂex and capital E grave: ??
<br>

capital 0 acute and small y acute: ??
<br>

small i circumﬂex and greek capital psi: ??
<br>

capital C cedilla and capital 1 grave: ??
<br>

small e acute and right gumet sign: ??

74

<br>
<br>

small i grave and greek capital gamma: ??
<br>

capital A grave and small e grave: ??

<br>

small e grave and capital C cedilla: ??
<br>

capital A ring and spacing macron sign: ??
<br>

small 0 circumﬂex and capital U acute: ??
<br>

<br>

capital A circumﬂex and greek small delta: ??
<br>

small a circumﬂex and capital E circumﬂex: ??
<br>

capital 0 stroke and three quarters sign: ??
<br>

small sharp s and greek capital beta: ??

<br>

capital U grave and small u circumﬂex: ??
<br>

<br>

capital E acute and small registered sign: ‘2?
<br>

small i acute and small y dasis: ??

<br>

capital U dasis and greek small beta: ??
<br>

small u grave and one half sign: ??
<br>

capital A acute and small sharp s: ??
<br>

<br>

capital I dasis and small u dasis: ??
<br>

small 0 acute and capital 0 circumﬂex: ??
<br>

75

capital A dasis and greek capital delta: ??
<br>

small 11 acute and plus-minus sign: ??
<br>

capital 0 circumﬂex and small e acute: ??
<br>

<br>

small e circumﬂex and greek small gamma: ??
<br>

capital U circumﬂex and small i grave: ??
<br>

small a acute and capital 0 dasis: ??

<br>

capital E circumﬂex and one quarter sign: ??
<br>

small 0 dasis and greek capital alpha: ??
<br>

<br>

capital I acute and small e circumﬂex: ??
<br>

small a dasis and greek small epsilon: ??
<br>

capital I circumﬂex and capital E acute: ??
<br>

small y acute and circumﬂex accent sign: ??

<fbody>
<fhtml>

76

<html>
<head><title>This page is authored by the Special Sign Specialist<ltitle></head>

<body>

This page is authored by the Special Sign Specialist.
<br>
<br>

Two symbols can be combined to form a unit. Below are 44 units:
<br>
<br>

book section sign and greek capital lambda: ??
<br>

feminine indicator sign and capital I acute: ??
<br>

superscript two sign and large OE sign: ??
<br>

small function sign and feminine indicator sign: ??
<br>

small division sign and greek capital mu: ??
<br>

<br>

little ae sign and large caron sign: ??

<br>

inverted question sign and small 0 cedilla: ??
<br>

little tilde sign and large THORN sign: ??
<br>

small registered sign and greek small sigma: ??
<br>

large pound sign and capital 0 grave: ??
<br>

<br>

superscript three sign and greek small psi: ??
<br>

three quarters sign and large AE sign: ??
<br>

little thorn sign and capital I circumﬂex: ??
<br>

n tilde sign and spacing dasis sign: ??

<br>

77

odd square sign and greek capital omicron: ??
<br>
<br>

book paragraph sign and small a circumﬂex: ??
<br>

broken bar sign and greek capital xi: ??

<br>

a tilde sign and odd square sign: ??

<br>

inverted exclamation sign and capital 0 acute: ??
<br>

Japanese yen sign and small copyright sign: ??
<br>

<br>

superscript one sign and greek small rho: ??
<br>

little eth sign and O tilde sign: ??

<br>

little oe sign and small a acute: ??

<br>

spacing macron sign and little tilde sign: ??
<br>

large caron sign and greek capital nu: ??
<br>

<br>

large OE sign and book section sign: ??
<br>

small cent sign and capital I dasis: ??
<br>

left gumet sign and superscript two sign: ??
<br>

large ETH sign and greek small tau: ??
<br>

one half sign and little caron sign: ??
<br>

<br>

large dasis sign and small a dasis: ??
<br>

spacing dasis sign and A tilde sign: ??
<br>

small copyright sign and little oe sign: ??
<br>

78

large THORN sign and greek capital omega: ??
<br>

small micro sign and superscript one sign: ??
<br>

<br>

plus-minus sign and book paragraph sign: ??
<br>

one quarter sign and broken bar sign: ??
<br>

little caron sign and greek small upsilon: ??
<br>

right gumet sign and large dasis sign: ??
<br>

large AE sign and capital U dasis: ??

<br>

<br>

O tilde sign and large ETH sign: ??

<br>

circumﬂex accent sign and small a grave: ??

<br>

small degree sign and small micro sign: ??

<br>

regular currency sign and superscript three sign: ??

<fbody>
</html>

79

Appendix B: Web Version of the Task

This page is authored by the Greek Letter Specialist.
Two symbols can be combined to form a unit. Below are 44 units:

greek small kappa and capital U circumﬂex: ??
greek capital tau and inverted exclamation sign: ??
greek capital delta and greek small zeta: ??

greek capital beta and greek capital zeta: ??

greek small omicron and capital A grave: ??

greek small zeta and greek capital kappa: ??
greek capital alpha and regular currency sign: ??
greek small rho and small u grave: ??

greek capital rho and greek small pi: ??

greek small sigma and little eth sign: ??

greek capital zeta and greek small nu: ??

greek small iota and capital A circumﬂex: ??
greek capital upsilon and Japanese yen sign: ??
greek small delta and small 0 acute: ??

greek capital omicron and little ae sign: ??

greek small nu and greek capital chi: ??
greek capital lambda and capital A ring: ??
greek small theta and small cent sign: ??
greek capital nu and greek small omicron: ??
greek small beta and greek capital rho: ??

greek capital theta and greek small theta: ??
greek small pi and greek capital eta: ??
greek capital iota and small i circumﬂex: ??
greek small xi and greek capital tau: ??
greek capital mu and small function sign: ??

greek capital chi and greek small kappa: ??
greek capital psi and capital A dasis: ??
greek small epsilon and greek capital pi: ??
greek capital xi and large pound sign: ??
greek capital phi and greek small iota: ??

greek small upsilon and small 0 circumﬂex: ??
greek capital gamma and greek capital upsilon: ??
greek small psi and greek capital theta: ??

greek capital sigma and greek small xi: ??

80

 

.L. ...u....«..rll~.. r

greek small tau and small division sign: ??

greek capital pi and greek small mu: ??

greek small gamma and greek capital sigma: ??
greek capital omega and small 0 dasis: ??
greek small alpha and greek capital iota: ??
greek capital epsilon and n tilde sign: ??

greek capital kappa and greek small lambda: ??
greek small lambda and capital A acute: ??
greek capital eta and greek capital phi: ??
greek small mu and little thorn sign: ??

81

This page is authored by the Latin Letter Specialist
Two symbols can be combined to form a unit. Below are 44 units:

capital I grave and left gumet sign: ??

small u dasis and greek small alpha: ??
capital Y acute and small u acute: ??

small a grave and capital Y acute: ??

capital U acute and inverted question sign: ??

capital 0 dasis and small i acute: ??

small y dasis and and capital U grave: ??
capital 0 grave and greek capital epsilon: ??
small c cedilla and capital 0 stroke: ??
capital E grave and small degree sign: ??

small u circumﬂex and capital E grave: ??
capital 0 acute and small y acute: ??

small i circumﬂex and greek capital psi: ??
capital C cedilla and capital I grave: ??
small e acute and right gumet sign: ??

small i grave and greek capital gamma: ??
capital A grave and small e grave: ??
small e grave and capital C cedilla: ??
capital A ring and spacing macron sign: ??
small 0 circumﬂex and capital U acute: ??

capital A circumﬂex and greek small delta: ??
small a circumﬂex and capital E circumﬂex: ??
capital 0 stroke and three quarters sign: ??
small sharp s and greek capital beta: ??

capital U grave and small 11 circumﬂex: ??

capital E acute and small registered sign: ??
small i acute and small y dasis: ??

capital U dasis and greek small beta: ??
small u grave and one half sign: ??

capital A acute and small sharp s: ??

capital I dasis and small 11 dasis: ??

small 0 acute and capital 0 circumﬂex: ??
capital A dasis and greek capital delta: ??
small u acute and plus-minus sign: ??
capital 0 circumﬂex and small e acute: ??

82

small e circumﬂex and greek small gamma: ??
capital U circumﬂex and small i grave: ??
small a acute and capital 0 dasis: ??

capital E circumﬂex and one quarter sign: ??
small 0 dasis and greek capital alpha: ??

capital I acute and small e circumﬂex: ??
small a dasis and greek small epsilon: ??
capital I circumﬂex and capital E acute: ??
small y acute and circumﬂex accent sign: ??

83

This page is authored by the Special Sign Specialist.
Two symbols can be combined to form a unit. Below are 44 units:

book section sign and greek capital lambda: ??
feminine indicator sign and capital I acute: ??
superscript two sign and large OE sign: ??

small function sign and feminine indicator sign: ??
small division sign and greek capital mu: ??

little ae sign and large caron sign: ??

inverted question sign and small c cedilla: ??
little tilde sign and large THORN sign: ??
small registered sign and greek small sigma: ??
large pound sign and capital 0 grave: ??

superscript three sign and greek small psi: ??
three quarters sign and large AE sign: ??

little thorn sign and capital I circumﬂex: ??

n tilde sign and spacing dasis sign: ??

odd square sign and greek capital omicron: ??

book paragraph sign and small a circumﬂex: ??
Broken bar sign and greek capital xi: ??

A tilde sign and odd square sign: ??

inverted exclamation sign and capital 0 acute: ??
Japanese yen sign and small copyright sign: ??

superscript one sign and greek small rho: ??
little eth sign and O tilde sign: ??

little oe sign and small a acute: ??

spacing macron sign and little tilde sign: ??
large caron sign and greek capital nu: ??

large OE sign and book section sign: ??
small cent sign and capital I dasis: ??

left gumet sign and superscript two sign: ??
large ETH sign and greek small tau: ??

one half sign and little caron sign: ??

large dasis sign and small a dasis: ??

spacing dasis sign and A tilde sign: ??

small copyright sign and little oe sign: ??

large THORN sign and greek capital omega: ??
small micro sign and superscript one sign: ??

84

plus-minus sign and book paragraph sign: ??
one quarter sign and broken bar sign: ??
little caron sign and greek small upsilon: ??
right gumet sign and large dasis sign: ??
large AE sign and capital U dasis: ??

O tilde sign and large ETH sign: ??

circumﬂex accent sign and small a grave: ??

small degree sign and small micro sign: ??

regular currency sign and superscript three sign: ??

85

Exhibit C: Training Manual

Learning to Write a Web Page (For Greek Letter Specialist)

Purpose: This tutorial will guide you on how to write a web page using html, the
main Internet programming language. It focuses on teaching you how to write
Greek Letter Symbols. Throughout the tutorial, you will see detailed instructions,
sample html codes, and their actual appearance on Internet Explorer. You will also
have opportunities to do some hands-on exercises.

HTML Document Structure

 

HTML files are just normal text ﬁles. They usually have the extension html.
HTML documents have two parts: the head and the body. The body is the larger
part of the document, as the body of a letter you would write to a friend would be.
The head of the document only contains the document's title and similar
information.

Example of basic HTML document structure:
<html>

<head><title>TItle goes here</title></head>
<body>

The body goes here. The body text is written in htrnl just as how it is written in Microsoft Word.
You can just type your text on and on. When you want to start a new line, you need to Insert
the following break tag.

<br>

This sentence will start in a new line. Without the tag, the next sentence just follows the
previous one until the end of the line where it wraps. When you want double space the next
line, you can just enter two break tags as following.

<br>

<br>

Symbols, however, are coded by a series of unique numbers. For example, &#936 is a Greek
capital psi.

<br>

A Greek capital phi is &#934.

</body>

</html>

86

i

Some Fundamentals: Whatever appears between the TITLE tags will be the title
of the document, which can be found in the title bar at the top of the screen of an
Internet Explorer window.

Whatever you place between the BODY tags will appear in the major area of the

document window. When you write a web page, you will write in the area between

<body> and </body>. You can just ignore the text between <header> and
</header>. If you want to start text in the new line, you need insert the break tag

<br>. If you want extra space between two lines, just enter two break tags, <br>
<br>.

Symbols: Symbols are special characters. Unlike a typical character we know
such as “a”, html language often requires a number of keystrokes to display a
symbol. Typically, a symbol can be coded by using a series of number. For
example, &#934 will display the Greek capital phi symbol, (D. &#936 will display
the Greek capital psi symbol, ‘1’.

 

Try it yourself #1
Type the following HTML code in Notepad.
To open Notepad, go to
Start
Programs
Accessories
Notepad

Note: Feel free to use COPY and PASTE if you feel comfortable with the code and
don't want to type it all yourself.

<html>
<head> <title>Title goes here</title> </head>
<body>

The body goes here. The body text is written in html just as how it is written in Microsoft Word.
You can just type your text on and on. When you want to start a new line, you need to insert
the following tag.

<br>

This sentence will start in a new line. Without the tag, the next sentence just follows the
previous one until the end of the line where it wraps. When you want double space the next
line, you can just enter two break tags as following.

<br>

<br>

87

Symbols are entered by a series of unique numbers. For example, &#936 is a Greek capital
psi.

<br>

A Greek capital phi is &#934.

</body>

</html>

Save the file as ExampIeOne.html in folder CleA
To see how this files appears on the Internet,

Open the Internet Explorer

In the Address Box, type: C: [AAA] ExampleOne.html

If you enter the right codes, you will see “Title go here" appear at the top of your
Internet Explorer. In the main page, you will see:

The body goes here. The body text is written in html just as how it is written in Microsoft Word.
You can just type your text on and on. When you want to start a new line, you need to Insert
the following tag.

This sentence will start in a new line. Without the tag, the next sentence just follows the
previous one until the end of the line where it wraps. When you want double space the next
line, you can just enter two break tags as following.

Symbols are entered by a series of unique numbers. For example, ‘1’ is a Greek capital psi.
A Greek capital phi is (D.

 

In this tutorial, you will learn how to write Greek Letter Symbols and
will become a Greek Letter Specialist. The html codes for 50 Greek
Letter Symbols are in the ﬁle C:/AAA/GreekLetters.ﬂf. Please, double
click on the preceding link to open the ﬁle and quickly read through its
content. Remember you gon’t have to memorize the html codes for
these symbols. as you can refer back to this reference ﬁle later any time

you need.

 

Try it yourself #2

Type the following HTML code in Notepad. To open Notepad, go to
Start

Programs

Accessories

Notepad

88

Note: Feel free to use COPY and PASTE if you feel comfortable with the code and
don't want to type it all yourself.

<html>

<head><title>Title goes here</title></head>
<body>

This is a Greek capital upsilon ??.

<br>

This is a Greek Capital lambda ??.
</body>

</html>

 

Now replace the question marks with the html codes for Greek capital upsilon and
Greek Capth lamba. You can find the html codes for these two symbols in file

C :/AAA/GreekLetters.pdf you just read.

Save the file as ExampIeTwo.html in folder C: [AAA
To see how this file appears on the Internet

Open the Internet Explorer

In the Address Box, type: C:/AAA/Example1\uo.html

The correct codes will show the following in the main page:

This is a Greek capital upsilon t).
This is a Greek capital lambda A.

 

Try it yourself #3
Type the following HTML code in Notepad. To open Notepad, go to
Start
Programs
Accessories
Notepad

Note: Feel free to use COPY and PASTE if you feel comfortable with the code and
don't want to type it all yourself.

<html>

<head> <title>11tle goes here</title></head>

<body>

89

This is a Greek capital chi ??.
<br>

This is a Greek capital pi ??.

</body>
</html>

Now replace the question marks with the html codes for Greek capital chi and Greek
Capital pi. You can find the html codes for these two symbols in file

C :/AAA/GreekLetters.gdf you just read.

Save the file as Exampleﬂrreehtml in folder C: [AAA

To see how this file appears on the Internet

Open the Internet Explorer

In the Address Box, type: C: [AAA] Example1'hree.html

The correct codes will show the following in the main page:

This is a Greek capital chi X.
This is a Greek capital pi 1'1.

90

Learning to Write a Web Page (For Latin Letter Specialist)

Purgose: This tutorial will guide you on how to write a web page using html, the
main Internet programming language. It focuses on teaching you how to write
Latin Special Letters. Throughout the tutorial, you will see detailed instructions,
sample html codes, and their actual appearance on Internet Explorer. You will also
have opportunities to do some hands-on exercises.

HTML Document Structure

HTML files are just normal text ﬁles. They usually have the extension html.
HTML documents have two parts: the head and the body. The body is the larger
part of the document, as the body of a letter you would write to a friend would be.
The head of the document only contains the document's title and similar
information.

Example of basic HTML document structure:
<html>

<head> <title>TitIe goes here</title> </head>
<body>

The body goes here. The body text is written in html just as how it is written in Microsoft Word.
You can just type your text on and on. When you want to start a new line, you need to insert
the following break tag.

<br>

This sentence will start in a new line. Without the tag, the next sentence just follows the
previous one until the end of the line where it wraps. When you want double space the next
line, you can just enter two break tags as following.

<br>

<br>

Symbols, however, are coded by a series of unique numbers. For example, &#249 is a small u
grave.

<br>

Capital A grave is &#192.

</body>

</html>
Some Fundamentals: Whatever appears between the TITLE tags will be the title

of the document, which can be found in the title bar at the top of the screen of an
Internet Explorer window.

91

-5‘11.‘ 'IJ‘ nu

 

 

 

Whatever you place between the BODY tags will appear in the major area of the
document window. When you write a web we, yog will write in the area between
<body> and </body>. You can just ignore the text between <header> and
</header>. If you want to start text in the new line, you need insert the break tag
<br>. If you want extra space between two lines, just enter two break tags, <br>
<br>.

Symbols: Symbols are special characters. Unlike a typical character we know
such as “a”, html language often requires a number of keystrokes to display a
symbol. Typically, a symbol can be coded by using a series of number. For
example, &#249 will display the small it grave, 0. &#192 will display the capital A
grave, A.

 

 

Try it yourself #1
Type the following HTML code in Notepad.
To open Notepad, go to
Start
Programs
Accessories
Notepad

Note: Feel free to use COPY and PASTE if you feel comfortable with the code and
don't want to type it all yourself.

<html>
<head><title>Trtle goes here</title></head>

<body>

The body goes here. The body text is written in html just as how it is written in Microsoft Word.
You can just type your text on and on. When you want to start a new line, you need to insert
the following tag.

<br>

This sentence will start in a new line. Without the tag, the next sentence just follows the
previous one until the end of the line where it wraps. When you want double space the next
line, you can just enter two break tags as following.

<br>

<br>

Symbols are coded by a series of unique numbers. For example, &#249 is a small u grave.
<br>

Capital A grave is &#192.

92

</body>

</html>

Save the file as ExampIeOne.html in folder C:/AAA
To see how this files appears on the Internet,

Open the Internet Explorer

In the Address Box, type: C:IAAA/ExampleOne.html

If you enter the right codes, you will see “Title go here" appear at the top of your
Internet Explorer. In the main page, you will see:

The body goes here. The body text is written in html just as how it is written in Microsoft Word.
You can just type your text on and on. When you want to start a new line, you need to insert
the following tag.

This sentence will start in a new line. Without the tag, the next sentence just follows the
previous one until the end of the line where it wraps. When you want double space the next
line, you can just enter two break tags as following.

Symbols are coded by a series of unique numbers. 1‘: is a small u grave.
<br>
Capital A grave is A.

 

In this tutorial, you will learn how to write Latin Special Letters and
will become a Latin Letter Specialist. The html codes for 50 Special
Signs are in the ﬁle C:/AAA/LatinSgecials.pdf. Please, open this ﬁle (go
to folder C:/AAA and double click on the ﬁle) and quickly read through
its content. Remember you don’t have to memorize the html codes for

these symbols, as you can refer back to this reference ﬁle later any time
you need.

 

Try it yourself #2

Type the following HTML code in Notepad. To open Notepad, go to
Start

Programs

Accessories

Notepad

Note: Feel free to use COPY and PASTE if you feel comfortable with the code and
don't want to type it all yourself.

93

<html>
<head><title>‘l‘rtle goes here</title></head>
<body>

This is a capital E acute ??.
<br>
This is a small i acute ??.

</body>
</htrn|>

Now replace the question marks with the html codes for capital E acute and small I
acute. You can ﬁnd the html codes for these two symbols in file

C:/AAA/LatinSpecials.pdf you just read.

Save the file as ExampleTwo.html in folder c:/AAA
To see how this file appears on the Internet

Open the Internet Explorer

In the Address Box, type: C:/AAA/ExampIeTWo.html

The correct codes will show the following in the main page:

This is a capital E acute E.
This is a small I acute i.

 

Try it yourself #3

Type the following HTML code in Notepad. To open Notepad, go to
Start

Programs

Accessories

Notepad

Note: Feel free to use COPY and PASTE if you feel comfortable with the code and
don't want to type it all yourself.

<html>

<head><title>1'rtle goes here</title></head>

<body>

This is a capital U circumﬂex ??.
<br>
This is a small a circumﬂex ??.

94

 

</body>
</html>

Now replace the question marks with the html codes for capital U circumflex and
small a circumflex. You can find the html codes for these two symbols In file

C:/AAA/LatinSgecials.gdf you just read.

Save the file as ExampleThree.html in folder C: [AAA
To see how this ﬁle appears on the Internet

Open the Internet Explorer

In the Address Box, type: C: /AAA/Example1hree.html

The comct codes will show the following In the main page:

This is a capital U circumﬂex U.
This is a small a circumﬂex a.

 

95

 

Learning to Write a Web Page (For Special Sign Specialist)

Purpose: This tutorial will guide you on how to write a web page using html, the
main Internet programming language. It focuses on teaching you how to write
Special Sign Symbols. Throughout the tutorial, you will see detailed instructions,
sample html codes, and their actual appearance on Internet Explorer. You will also
have opportunities to do some hands-on exercises.

HTML Document Structure

HTML ﬁles are just normal text files. They usually have the extension html.
HTML documents have two parts: the head and the body. The body is the larger
part of the document, as the body of a letter you would write to a friend would be.
The head of the document only contains the document's title and similar
information.

Example of basic HTML document structure:
<html>

<head><title>T|tie goes here</tit|e></head>
<body>

The body goes here. The body text is written in html just as how it is written In Microsoft Word.
You can just type your text on and on. When you want to start a new line, you need to insert
the following break tag.

<br>

This sentence will start in a new line. Without the tag, the next sentence just follows the
previous one until the end of the line where it wraps. When you want double space the next
line, you can just enter two break tags as following.

<br>

<br>

Symbols, however, are coded by a series of unique numbers. For example, &#165 is a
Japanese Yen sing.

<br>

Book section sign is &#167.

</body>

</html>
Some Fundamentals: Whatever appears between the TITLE tags will be the title

of the document, which can be found in the title bar at the top of the screen of an
Internet Explorer window.

96

 

—l

Whatever you place between the BODY tags will appear in the major area of the

document window. When you write a web page, you will write in the area between

<body> and </body>. You can just ignore the text between <header> and
</header>. If you want to start text in the new line, you need insert the break tag

<br>. If you want extra space between two lines, just enter two break tags, <br>
<br>.

Symbols: Symbols are special characters. Unlike a typical character we know
such as “a”, html language often requires a number of keystrokes to display a
symbol. Typically, a symbol can be coded by using a series of number. For
example, &#165 will display the Japanese Yen sign, ¥. &#167 will display the book
section sign, §.

 

Try it yourself #1

Type the following HTML code in Notepad.
To open Notepad, go to

Start

Programs

Accessories

Notepad

Note: Feel free to use COPY and PASTE if you feel comfortable with the code and
don't want to type it all yourself.

<html>
<head><title>‘l'itle goes here</title></head>
<body>

The body goes here. The body text is written In html just as how it is written in Microsoft Word.
You can just type your text on and on. When you want to start a new line, you need to insert
the following tag.

<br>

This sentence will start in a new line. Without the tag, the next sentence just follows the
previous one until the end of the line where it wraps. When you want double space the next
line, you can just enter two break tags as following.

<br>

<br>

Symbols are coded by a series of unique numbers. For example, &#165 is a Japanese Yen
sign.

<br>

Book section sign is &#167.

97

 

</body>

</html>

Save the ﬁle as ExampleOne.html in folder C: [AAA
To see how this ﬁles appears on the Internet,
Open the Internet Explorer

In the Address Box, type: C: lAAA/ ExampIeOne.html

If you enter the right codes, you will see “Title go here" appear at the top of your
Internet Explorer. In the main page, you will see:

The body goes here. The body text is written in html just as how it is written in Microsoft Word.
You can just type your text on and on. When you want to start a new line, you need to Insert
the following tag.

This sentence will start In a new line. Without the tag, the next sentence just follows the
previous one until the end of the line where it wraps. When you want double space the next
line, you can just enter two break tags as following.

Symbols are coded by a series of unique numbers. ¥ is a Japanese Yen sign.
<br>
Book section sign is §.

 

In this tutorial, you will learn how to write Special Signs and will
become a Special Sign Specialist. The html codes for 50 Special Signs
are in the ﬁle C:/AAA/SpecialSigns.pdf. Please, open this ﬁle (go to
folder C:/AAA and double click on the ﬁle) and quickly read through its
content. Remember you gon’t have to memorize the html codes for

these symbols, as you can refer back to this reference ﬁle later any time
you need.

 

Try it yourself #2

Type the following HTML code in Notepad. To open Notepad, go to
Start

Programs

Accessories

Notepad

Note: Feel free to use COPY and PASTE if you feel comfortable with the code and
don't want to type it all yourself.

98

<html>

<head><title>Title goes here</tit|e></head>

<body>

This is a small copyright sign ??.
<br>

This is an n tilde sign ??.

</body>
</html>

Now replace the question marks with the html codes for small copyright sign and n
tilde sign. You can ﬁnd the html codes for these two symbols in ﬁle

C :/AAA/SpecialSigns.pdf you just read.

Save the file as ExampleTwo.html in folder C: [AAA
To see how this ﬁle appears on the Internet

Open the Internet Explorer

In the Address Box, type: c:/AAA/ExampleTwo.html

The correct codes will show the following In the main page:

This is a small copyright sign ©.
This is an n tilde sign it.

 

Try it yourself #3

Type the following HTML code in Notepad. To open Notepad, go to
Start

Programs

Accessories

Notepad

Note: Feel free to use COPY and PASTE if you feel comfortable with the code and
don't want to type it all yourself.

<html>

<head><title>TitIe goes here</title></head>

<body>

This is a book paragraph sign ??.
<br>

99

This is a large OE sign ??.

</body>
</html>

Now replace the question marks with the html codes for book paragraph sign and
large OE sign. You can ﬁnd the html codes for these two symbols In ﬁle

C :/AAA/SpecialSigns.pdf you just read.

Save the ﬁle as ExampleThree.html in folder C: [AAA
To see how this ﬁle appears on the Internet

Open the Internet Explorer

In the Address Box, type: C: lAAA/ ExampleThree.html

The correct codes will show the following in the main page:

This is a book paragraph sign.
This is a large OE sign.

 

100

 

Exhibit D: Coding Manual
50 Greek Letters

&#927 greek capital omicron
&#965 greek small upsilon
&#913 greek capital alpha
&#963 greek small sigma
&#915 greek capital gamma

&#961 greek small rho
&#919 greek capital eta
&#959 greek small omicron
&#921 greek capital iota
&#957 greek small nu

&#914 greek capital beta
&#955 greek small lambda
&#926 greek capital xi
&#953 greek small iota
&#922 greek capital kappa

&#950 greek small zeta
&#928 greek capital pi
&#948 greek small delta
&#929 greek capital rho
&#946 greek small beta

&#931 greek capital sigma
&#932 greek capital tau
&#937 greek capital omega
&#945 greek small alpha
&#933 greek capital upsilon

101

 

 

&#935 greek capital chi
&#947 greek small gamma
&#936 greek capital psi
&#949 greek small epsilon
&#934 greek capital phi

&#952 greek small theta
&#923 greek capital lambda
&#954 greek small kappa
&#917 greek capital epsilon
&#956 greek small mu

&#924 greek capital mu
&#958 greek small xi
&#925 greek capital nu
&#960 greek small pi
&#920 greek capital theta

&#968 greek small psi
&#916 greek capital delta
&#964 greek small tau
&#918 greek capital zeta
&#966 greek small phi

&#967 greek small chi
&#962 greek ﬁnal sigma
&#963 greek small sigma
&#998 greek pi square
&#978 greek hook upsilon

102

50 Latin Special Letters

&#233 small e acute
&#193 capital A acute
&#236 small i grave
&#196 capital A dasis
&#243 small 0 acute

&#192 capital A grave
&#225 small a acute
&#204 capital I grave
&#234 small e circumﬂex
&#221 capital Y acute

&#223 small sharp s
&#205 capital I acute
&#255 small y dasis
&#218 capital U acute
&#226 small a circumﬂex

&#201 capital E acute
&#228 small a dasis
&#214 capital 0 dasis
&#232 small e grave
&#207 capital I dasis\

&#250 small 11 acute
&#206 capital I circumﬂex
&#253 small y acute
&#211 capital 0 acute
&#249 small u grave

&#210 capital 0 grave

&#251 small 11 circumﬂex
&#220 capital U dasis

103

&#252 small u dasis
&#216 capital O stroke

&#237 small i acute

&#217 capital U grave
&#244 small 0 circumﬂex
&#194 capital A circumﬂex
&#231 small c cedilla

&#219 capital U circumﬂex
&#246 small 0 dasis

&#197 capital A ring
&#199 capital C cedilla
&#224 small a grave

&#200 capital E grave
&#202 capital E circumﬂex
&#238 small i circumﬂex
&#212 capital O circumﬂex
&#230 small ae diphthong

&#107 small k letter
&#100 lowercase (1 letter
&#114 small r cursy
&#122 lower 2 cursy
&#100 small n cursy

104

r

h
,_

50 Special Signs

&#168 spacing dasis sign
&#732 little tilde sign
&#169 small copyright sign
&#978 odd square sign
&#222 large THORN sign

&#166 Broken bar sign
&#710 circumﬂex accent sign
&#167 Book section sign
&#254 little thorn sign

&#170 feminine indicator sign

&#402 small function sign
&#176 small degree sign
&#352 large caron sign
&#178 superscript two sign
&#240 little eth sign

&#338 large OE sign
&#177 plus-minus sign
&#339 little oe sign
&#189 one half sign
&#376 large dasis sign

&#188 one quarter sign
&#353 little caron sign
&#190 three quarters sign
&#187 right gumet sign
&#198 large AE sign

&#230 little ae sign

&#179 superscript three sign
&#208 large ETH sign

105

 

 

 

&#191 inverted question sign
&#247 small division sign

&#171 left gumet sign
&#162 small cent sign
&#175 spacing macron sign
&#165 Japanese yen sign
&#182 book paragraph sign

&#185 superscript one sign
&#241 n tilde sign
&#195 A tilde sign
&#213 O tilde sign

&#161 inverted exclamation sign

&#174 small registered sign
&#163 large pound sign
&#181 small micro sign
&#164 regular currency sign
&#130 left single quote

&#132 right single quote
&#093 right square bracket
&#092 left square bracket
&#125 right curly bracket
&#123 left curly bracket

106

 

 

 

Appendix E: Team Task Instruction
PLEASE, READ THE INSTRUCTION FOR THE NEXT TASK.
Your Team Task

You have been trained to become an expert in one area while your two other team
members have learned to become specialists in two other areas. In the next team task,
you will work with the other two members to ﬁx a website. The website has 3 pages.
Each page has 44 errors, so there are 132 errors in total. Your team will have a 25 %
chance to win $90 cash bonus for this task.

Your Assignment

Each of you will be assigned to ﬁx one page of the website for which you have the most
expertise. The page is set up so that you don’t have the knowledge to ﬁx all the errors
yourself. If you don’t know how to ﬁx an error, one of your team members does. So you
need to communicate and work with your team members if you want to ﬁx all the errors.
You guys can work in any way you want. The only restriction is that you cannot
physically type in the codes for your other team members.

Your Team and Your Own Performance

Your team performance will be the total number of errors that all three team members
correctly ﬁx. Your performance will be the total number of errors that you correctly ﬁx
in your assigned page. Remember that a ﬁx should contain a unit of two correct symbols.
If one of the symbols in the unit is not corrected, it is still considered an error.

If your team performance is in the top 25 % of all performing teams, your team will
win $90.
The following lines are different for each condition
High positive reward interdependence:

If your team wins $90, each one of you will get $30.
Low positive reward interdependence:

If your team wins $90, the top performer of your team will get $45; the middle
performer will get $30; and the bottom performer will get $20.

You are assigned to work on page C:/AAA/member#.html. Please, open this ﬁle in
both Notepad and Internet Explorer.

PLEASE, DON’T START YET. LET THE EXPERIMENTER KNOW ONCE
YOU HAVE THE FILE OPEN SO HE/SHE CAN START EVERYONE AT THE
SAME TIME.

107

 

 

 

Appendix F: Information Exchange Checklist

 

 

 

  
    
  

 

 

 

108

 

 

Symbol Codes Latin Ask Response Sign ask Response

reek capital beta &#914
greek capital chi &#935
greek capital delta &#916
greek capital epsilon &#917
greek capital eta &#919
greek capital gamma &#915

reek capital iota &#921

reek capital kappa &#922

reek capital lambda &#923
greek capital mu &#924
greek capital nu &#925
greek capital omega &#937
greek capital omicron &#927
greek capital phi &#934
greek capital pl &#928
greek capital psi &#936
greek capital rho &#929
greek capital sigma &#931
greek capital tau &#932
greek capital theta &#920
greek capital upsilon &#933
greek capital xi &#926
greek capital zeta &#918
greek capital alpha &#913
greek final sigma &#962
greek hook upsilon &#978
greek pl square &#998
greek small alpha &#945
greek small beta &#946
greek small delta &#948
greek small epsilon &#949
greek small gamma &#947
greek small iota &#953
greek small kappa &#954
greek small lambda &#955
greek small mu &#956
reek small nu &#957

 

 

greek small
greek small
greek small

greek small
greek small
greek small
greek small
greek small

greek small

 

omicron
pi
rho

sigma
tau
theta
upsilon
xi

zeta

greek small chi

greek small phi
greek small psi
reek small si ma

 

 

 

 

109

 

 

 

 

Appendix G
Questionnaire 1: Demographic Information
Demoggaphic Information

What is your age? Gender? GPA?

 

SAT score? Or ACT score?

 

What is your ethnicity? (Please check one):

Caucasian American
Aﬁ'ican American
Hispanic American

 

Native American
Asian American

DECIDE]

110

f

Appendix H

Questionnaire 2: Manipulation Checks and Social Orientation Scales

Manipulation Checks:
To what extent do you agree or disagree with the following statements

 

1. A correct ﬁx will require the correct codes for two symbols. If one of the
symbols of the unit is not the correct, it is still an error.

2. Your team performance will be the total number of errors all three members of the
team correctly ﬁx.

3. Your performance will be the total number of errors you correctly ﬁx in your
assigned page

4. Your team will win $90 if your team performance is in the top 25 % of the all
performing teams. .

For high reward interdependence condition .1

5. If you team wins $90, you will receive $30.

For low reward interdependence condition

5. If you team wins $90, top performer of the team will receive $40; middle
performance will receive $30; and bottom performer will receive $20

Social Orientation Scales
To what extent do you agree or disagree with the following statements regarding your
expectations and intentions for this task

Cooperative Orientation

I like to help other team members perform.

I like to share my knowledge and materials with other team members.

I like to cooperate with other team members.

I beneﬁt greatly by working with other team members.

I try to shard my knowledge and materials with other team members when I think
it will help them.

Team members can beneﬁt greatly by working with one another.

7. It is a good idea to help other team member perform.

MPP’NT‘

9‘

Competitive Orientation

8. I like to do better work than other team members.

9. I work to get better rewards than other team members do.

10. I like to be the best performer in the team.

11. I don’t like to be second.

12. I like to compete with other team members to see who can do the best work.
13. I am happiest when I am competing with other team members.

14. I like the challenge of seeing who is best.

15. Competing with other team members is a good way to work.

111

 

BIBLIOGRAPHY

112

BIBLIOGRAPHY

Barrick, M. R., Stewart, G. L., Neubert, M. J ., & Mount, M. K. (1998). Relating member
ability and personality to work-team processes and team effectiveness. Journal of
Applied Psychology, 83, 377-391.

Bailey, P., Onwuegbuzie J. A., & Daley, E. C. (2000). Correlates of anxiety at three
stages of the foreign language learning process. Journal of Language and Social
Psychology, 19, 474-490.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51, 1 173—1182.

Bell, B. S., & Kozlowski, S. J. (2002). A typology of virtual teams: Implications for
effective leadership. Group and Organization Management, 27, 14-49.

Bliese, P. D. (2000). Within-group agreement, non-independence, and reliability:
Implications for data aggregation and analyses. In K. J. Klein & S.W. J. Kozlowski
(Eds.), Multilevel theory, research, and methods in organizations: Foundations,
extensions, and new directions (pp. 349—381). San Francisco, CA: J ossey—Bass.

Bowers, C. A., Braun, C. C., & Morgan, B. B., Jr. (1997). Team workload: Its meaning
and measurement. In M. T. Brannick, E. Salas, & C. Prince (Eds.), Team Performance
Assessment and Measurement. Mahwah, NJ: Lawrence Erlbaum Associates.

Bryk & Raudenbush (1992). Hierarchical linear models: Applications and data analysis
methods. Thousand Oaks, CA: Sage Publications, Inc.

Campion, M. A., Papper, E. M., & Medsker, G. J. (1993). Relations between work team
characteristics and effectiveness: A replication and extension. Personnel-Psychology, 49,
429-452.

Campion, M. A., Medsker, G. J ., & Higgs, A. C. (1993). Relations between work team
characteristics and effectiveness: Implications for designing effective work groups.
Personnel Psychology, 46, 823-850.

Cannon-Bowers, J. A., & Salas, E. (1998). Individual and team decision making under
stress: Theoretical underpinnings. In J. A. Cannon-Bowers & E. Salas (Eds.), Making
Decisions under Stress: Implications for Individual and Team Training. Washington, DC:
American Psychological Association.

Chertkoff, J. M., & Mesch, D. J (1997). Performance under different reward systems: A
reconceptualization of the cooperative-competitive-individualistic literature. Advances in
Group Processes, 14, 1-27.

113

 

 

Cohen, S. G., & Bailey, D. E. (1997). What makes teams work: Group effectiveness
research from the shop ﬂoor to the executive suite, Journal of Management, 23, 239-290.

Cotton, J ., & Cook M. (1982). Meta-analyses and the effects of various reward systems:
Some different conclusions from Johnson et al. Psychological Bulletin, 92, 176-183.

Dansereau F., Alutto, J. A., & Yammarino, F. J. (1984). Theory testing in organizational
behavior: The variant approach. Englewood Cliffs, MJ: Prentice Hall.

DeMatteo J. S., Eby, L. T., & Sundstrom, E. (1998). Team-based rewards: Current
empirical evidence and directions for future research. Research in Organizational
Behavior, 20, 141-183.

Deutsch, M. (1949). A theory of cooperation and competition. Human Relation. 2, 129—
151.

Deutsch, M. (1985). Distributive Justice, Yale University Press, New Haven.
Dumaine, B. (1994). The trouble with teams. Fortune, 130, 86-90.

F leishman, E. A., & Zaccaro, S. J. (1992). Toward a taxonomy of team performance
functions. In R. W. Swezey & E. Salas (Eds.), Teams: Their Training and Performance.
Norwood, NJ: Ablex.

Gladstein, D. (1984). Groups in context: A model of task group effectiveness.
Administrative Science Quarterly, 29, 499-517.

Gordon, F. M., Welch, K. R., Ofﬁ'inga, G., & Katz, N. (2000). The complexity of social
outcomes from cooperative, competitive, and individualistic reward systems. Social
Justice Research, 13. 237-269.

Gross, S. E. (1995). Trends in compensations. Benefits and Compensation International,
23 (1), 11-15.

Guzzo, R. A., & Dickson, M. W. (1996). Teams in organizations: Recent research on
performance and effectiveness. Annual Review of Psychology, 47, 307-338.

Hackman, J. R. (1998). Why teams don’t work. In R. S. Tindale et al. (Eds), Theory and
research on small group. New York, NY: Plenum Press.

Hackman, J. R. (1987). The design of work teams. In J. W. Lorsch (Ed.), Handbook of
Organizational Behavior. Englewood Cliffs, NJ: Prentice-Hall.

Hackman J. R., & Morris, C. G. (1975). Group tasks, group interaction process, and

group performance effectiveness: A review and proposed integration. In L. Berkowits
(Ed.), Advances in experimental social psychology. New York, NY: Academic Press.

114

 

 

Hayes, L. (1976). The use of group contingencies for behavioral control: A review.
Psychological Bulletin, 83, 628-648.

Hoﬁnann, D. A., Grifﬁn, M. A., & Gavin, M. B. (2000). The application of hierarchical
linear modeling to organizational research. In K. Klein & S. W. J. Kozlowski (Eds),
Multilevel theory, research, and methods in organizations: Foundations, extensions, and
new directions. (pp. 467-511). San Francisco, CA: Jossey—Bass.

Ilgen, D. R., & Sheppard, L. (1997). Motivation in teams. In M. Erez, H. Thierry, & U.
Kleinbeck (Eds.), Motivational issues in multi-national organizations. New York:
Erlbaum.

James, L. R. (1982). Aggregation bias in estimates of perceptual agreement. Journal of
Applied Psychology, 67, 219-229.

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

James L. R., Demaree R. G., & Wolf, G. (1993). ng: An assessment of within-group
interrater agreement. Journal of Applied Psychology, 78, 306-309.

Johnson, D. W., & Johnson, R. T. (1989). Cooperation and Competition: Theory and
Research. Edina, MN: Interaction Book Company

Johnson, D. W., Maruyama G., Johnson, R., Nelson D., & Skon L. (1981). Effects of
cooperative, competitive, and individualistic goal structures on achievement: A meta-
analysis. Psychological Bulletin, 89, 47-62.

Johnson, D. W., & Norem-Hebeisen, A. A. (1979). A measure of cooperative,
competitive, and individualistic attitudes. The Journal of Social Psychology, 109, 253-
261.

Jones, C. H., Slate, J. R., & Marini, I. (1995). Locus of control, social interdependence,
academic preparation, age, study time, and the study skills of college students. Research
in the Schools, 2, 55-62.

Kanfer, R. (1992). Motivation theory and industrial and organizational psychology. In
M. D. Dunnette & L. M. Hough (Eds.), Handbook of industrial and organizational
psychology (pp 75-170). Palo Alto, CA: Consulting Psychologist Press.

Kirkman, G. L. (2001). Assessing the incremental validity of team consensus ratings

over aggregating of individual-level data in predicting team effectiveness. Personnel
Psychology, 54, 645-663.

115

 

 

 

Klein, K. J ., & Kozlowski, S. W. J. (2002). Multilevel theory, research and methods in
organizations: Foundations, extensions, and new directions. San Francisco, CA: J ossey-
Bass.

Klimoski,R., & Mohammed, S. (1994). Team mental model; Construct or metaphor?
Journal of Management, 20, 403-437.

Konovsky, M. A., & Podsakoff, P. M. (1993). Effects of individual and group incentive
systems and task interdependence on group productivity. Paper presented at the annual
meeting of the Academy of Management, Atlanta, GA.

Kozlowski, S. W. J. & Bell, B. (2003). Work groups and teams in organization. In W.
C. Borman, D. R. Ilgen, & R. J. Klimoski (Eds.), Comprehensive Handbook of
Psychology (Vol 12): Industrial and Organizational Psychology. New York, NY: Wiley.

Lawler, E. E., Mohrman, S., & Ledford, G. (1992). Creating high performance
organizations: Practices and results of employee involvement and TQM in Fortune 1000
companies. San Francisco: Jossey-Bass.

Lawler, E. E., Mohrman, S., & Ledford, G. (1995). Employee involvement and total
quality management: Practices and results in Fortune 1000 companies. San Francisco:

Jossey-Bass. Administrative Science Quarterly 34, 252—276.

London M., & Oldham, R. G. (1977). A comparison of group and individual incentive
plans. Academy of Management Journal, 20, 34-41.

MacKinnon, D. P., Krull, L. J ., & Lockwood, C. M. (2000). Equivalence of the
mediation, confounding, and suppression effect. Prevention Science, 1, 173-181

Marks, M. A., Mathieu, J. E, & Zaccaro, S. J. (2001). A temporally based framework
and taxonomy of team processes. Academy of Management Review, 26, 356-376.

Marks, M. A., Sabella, M. J ., Burke, C. S., & Zaccaro, S. J. (2002). The impact of cross-
training on team effectiveness. Journal of Applied Psychology, 87, 3-13.

Mathieu, J. E., Heffner, T. S., Goodwin, G. F ., Salas, E., & Cannon-Bowers, J. A. (2000).
The inﬂuence of shared mental models on team process and performance. Journal of
Applied Psychology, 85, 273-283.

Michaels, W. J. (1977). Classroom reward structures and academic performance.
Review of educational research, 47, 87-98.

Michaels, W. J. (1978). Effects of differential rewarding and sex on math performance.
Journal of Educational Psychology, 70, 565-573.

116

 

 

 

Miller, L. K. & Hamblin, R. L. (1969). Interdependence, differential rewarding, and
productivity. American Sociological Review, 28, 768-778.

Mischel (1973). Toward a cognitive social learning reconceptualization of personality.
Psychological Review, 80, 252-253.

Morgan, B. B., Jr., Salas, E., & Glickman, A. S. (1994). An analysis of team evolution
and maturation. Journal of General Psychology, 120, 277-291.

Newcomb, A., Brady, J ., & Hartup (1979). Friendship and incentive condition as

determinants of children’s task-oriented social behavior. Child Development, 50, 878-
881.

Onwueqbuzie, A. J ., & Daley, C. E. (1998). Study skills of undergraduates as a function
of academic locus of control, self-perception, and social interdependence. Psychological
Reports, 83, 595-598.

Prince, C. & Salas, E. (1993). Training research for teamwork in the military aircrew. In
E. L Wiener, B. G. Kanki, & R. L. Helmreich (Eds.), Cockpit Research Management.
Orlando, FL: Academic Press.

Scott, w. E., J r., & Chen'ington, D. J. (1974). Effects of competitive, cooperative, and
individualistic reinforcement contingencies. Journal of Personality and Social
psychology, 30, 748-758.

Shea, R. A., & Guzzo, G. B. (1992). Group performance and intergroup relations in
organizations. In Dunnette & Hough (Eds) Handbook of Industrial and Organizational
Psychology (pp. 269-313). Palo Alto, CA: Consulting Psychologists Press, Inc.

Rosenbaum, M. Moore, D., Cotton, J ., Cook, M., Heiser, R., Shovar, N., & Gray, M.
(1980). Group productivity and process: Pure and mixed reward structures and task
interdependence. Journal of Personality and Social Psychology, 39, 626-642.

Salas, E; Dickinson, T. L.; Converse, S. A.; Tannenbaum, S. I. (1992). Toward an
understanding of team performance and training. In R. W. Swezey & E. Salas (Eds.).
Teams: Their Training and Performance (pp. 3-29). Norwood, NJ: Ablex.

Slavin, R. (1983). Cooperative Learning. New York: Longrnan.

Smith-Jentsch, K. A., Johnston, J. H, & Payne, S. C. (1997). Measuring team-related

expertise in complex environments. In J. A. Cannon-Bowers & E. Salas (Eds), Making
Decisions under Stress: Implications for Individual and Team Training. Washington, DC:
American Psychological Association.

117