-
Personnel ReviewInterdependence and fit in team performance
managementHarm van Vijfeijken Ad Kleingeld Harrie van Tuijl Jen A.
Algera Henk Thierry
Article information:To cite this document:Harm van Vijfeijken Ad
Kleingeld Harrie van Tuijl Jen A. Algera Henk Thierry,
(2006),"Interdependence andfit in team performance management",
Personnel Review, Vol. 35 Iss 1 pp. 98 - 117Permanent link to this
document:http://dx.doi.org/10.1108/00483480610636812
Downloaded on: 08 March 2015, At: 00:50 (PT)References: this
document contains references to 34 other documents.To copy this
document: [email protected] fulltext of this
document has been downloaded 3444 times since 2006*
Users who downloaded this article also downloaded:Arthur Morgan,
Kath Cannan, Joanne Cullinane, (2005),"360 feedback: a critical
enquiry", PersonnelReview, Vol. 34 Iss 6 pp. 663-680
http://dx.doi.org/10.1108/00483480510623457Caroline Rowland,
(2013),"Managing team performance: saying and paying",
International Journal ofOrganizational Analysis, Vol. 21 Iss 1 pp.
38-52 http://dx.doi.org/10.1108/19348831311322524Vidhi Agrawal,
(2012),"Managing the diversified team: challenges and strategies
for improvingperformance", Team Performance Management: An
International Journal, Vol. 18 Iss 7/8 pp. 384-400
http://dx.doi.org/10.1108/13527591211281129
Access to this document was granted through an Emerald
subscription provided by 471881 []
For AuthorsIf you would like to write for this, or any other
Emerald publication, then please use our Emerald forAuthors service
information about how to choose which publication to write for and
submission guidelinesare available for all. Please visit
www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.comEmerald is a global
publisher linking research and practice to the benefit of society.
The companymanages a portfolio of more than 290 journals and over
2,350 books and book series volumes, as well asproviding an
extensive range of online products and additional customer
resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The
organization is a partner of the Committeeon Publication Ethics
(COPE) and also works with Portico and the LOCKSS initiative for
digital archivepreservation.
*Related content and download information correct at time of
download.
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
Interdependence and fit in teamperformance management
Harm van VijfeijkenCPS, Amersfoort, The Netherlands
Ad Kleingeld, Harrie van Tuijl and Jen A. AlgeraTechnische
Universiteit Eindhoven, Department of Technology Management,
Eindhoven, The Netherlands
Henk ThierryUniversity of Tilburg, The Netherlands
Abstract
Purpose To evaluate a proposed prescriptive model for the design
of effective combinations ofperformance goals and
pay-for-performance plans for the performance management of
teams.
Design/methodology/approach The idea underlying the model in
which task, goal, andreward interdependence and their fit play a
dominant role is that a pay-for-performance planshould support the
team goals and the goals of individual team members as well as
support the wayin which team members need to cooperate. To obtain a
first notion on the models validity, it wasapplied to evaluate a
pay-for-performance plan for management teams at a large IT
company. Thisevaluation consisted of an in-depth study of three
management teams, using a case studymethodology.
Findings Combinations of fit among type of team, performance
goals, and pay-for-performanceplan (established by a fit between
the interdependence constructs and/or by an overlap in the
contentof the goal and pay indicators) are more effective than
combinations of misfit.
Research limitations/implications The case study was limited to
intra-team interdependencerelationships and did allow for a
analysis of the separate effects of a fit between the
interdependenceconstructs versus content fit.
Practical implications This study shows that pay-for-performance
plans should not be designedin isolation, but rather in alignment
with performance goals and existing task interdependencies.
Originality/value This is the first study to investigate the
three inter-dependence constructs inconjunction in a field
setting.
Keywords Team performance, Team management, Performance-related
pay, Performance measures
Paper type Research paper
IntroductionTeams and pay-for-performance are two possibly
conflicting phenomena that areincreasingly present in modern
organizations (Kozlowski and Bell, 2003; Prendergast,1999). The
possible conflict lies in the fact that teams are generally
associated withcooperation among the team members (West, 1996),
whereas pay-for-performanceplans can easily result in
individualistic or competitive behaviour (Ilgen and Sheppard,2001).
The difficulty of designing a pay-for-performance plan for teams
can be inferredfrom the reviews of Thierry (2002a, b) and
Prendergast (1999), who both found mixedresults for the
effectiveness of pay-for-performance plans for teams. In
thiscontribution we propose and evaluate a prescriptive model for
the design of
The current issue and full text archive of this journal is
available at
www.emeraldinsight.com/0048-3486.htm
PR35,1
98
Received December 2003Accepted July 2004
Personnel ReviewVol. 35 No. 1, 2006pp. 98-117q Emerald Group
Publishing Limited0048-3486DOI 10.1108/00483480610636812
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
pay-for-performance plans that overcome the above conflict and
support teamwork.Teams, in this context, are defined as three or
more people who operate in anorganizational context, see themselves
as a team, are seen by others in the organizationas a team, and
have one or more tasks to perform resulting in team products
(Hackman,1987; Shea and Guzzo, 1987).
The idea underlying this model is that a pay-for-performance
plan should supportthe goals of a team and those of its individual
members, as well as support the way inwhich team members need to
cooperate for the completion of their work. These ideas offit
between the team task, the performance goals and the
pay-for-performance plan areoperationalized via the interdependence
constructs that stem from each of theseelements, i.e. task, goal
and reward interdependence, and via the content of goal andpay
indicators (see Figure 1).
In this model, interdependence relationships are dominant
because we expect thatthey play a crucial role in the design of
effective combinations of pay-for-performanceand performance goals
for teams. Earlier research showed, for example, thatcombinations
of:
(1) task and reward interdependence (e.g. Miller and Hamblin,
1963;Rosenbaum et al., 1980; Wageman, 1995; Wageman and Baker,
1997;
(2) task and goal interdependence (Van der Vegt et al., 2000;
Saavedra et al., 1993;Mitchell and Silver, 1990; Gowen, 1987;
Weldon and Weingart, 1993); and
(3) reward and goal interdependence impact (elements of) team
effectiveness (Lewet al., 1986a, b; Mesch Johnson and Johnson,
1989).
To date, however, these three constructs have not been studied
in conjunction, eventhough the importance of doing so has been
acknowledged (Hollensbe and Guthrie,2000). Further, most of these
studies were conducted in laboratory settings, while theexisting
field studies were mostly conducted via surveys, resulting in
rather abstractknowledge. The aim of the present study is to apply
this knowledge in practicalsettings and evaluate its practical
applicability.
In this paper, the following topics will be addressed. We will
start with a furtherelaboration on the prescriptive model, which
will be followed by a description of thecase-study setting.
Subsequently we will discuss the case-study method for thein-depth
study of three top management teams. Then we will report on
theinterdependence analysis, and evaluate the prescriptive model.
The paper will beconcluded with a discussion.
Figure 1.Proposed prescriptive
model on effectivecombinations of task, goal
and rewardinterdependence and thecontent of goal and pay
indicators
Teamperformancemanagement
99
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
A prescriptive modelThe general idea behind the model as
depicted in Figure 1 is that if (a) the goal and payindicators fit
together in terms of content, and if (b) the levels of goal and
rewardinterdependence fit together in terms of signals they give on
desired behaviour, and if(c) the levels of goal and reward
interdependence fit with the level of taskinterdependence, this
combination is more effective than combinations in which one ormore
of these fit requirements are not satisfied. Thus, the extent of
fit or misfit betweenthe elements as depicted in Figure 1
constitutes the independent variables in thismodel, whereas the
dependent variable refers to effects of variation on
theseindependent variables, i.e. the effectiveness of the
combination of the three elements.This combination may impact many
elements of effectiveness, such as team outcomes(e.g. performance)
and team interaction processes (e.g. cooperation and
competition)(Hackman, 1987; McGrath, 1964). In this study we are
primarily interested incooperation as a team interaction process
and motivation as an outcome. We expect thetwo different types of
fit to be associated with different elements of effectiveness,
henceour specific interest in these variables.
First, we expect content fit the fit that is concerned with a
clear linkage between goaland reward attainment to be specifically
associated with motivation. The theoreticalbackground for this idea
can be found in the expectancy theory (Vroom, 1964). Accordingto
this theory, content fit is the situation in which the attainment
of a performance goal isinstrumental for the attainment of a
(valued) reward. This, together with the expectancythat a persons
efforts will lead to goal attainment, will positively impact
motivation.Second, we expect a fit between the interdependence
relationships to be specificallyassociated with cooperation. In
line with Deutsch theory of cooperation and competition(Deutsch,
1949a, b), the underlying idea is that teamwork is always
associated withinteraction patterns (e.g. cooperation,
competition), and that performance goals and apay-for-performance
plan should give signals as to the desired interaction
pattern(cooperation). In situations of a fit between the
interdependence constructs, both theperformance goals and
pay-for-performance plan stimulate cooperation by creating
positiveinterdependence. In situations of misfit, the performance
goals and pay-for-performanceplan give mixed signals or
consistently stimulate competition, which in turn negativelyimpacts
cooperation. Thus, we expect that both types of fit positively
influence elements ofeffectiveness, and that their effects are
cumulative, i.e. the more fit, the better. In theremainder of this
section, the various elements of the model will be further
specified.
Task interdependence reflects the extent to which team members
have to exchangeinformation and/or means for the completion of
their contribution to the team task (e.g.,Thompson, 1967). Task
interdependence can be classified as low or high. In situations
ofhigh task interdependence, team members critically depend on
information and/or meansfrom each other for the completion of
(their part of) the team task, i.e. absence of thisinformation
and/or means hampers the achievement of satisfying levels of
taskcompletion or renders task completion impossible. Think, for
instance, of the members ofa surgical team where the surgeon cannot
complete his/her task without information onthe patients condition
and the anaesthesia administered by the anaesthetist. In
situationsof low task interdependence, team members hardly depend
on information or means fromeach other for the completion of (their
part of) the team task. While team members mayexchange information
or means within the scope of task completion, the dependence onthis
information and/or means is not critical: even without this
information and meanssatisfying levels of task performance can be
achieved. For instance, telemarketers
PR35,1
100
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
exchange information about specific advice or tricks on how to
deal with customers,although they can also complete their task
without this information.
Goal interdependence is the interdependence created by the way
in which theattainment of performance goals of a team member is
related to the attainment ofperformance goals by other team
members. Goal interdependence can vary fromnegative to positive. In
the case of positive goal interdependence, the attainment ofones
individual goals is positively influenced by the attainment of
goals by other teammembers. Logically, team goals create positive
goal interdependence as well, becauseindividual team members are
positively interdependent on one another for theattainment of this
goal. In the case of negative goal interdependence, the attainment
ofones individual goals is negatively influenced by the attainment
of goals by other teammembers. An example of neutral
interdependence is a team of organizational trainers,where all team
members are assigned the goal of giving a pre-specified number
of(hours) of reference courses per year. In this case, attainment
of the pre-specifiednumber of hours by one team member does not
influence the attainment of the samegoal by other team members.
Reward interdependence is the interdependence created by the way
in which the payindicators of a team member are related to the pay
indicators of other team members.The type of reward interdependence
is determined by the type of pay-for-performanceplan, which means
that changes in the level of reward interdependence can be realized
bychanging the pay-for-performance plan. Reward interdependence can
vary from negativeto positive. An example of negative reward
interdependence is a ranking system, inwhich a reward is based on
the relative performance of the team members: highperformance by
other team members reduces ones own possibilities to earn a
reward.An example of positive reward interdependence is a case in
which a reward is basedupon team performance: if my co-team members
receive a reward, I will receive a rewardas well. However, positive
reward interdependence can also result from individual
payindicators that are positively related. An example is a
management team where the HRmanager is assigned the goal of
recruiting more and better-equipped sales personnel.The sales
manager on the other hand, is assigned the goal to increase market
share andcustomer satisfaction. In this situation, the indicators
of the HR and sales manager arepositively related: if the HR
manager attains his/her goals, it will positively impact
theattainment of the sales managers goals, because better sales
personnel facilitates anincrease in market share and customer
satisfaction. Finally, neutral rewardinterdependence can be created
via individual level pay indicators in a similar way asdescribed in
the example under goal interdependence.
From the above discussion it appears that the actual type of
goal or rewardinterdependence that is created depends on the
combination and type of individualand/or team level indicators.
Figure 2 shows the different types of goal and reward
interdependence that canresult from individual and/or team level
indicators goal and pay indicators[1]. First,team level indicators
result in positive interdependence, for reasons mentioned
earlier.
Second, individual level indicators can result in different
types of interdependence,depending on the relation between the goal
or pay indicators of individual teammembers. Examples of neutrally
and positively related indicators have been givenabove. An example
of individual indicators that create negative interdependence is
amanagement team where the financial manager is assigned the goal
to cut down theoverhead costs, while the marketing manager is
assigned the goal to increase brand
Teamperformancemanagement
101
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
recognition by expanding the marketing activities. The latter
goal conflicts with theformer: expanding marketing activities
brings along overhead costs, therebynegatively influencing the
financial managers goal attainment. Thus, depending onthe way in
which different individual indicators are related, they create
different typesof interdependence.
The third possibility is that indicators exist both at the
individual and the teamlevel, which can result in combinations of
the above-mentioned forms ofinterdependence, i.e. positive
interdependence created via the team level indicator,and neutral,
positive or negative interdependence created via the individual
levelindicators. We will classify these combinations later on.
Content fit is the extent to which goal and pay indicators are
related in terms ofcontent, where content is defined as the
attribute to which the indicators refer. Anexample of goal and pay
indicators that refer to different attributes is a situation in
whichthe performance goal is to increase production quality and
where the pay indicators focuson decreasing production costs and
increasing output. In this situation, the performancegoal refers to
quality whereas the pay indicators refer to costs and quantity.
Fit between task, goal and reward interdependence is the
situation where performancegoals and a create the same type of
interdependence (i.e. the same direction), therebygiving consistent
signals with respect to the desired behaviour (cooperation,
competition,or individualistic). In this situation, the signals are
consistent with the level of taskinterdependence and the need to
cooperate resulting from this: high task interdependencerequires
cooperative behaviour, while low task interdependence requires a
mixture ofcooperative and individualistic behaviour. Thus, the
situation in which goal and rewardinterdependence are both negative
is not considered as fit, since it conflicts with a basicpremise of
a team, namely the presence of a minimum of task interdependence.
Becausethe present study was conducted with top management teams of
an IT company thatwere characterized by high levels of task
interdependence, the evaluation of the proposedmodel was limited to
high task interdependent settings.
Figure 2.Causes of different typesof goal and
rewardinterdependence
PR35,1
102
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
Case descriptionSettingThe setting of this study is a global
IT-services company, which we will call Itech. Theorganization is
structured around 32 country organizations that are headed by
amanagement team. In situations where local market sizes are small,
which is particularlythe case in countries only recently entered by
Itech, one management team (MT) headsseveral country organizations.
The board of Itech consists of three members, who arealso member of
a local MT. An additional division is the Global Clients division,
which isresponsible for all global clients of Itech. Most global
clients are multinationals withoffices across the world that cannot
be served effectively by the country organizations.
TeamsThree management teams participated in the first part of
the case study: The GlobalClients MT (n 7; located in Brussels,
Belgium), which headed the Global Clients servicedivision (about
350 employees); the Netherlands MT (n 7), which headed the
singlelargest country organization (about 9,000 employees); and the
UK MT (n 8), whichheaded a medium-sized country organization (about
2,000 employees).
Performance goals and pay-for-performance planTeam-level
performance goals were formulated annually by the board of
directors andcommunicated to the MTs before the beginning of a new
year. The performance goalswere specific and differed substantially
per team. Examples of the sort of performancegoals that were
formulated are: 1. successfully complete merger; 2. create strong
businessrelationships with Latin-America; and 3. recruitment (515
new hires in 2001). After thistop-down communication of the
performance goals, teams were free to implement, furtherspecify or
cascade the performance goals in accordance with their own
view.
The pay-for-performance plan of Itech, which applied to all
members of themanagement teams as described above (n 168, divided
over 20 teams), was divided intothree organisational levels. At the
organizational level (25 per cent), one indicator wasformulated:
earnings per share. At the team level (50 per cent), approximately
threefinancial indicators were set, generally with equal weights,
like cash flow, net income,budget attainment, and revenue. The
exact set of indicators and their relative weightrelated indicators
applied to team members responsible for sales or delivery,
whereascost-related indicators applied to team members in a support
function. The individuallevel indicators (25 per cent) generally
were non-financial, for instance set-up an effectivemarketing and
communication organization and execute a restructuring program.
The average bonus amounted to approximately 40 per cent of an
annual salary; themaximum bonus attainable was approximately 80 per
cent of an annual salary. TheMTs were free to fill in the
individual level indicators at their own discretion.
MethodSelection criteriaThe three teams that participated in
this study were selected on theoretical grounds.According to Yin
(1994) and Eisenhardt (1989) such a theoretical selection approach
ispreferable over random sampling (which fits better in a sampling
logic approach),since this allows researchers to choose cases that
are likely to replicate or extend the
Teamperformancemanagement
103
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
emergent theory. As such, a theoretical sampling approach is in
line with thereplication logic that underlies case-study research
(Yin, 1994).
The first criterion was that there should be differences between
the types oforganizations an MT is heading. Second, the teams were
selected such that differencesin interpretation (e.g. filling-in of
the individual level indicators) of thepay-for-performance plan
could be expected. The rationale for the first criterion wasto
acquire a broader picture of the MTs within the organization and to
explore possibledifferences across these types of teams. The
rationale for the second criterion was tofind differences in the
types of reward interdependence.
Information on these two criteria was collected via the
organizations key informant,who was also closely involved in the
final selection of the teams. Practical criteria, suchas
willingness of MTs to participate and geographical proximity,
played a role as well.
Data collectionThe data for this study were collected via
interviews (n 21), documents describing theorganization, the
performance goals and the pay-for-performance plan (such as
annualreports and internal publications), and discussions with the
Vice President of CorporateHuman Resources and with the Director of
Compensation and Benefits who acted as keyinformants. The
interviews were the main data source for this case study.
Theinformation collected via documents and discussions mainly
provided backgroundinformation and were primarily used to verify
information collected via interviews.
The interviews followed the structure of the prescriptive model,
discussing the work,the performance goals and pay-for-performance
plan, the fit between performance goalsand pay-for-performance
plan, and effectiveness. The interview schemes were sent to
theinterviewees a couple of days before the interview, with the
request to go through thescheme in advance. The interviews took one
or two hours and the response rate was 86per cent for The
Netherlands MT, and 100 per cent for the other two MTs.
Measurement of the constructsTask interdependence was measured
by asking respondents to indicate their keyindividual tasks and the
extent and cruciality to which completion of these tasksdepended on
information and/or means from other team members.
To measure goal and reward interdependence, we first developed a
classificationframework for the different combinations of team
and/or individual level goal and payindicators as depicted in
Figure 2. This framework is shown in Table I. As can be
seen,combinations 1-4 have in common that a team level indicator is
present, while incombinations 5-8 a team level indicator is absent.
Combination 3 is special in the sensethat in this situation the
team and individual level indicators create opposing types
ofinterdependence. Although one could argue that these two
interdependencerelationships cancel each other out, resulting in
neutral interdependence, we classifythis combination as slightly
negative. The reason for doing so is that we expect thenegatively
related individual level indicators to be more dominant than the
positiveinterdependent team level indicator, which is also in line
with earlier studies. Forinstance, in line with Miller and Hamblin
(1963), Rosenbaum et al. (1980) evaluated theimpact of different
mixtures of negatively and positively interdependent rewards
ongroup productivity and processes. They concluded that even a
modicum of acompetitive (negatively interdependent, HvV et al.)
reward led to lowered efficiencyand productivity.
PR35,1
104
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
Starting point for the measurement of goal interdependence were
the team-levelperformance goals. First, interviewees were asked
whether these performance goals werecascaded to the individual
level, or whether other individual performance goals
wereformulated. Second, they were asked to indicate how the
attainment of the individualperformance goals was influenced by the
different goals of colleagues within the team(positively,
neutrally, or negatively). This yielded rather complex information,
becausesome individual performance goals were positively related,
while others were negativelyor unrelated. Thus, for one team
member, various types of goal interdependence existed.To
incorporate this information, the response of each interviewee on
how the individuallevel performance goals were related to each
other was placed in one of three categories,i.e. (predominantly)
positive, neutral or negative, by two of the authors. The raters
madetheir assessments independently of each other. In this rating
process more weight wasattached to negatively related indicators,
since we expected these relationships to bemore dominant than
neutral or positive relationships (Rosenbaum et al., 1980).
Theinter-rater reliability was computed using Cohens k (Cohen,
1960), and was 0.57. In linewith the benchmark ratings of Landis
and Koch (1977), we interpreted this as moderateagreement on the
goal interdependence measure. Subsequently, the raters discussed
thecases on which they initially disagreed, until full agreement
was achieved. Thisinformation, in combination with the knowledge
that explicit team level performancegoals were present, facilitated
a classification of goal interdependence following theframework as
presented in Table I.
A similar approach was used for the measurement of reward
interdependence.As team level indicators were present, the question
was how the individual levelindicators were related (to each
another). Interviewees were asked to indicate howtheir individual
level pay indicators were related to the pay indicators of
otherteam members (positively, neutrally, or negatively). Again,
this yielded rathercomplex information, because some individual pay
indicators were positivelyrelated, while others were negatively or
not related, i.e. a single team member wasconfronted with different
types of reward interdependence. The response of eachinterviewee
was placed in one of three categories, i.e. (predominantly)
positive,neutral or negative, by two independent raters. Again,
more weight was attachedto negatively related indicators, since we
expected these relationships to be moredominant. The inter-rater
reliability coefficient k was 0.76, which is interpreted
assubstantial agreement among the raters (Landis and Koch, 1977).
After this, the
CombinationTeam level goal orpay indicator
Relationship between individuallevel goal or pay indicators
Classification of goalor reward interdependence
1 Yes Positive 2 Yes Neutral 3 Yes Negative 4 Yes Absent 5 No
Positive 6 No Neutral 07 No Negative 8 No Absent 0 (no
interdependence)
Notes: positive; slightly positive; 0 neutral; - slightly
negative; negative
Table I.Combinations of team
and individual level goalor pay indicators and a
classification of the typeof goal or reward
interdependence theycreate
Teamperformancemanagement
105
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
raters discussed the cases on which they initially disagreed,
until full agreementwas achieved. This information, in turn,
facilitated a classification of the type ofreward interdependence
following the framework described in Table I.
To measure the content fit, respondents were asked to indicate
the influence of theattainment of individual performance goals on
the attainment of their pay indicators.Again, the responses did not
always allow for a univocal interpretation, because ofdifferent
relationships between the various goal and pay indicators.
Therefore, theresponse of each interviewee was placed in one of
three categories, i.e. (predominantly)high, modest or low, by two
independent raters. The k of 0.67 indicated a substantialagreement
among the raters (Landis and Koch, 1977). Again, the raters
discussed thecases they initially disagreed on, until full
agreement was achieved.
The effect variables were determined and formulated by the
organization in whichthe case study was carried out, with the
objective to evaluate their intervention in thepay-for-performance
plan. Nine effect variables were included in this study, whichwere
measured via nine single-item questions on a five-point Likert
scale. These nineitems included the variables that were of
particular interest to us (i.e. cooperation andmotivation).
Cooperation, for example, was covered by the question To what
extentdoes the pay-for-performance plan drive individual
contributions into team play?. Anexample of a question that was
included but falls outside our primary field of interest isTo what
extent does the pay-for-performance plan increase the
attractiveness of Itechin the marketplace?. We consider these items
to be a valid operationalization of theeffectiveness of the
combination of task, goal and reward interdependence and thecontent
of goal and pay indicators, because the scores on these
effectiveness criteriawill not be determined by the
pay-for-performance plan in isolation, but by the way inwhich it
interacts with the performance goals and type of team work, i.e. by
the extentof fit or misfit.
An interdependence analysisThis section reports on the results
of an interdependence analysis as conducted in thethree MTs we
studied.
Team A: Global Clients MTTask interdependence. Team members
within the Global Clients MT[2] needed toexchange information for
the satisfactory completion of their individual contribution tothe
team task, but the intensity of information exchange differed per
team member. Onthe one hand, the general manager and managers in
the support functions (finance andHR) indicated that the
information exchange with other team members was intense
andcritical to task completion. For instance, the financial manager
needed (financial)information from other team members, and
especially from the line managers (delivery,technology support and
competencies and alliances) for the completion of his/her work.On
the other hand, the line managers, who were responsible for the
actual delivery andsale of products or services, indicated that the
information exchange was less intensivebut usually critical for the
cases in which it took place. An example would be themanager of
technology support, who was responsible for delivery of support
serviceson products delivered by the Global Clients MT. One of
his/her tasks was to establish acompetence centre on a specific
software application. To realize this, information andmeans (human
capacity) was needed from among others the delivery managerswithin
the team. Based on the above discussion, from which it appeared
that team
PR35,1
106
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
members were critically dependent on information to complete
their task, we concludethat the task interdependence within team A
is high.
Goal interdependence. The types of goal interdependence in this
team are depicted inTable II. The results are based on a
combination of the knowledge that team level goalindicators were
present (verified in the interviews) with the rating results on
therelationship between the individual level goal indicators. For
example, the columnpositive results from a combination of team
level indicators and (predominantly)positively related individual
level indicators[3].It can be gathered from the Table thatthere was
a mixture of positive and slightly positive goal interdependence
within teamA, which can be traced back to different types of
relationships among individual levelindicators. Some respondents
indicated that the individual level goal indicators wereunrelated,
for instance: [individual, HvV] performance goals relate to each
other like1 1 2; results are merely cumulative and do not reinforce
each another. Thisrespondent, for example, was classified as
slightly positively goal interdependent,resulting from the
combination of neutrally related individual level goal indicators
andteam goals (see Table I). Others indicated that goal attainment
of colleagues positivelyinfluenced their own goal indicators,
although direct relationships were absent.
Reward interdependence. The types of reward interdependence
present in this teamare depicted in Table II and were established
analogously to the classification of goalinterdependence, i.e. by
relating the information on the presence of team level
payindicators and the relationship among individual level pay
indicators to theclassification framework. For example, the column
slightly negative is based on thefinding that team level indicators
were present and that the individual level indicatorswere
(predominantly) negatively related. From the Table it appears that
the levels ofreward interdependence were mixed and varied between
slightly positive and slightlynegative. An important source of the
slightly negative interdependence was the conflict
n Positivea Slightly positiveb Slightly negativec
Team AGoal interdependence 7 3 3 1Reward interdependence 7 0 4
3
n High Modest LowContent fit 7 2 2 3Team BGoal interdependence 4
1 2 1Reward interdependence 5 0 3 2
n High Modest LowContent fit 4 1 2 1Team CGoal interdependence 7
4 2 1Reward interdependence 7 4 2 1
n High Modest LowContent fit 7 4 3 0
Notes: aCreated via a team level indicator in combination with
positively related individual levelindicators; bcreated via a team
level indicator in combination with unrelated individual
levelindicators; ccreated via a team level indicator in combination
with negatively related individual levelindicators
Table II.Goal and reward
interdependence andcontent fit
Teamperformancemanagement
107
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
between the individual level indicators of the manager alliances
and the individuallevel indicators of the two delivery managers and
the technology support manager.Among other things, all four
managers had to Attain the business plan for the period2001-2003 as
an individual level indicator. This applied to each individual
managersbusiness. For the alliance manager, it boiled down to
setting up alliances andpartnerships with other companies. Such
partnerships negatively influenced thebusinesses of the other team
members with the Global Clients MT, since it stimulatedthe
outsourcing of services that were supposed to be delivered by the
deliverymanagers. In a sense, it undermined their day-to-day
activities by bringing incompetitors products that competed with
their own products.
The team members that were classified as slightly negatively
rewardinterdependent were the delivery managers and the technology
support manager.Interestingly, the manager of alliances was not
classified in this category, as heindicated not to be negatively
interdependent on other team members for theattainment of his/her
individual level pay indicators. The team member who wasclassified
as slightly negative goal interdependent was one of the
threeabove-mentioned managers.
Fit between the interdependence constructs. The types of
interdependence as presentin this team are depicted in Figure 3.
Relating these findings to the definition of fit, weconclude that
there is a misfit between the three constructs. First, goal and
rewardinterdependence have partly conflicting directions.
Furthermore, we see that theslightly negative reward
interdependence relationships in particular conflict with thelevel
of task interdependence.
Content fit. Table II also shows the rating results for the
responses on the level ofcontent fit. The findings are mixed,
running from high to low. As a result, the level ofcontent fit
cannot be univocally classified. In addition, no clear pattern
could be foundbetween the type of content fit and, for example, the
type of function a team member holds.
A possible reason for the five cases of modest and low content
fit (2 3), is the loweffort that was put into the development of
individual level pay indicators, which wasillustrated by the fact
that there was hardly any differentiation between the individualpay
indicators of the various team members. As a result, the
opportunity was notutilized to establish a link between the
performance goals and the pay-for-performanceplan via the
individual level pay indicators.
Propositions on the effects of fit and misfit. Comparing the
observed pattern with thedefinition of fit, leads us to the
following expectations. Concerning the interdependenceconstructs,
it was the negative reward interdependence in particular that
conflictedwith the (slightly) positive goal interdependence and the
high task interdependence.Where the last two types of
interdependence required and stimulated cooperation, in
Figure 3.Fit between theinterdependenceconstructs for Team A
PR35,1
108
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
half of the cases the pay-for-performance plan created a type of
interdependence thatstimulated competition among team members.
Following the model, we expect thiscombination to be ineffective:
it will not stimulate cooperation among team membersnor motivate
team members. In addition, the mixed levels of content fit will
notcontribute to the effectiveness of the combination either.
Team B: The Netherlands MTThe findings for Team B are highly
similar to the findings of Team A (see Table II). Asa result, the
propositions concerning the effectiveness of the constellation
ofinterdependence constructs and content fit are similar to the
propositions for TeamA. To avoid redundancy, the discussion of Team
B will be confined to an example ofhow the reward interdependence
was actually created in this team.
Reward interdependence. None of the interviewees indicated that
the individual levelpay indicators of team members were positively
related, resulting in the absence ofpositive reward interdependence
(see Table II). The set of individual level payindicators of the
legal manager provides a good example of a situation in which
theindividual level pay indicators were unrelated. This manager had
two individual payindicators: improve contract management within
Itech; and align the legal processeswithin two specific divisions
of Itech. The manager indicated that the individual levelpay
indicators of the other team members (13 in total) were mostly
unrelated to his/herown indicators: ten out of 13 indicators were
unrelated to the attainment of his/her firstpay indicator (contract
management), the other three were positively related. Only oneout
of the 13 other individual level indicators was positively related
to the attainmentof his/her second pay indicator (align legal
processes), while the scores on the other 12individual level pay
indicators did not affect his/her second indicator.
A possible reason for the presence of unrelated individual level
indicators is thatthey were formulated irrespective of each other,
leaving not much room to gear theindividual indicators of team
members. During the interviews, it became apparent thatteam members
did not know one anothers individual level indicators.
Beingconfronted with the individual level indicators of colleagues
sometimes resulted inamazement on the conflicting nature of these
indicators with their own indicators.
Team C: UK MTTask interdependence. The patterns of information
exchange among team members inthe UK MT were highly similar to the
ones described for the Global Clients MT, both interms of the
intensity and cruciality, and in terms of the type of information
that wasexchanged. The fact that the UK MT was split into two parts
that operated fromdifferent geographical locations (about 200 km
apart), did not significantly affect theinformation exchange, nor
did it result in two sub-teams. We therefore classify the
taskinterdependence in this team as high.
Goal interdependence. Table II shows that the goal
interdependence was (slightly)positive. An example of positively
related individual level goal indicators are theperformance goals
in the customer quadrant of the balanced score cards (BSCs) of
theHR and legal manager. Among other things, the HR manager was
assigned theperformance goals to: improve and align HR processes to
the needs of the business; andrationalise the HR systems (e.g.
improve transparency by removing old systems andprocedures). The
legal manager, on the other hand, was assigned the goals to:
prepare a
Teamperformancemanagement
109
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
new contract of employment within the timescale agreed with the
HR manager; andliaise with the HR manager to ensure that new
starters attending introduction coursesare made aware of HR
policies and procedures. Thus the legal aspects of theperformance
goals of the HR manager were reflected in the goals of the legal
manager,resulting in positive interdependence among these managers:
attainment of theperformance goals by the legal manager positively
influenced the performance goals ofthe HR manager and vice
versa.
One of the reasons for these rather univocal findings is that UK
MT developed aBSC (Kaplan and Norton, 1996) for the team as a
whole, based on the performancegoals as assigned by the board, and
additionally developed a BSC for each individualteam member. These
BSCs were developed on a participative basis in several
teamsessions. A direct result from these sessions was that
inconsistencies among individualperformance goals came to light and
could be resolved. Or, as one interviewee put it:The reason why
there are no conflicting goals is that the BSCs are made on
aparticipative basis (i.e. via team decision making, HvV et al.),
so eventually conflictingobjectives are immediately detected.
Other interviewees downplayed this somewhat by stating that
there was alwayssome conflict between individual performance goals.
However, team members wereaware that full consistency was hard to
achieve, as appeared from the fact that theperformance goals
formulated in the individual BSCs were agreed upon via
consensusamong all team members.
Reward interdependence. As can be gathered from Table II, the
type of rewardinterdependence is similar to the type of goal
interdependence, i.e. predominantly(slightly) positive. This is a
direct result of the fact that the non-financial elements ofthe BSC
of each individual team member were translated one-to-one into
individuallevel pay indicators. Thus, the individual performance
goals and individual payindicators were highly similar. The only
difference was that the financial component ofthe BSC (accounting
for 25 per cent) was not used as an individual level pay
indicator,because the pay plan already contained a considerable
financial component via the(financial) team level pay indicators.
However, this difference did not result in differenttypes of reward
interdependence.
Fit between interdependence constructs. Figure 4 gives the
predominant types ofinterdependence present in team C. Relating
these to the definition of fit, we classifythis situation as fit.
First, the performance goals and pay-for-performance plan
createdthe same types (i.e. direction) of interdependence.
Furthermore, the types of goal andreward interdependence were in
line with the level of task interdependence.
Content fit. The findings in Table II indicate that the content
fit betweenperformance goals and pay indicators was modest to high.
Team members indicatedthat the content fit was mainly established
by the presence of the (non-financial)
Figure 4.Fit between theinterdependenceconstructs for Team C
PR35,1
110
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
individual performance goals in the pay-for-performance plan. In
addition, theyindicated that the team level performance goals were
positively related to the teamlevel pay indicators. Other team
members subscribed to the above viewpoint, with theremark that
there was some imbalance between the performance goals and
thepay-for-performance plan, i.e. the BSCs consisted of 25 per cent
financial indicators,while the pay-for-performance plan consisted
of 75 per cent financial indicators. Inthese cases, the content fit
was classified as modest.
Propositions on the effects of fit and misfit. It is expected
that the combination of task,goal and reward interdependence, and
the goal and pay indicators will be effective in thissituation.
First, the levels of task, goal and reward interdependence fit
together: high taskinterdependence required cooperation among team
members, which was supported byperformance goals and a
pay-for-performance plan that created positiveinterdependence,
thereby stimulating cooperation as well. In addition, the
performancegoals and pay-for-performance plan fit in terms of
content, thus establishing a linkbetween goal attainment and the
attainment of a pay-for-performance bonus.
Evaluation of the prescriptive modelFrom the discussion above,
it appears that only team C satisfied the modelspropositions on fit
between the interdependence constructs and content fit, while teamA
and B did not satisfy all propositions. Therefore we expect team C
to be moreeffective than the other two teams.
Table III shows the effects of fit and misfit on the nine
dependent variables.Non-parametric statistical tests were used to
evaluate the differences between the teams.First, the
Kruskal-Wallis Chi-square was computed for each variable to test
whether therewere significant differences among teams. If this was
the case, a Mann-Whitney U test wasconducted to investigate which
pairs of teams had significantly different mean scores.Concerning
the effect variables in which we were primarily interested in from
a theoreticalpoint of view (i.e. cooperation (covered by item 1)
and motivation (covered by item 3)),significant differences among
teams were found on cooperation while no significant results
Aa
(n 8)B
(n 6)C
(n 7)Kruskal-Wallis
x2
The pay-for-performance plan: Mean score1. Drives individual
contributions into team playb, c 2.75 3.50 4.14 5.94 *
2. Enhances attainment of financial resultsb, c 3.63 3.83 4.57
4.88 *
3. Increases shareholder value 3.43 4.00 4.43 3.014. Motivates
people 3.00 3.00 3.86 3.595. Retains people 2.88 2.17 3.14 3.756.
Rewards in a fair way 2.86 2.83 3.57 2.817. Increases attraction of
IT in the marketplace 2.50 2.67 3.14 1.198. Facilitates IT to
become a world class player with
excellent customer satisfaction 2.38 2.67 2.86 1.179.
Facilitates IT to become a world class player withexcellent
employee satisfaction 2.38 2.80 3.14 2.37
Notes: an 7 for variables 3, 7 and 8; bMann-Whitney U test
yielded statistically significantdifferences ( p , 0.05) for teams
A and C; cMann-Whitney U test yielded statistically
significantdifferences ( p , 0.10) for teams B and C; *p , 0.10
Table III.Effects of fit and misfit onnine dependent
variables
Teamperformancemanagement
111
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
were found on motivation. With regard to cooperation, team C
scored significantly higherthan team A (U (n 15 10:00, p , 0:05)
and team B (U n 13 10:50, p , 0:10).This finding, indicating that
the combination of fit as present in team C enhanced
cooperationmore than the combinations of misfit of the other two
teams, is in line with the proposition.
With regard to motivation, we notice that the findings point in
the hypothesizeddirection (Team C scores 0.86 scale point higher
than teams A and B). A possible reasonfor not finding statistical
differences may be a lack of statistical power. All in all,
weinterpret these findings as some first support for the
proposition that combinations of fit(Team C) are more effective
than combinations of misfit (Teams A and B).
Concerning the other variables that were not explicitly
addressed in thepropositions, significant differences were found on
the variable financialperformance (covered by item 2). Again, team
C scored significantly higher thanteam A (U (n 15 12:50, p , 0:10)
and team B (U (n 13 9:50, p , 0:10).Although no explicit
propositions were made concerning the extent to which
financialperformance was enhanced, this finding fits to the general
idea that a situation of fit ismore effective than a situations of
misfit. This idea is further supported by the fact thatTeam C
scores higher on all dependent variables than Teams A and B even
thoughthese differences are not statistically significant. Two
simple sign tests comparingTeams A and C, and Teams B and C tell us
that the probability of finding ninedifferences in the same
direction by chance is p , 0:01, i.e. the hypothesis that themean
scores of Teams A and B are equal to Team C is rejected.
In addition to the findings presented in Table III, the
interviews provided someanecdotal evidence for the notion that the
above findings are especially related to fitand misfit between
different types of interdependence.
In the Netherlands MT (team B), for example, one respondent
noted that the currentpay plan did not reflect the interdependence
relationships that were present within theteam. The respondents
suggestion was to first analyse these relationships and
thenformulate pay indicators for the different sets of team members
that depend on oneanother for the completion of a specific job. An
example would be the managers ofsales, marketing and HR, who
together were responsible for the development of amarketing
strategy (i.e. sales for market information and HR for the new
hires to set upa marketing department). According to the
interviewee, an important reason for thelow effectiveness was that
these sets of interdependent team members were notrecognized by the
current pay-for-performance plan, while they should be the basis
onwhich pay indicators should be formulated.
In contrast, the respondents in team C indicated that one of the
main reasons whytheir situation was effective was that the
pay-for-performance plan reflected theseinterdependence
relationships between team members. For example, the Legalmanager
had to work together with the financial manager in restructuring
the systemsfor monthly reporting to headquarters. This task
interdependence relationship wasreflected in the BSC by assigning
both team members the performance goal (andthereby automatically
the pay indicator) to complete this restructuring within one
year.
DiscussionThe aim of this study was to apply existing knowledge
on effective combinations ofinterdependence constructs mainly
generated via laboratory experiments andsurveys to the practical
issue of designing effective pay-for-performance plans for
PR35,1
112
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
teams. To this end, we integrated this knowledge into a
prescriptive model, which wasevaluated in three top management
teams. The key proposition that underlies thismodel is that
situations in which a pay-for-performance plan fits with the
performancegoals as well as with the team task are more effective
than situations of misfit. Thesefit relationships are
operationalized by means of the interdependencies that result
fromthese three elements, i.e. task, goal and reward
interdependence, and by the content ofthe goal and pay indicators.
This study yielded at least three interesting findings.
First, some preliminary support is found for the proposition
that situations of fit aremore effective than situations of misfit.
Significant differences were found betweenTeam C (fit) and Teams A
and B (misfit) on the effectiveness criteria cooperation
andfinancial performance. Most differences were found at p , 0:10,
which we consider adefendable significance level given the small
sample size. However, these findingsshould be interpreted with some
reservation for at least two reasons. Contrary to whatwe expected,
no significant differences were found on the effectiveness
criterion ofmotivation. Also, the effectiveness criteria were
measured with single items therebylowering the reliability of these
measures.
Second, this study showed that negative reward interdependence
relationshipsplayed a dominant role in the combinations of misfit,
i.e. the misfits of Teams A and Bwere both associated with negative
reward interdependence. Therefore, we concludethat negative reward
interdependence relationships should be avoided
inpay-for-performance plans for teams, which is in line with
earlier findings fromexperimental studies (e.g. Miller and
Hamblin;, 1963; Rosenbaum et al., 1980; Thurkowet al., 2000). An
important and robust contribution of this study lies in the fact
that it isthe first study to investigate specific sources of reward
interdependence relationshipsin practical settings. It shows that
negative reward interdependence not only stemsfrom the distribution
method that is used, which has been the traditional vehiclethrough
which reward interdependence was manipulated in former studies, but
alsofrom the individual level pay indicators. For instance, we
found that the negativereward interdependence was not created on
purpose (e.g. via a ranking system), butthat it was a result of a
negligent design process, i.e. the individual level pay
indicatorswere developed irrespective of each other, without
checking for the presence ofconflicting indicators (see, for
example, Team B), which resulted in negative rewardinterdependence.
On the other hand, Team C provided a good example of howindividual
indicators can be developed such that they do not result in
negative rewardinterdependence. In this team, the individual level
pay indicators were developedcooperatively (i.e. via the
development of individual performance goals that served asan input
for the pay-for-performance plan) to make sure that the indicators
did notconflict and reflected, where possible, the task
interdependence relationships withinthe team. Thus, awareness is
needed in designing individual level pay indicators forteam
members. Managerial influence can play an important role here, by
coordinatingan integral design via, for instance, team decision
making.
Third, this study provided insight into the complexity of
interdependence relationshipsin real-life settings and demonstrated
the difficulty of applying existing knowledgeone-on-one in
practice. Van der Vegt et al. (2000) and Van der Vegt and Van der
Vliert(2002) have already demonstrated that the levels of task
interdependence can vary acrossteam members. The present study
shows that not only task interdependence may varyacross team
members, but that the type of goal and reward interdependence may
varyacross team members as well. Moreover, this study shows that
even a single team
Teamperformancemanagement
113
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
member may be confronted with different types of goal and/or
reward interdependence atthe same time. These findings, in turn,
have implications for the applicability of the designrules as
incorporated in the prescriptive model, as it makes it more
difficult to univocallydetermine goal and reward interdependence of
a single team member, let alone of a teamas a whole. The practical
implication is that in heterogonous situations the
prescriptivemodel may not be applicable, and that a univocal
team-level solution for the design ofperformance goals and
pay-for-performance is not something to aim for.
In line with most existing research on interdependence
relationships, this study wasconfined to internal interdependence
relationships. However, in contrast to laboratorysettings,
real-life settings are not limited to internal interdependence
relationships (McCannand Ferry, 1979). External interdependencies
may impact the effectiveness of combinationsof task, goal and
reward interdependence as well. To date, these external
interdependencerelationships have been left largely unaddressed.
Further research is needed here toimprove our understanding of
effective combinations of interdependence relationships.
An issue that was left unresolved in this study is how the two
types of fit (i.e. fitbetween the interdependence constructs and
content fit) contribute to the effectivenessof combinations of
task, goal and reward interdependence and goal and pay
indicators.This study provided some anecdotal evidence for the
notion that it was especially thefit between the interdependence
constructs that influenced the effectiveness criteria.However, it
did not allow for an exact analysis of the separate effects,
because of thelimited number of teams in this study and because of
the case study design. Alarger-scale field-experimental design, as
applied by Wageman (1995) to evaluate theeffectiveness of different
combinations of task and reward interdependence, would bemore
appropriate here. Such a design would allow us to evaluate the
proposition as putforward in this study was that the effects of the
two types of fit are additive, which isinteresting to do as one can
imagine other relationships as well. For instance, one mayexpect a
moderating effect of content fit on the relationship between a fit
between theinterdependence constructs and effectiveness. That is,
the effects of a fit or misfitbetween the interdependence
constructs may become stronger with increasing levels ofcontent
fit, because with increasing levels of content fit the link between
performancegoals and pay indicators is more clear, thereby
accentuating the fit or misfit betweenthe interdependence
constructs. We consider this an interesting direction for
futureresearch to disentangle the separate effects of different
types of fit, with boththeoretical and practical relevance.
Finally, despite the drawbacks that are associated with the
design of the presentstudy as regards the generalizability and
causal inferences that can be made, thisresearch approach
facilitated a very detailed analysis of interdependence
relationshipsin practical settings.
Notes
1. We realize that performance goal indicators and
pay-for-performance indicators can be definedat many other
organizational levels as well. However, in this study we are
primarily interestedin goal and pay indicators at the team and
individual level, because these indicators have adirect impact on
the type of goal and reward interdependence among team members.
2. In the discussion of Team A we will talk about the
classification procedure of theinterdependence constructs in
greater detail. This procedure which was also used for theother
teams will not be repeated in the discussions of team B and C.
PR35,1
114
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
3. Please note that the absence of neutral and negative forms of
goal and rewardinterdependence is due to the presence of team level
goal and pay indicators, which makethese forms of interdependence
inapplicable here (see Table I).
References
Cohen, J. (1960), A coefficient of agreement for nominal scales,
Educational and PsychologicalMeasurement, Vol. 20 No. 1, pp.
37-46.
Deutsch, M. (1949a), A theory of co-operation and competition,
Human Relations, Vol. 2, pp. 129-52.
Deutsch, M. (1949b), An experimental study of the effects of
co-operation and competition upongroup processes, Human Relations,
Vol. 2, pp. 199-231.
Eisenhardt, K.M. (1989), Building theories from case study
research, Academy of ManagementReview, Vol. 14, pp. 532-50.
Gowen, C.R. (1987), Managing work group performance by
individual goals and group goals foran interdependent task, Journal
of Organizational Behavior Management, Vol. 7, pp. 5-27.
Hackman, J.R. (1987), The design of work teams, in Lorsch, W.J.
(Ed.), Handbook ofOrganizational Behavior, Prentice-Hall, Englewood
Cliffs, NJ, pp. 315-42.
Hollensbe, E.C. and Guthrie, J.P. (2000), Group
pay-for-performance plans: the role ofspontaneous goal setting,
Academy of Management Review, Vol. 25, pp. 864-72.
Ilgen, D.R. and Sheppard, L. (2001), Motivation in work teams,
in Erez, M. and Kleinbeck, U.(Eds), Work Motivation in the Context
of a Globalizing Economy, Lawrence ErlbaumAssociates, Mahwah, NJ,
pp. 169-79.
Kaplan, R.S. and Norton, D.P. (1996), The Balanced Score Card:
Translating Strategy Into Action,Harvard Business School Press,
Boston, MA.
Kozlowski, S.W.J. and Bell, B.S. (2003), Workgroups and teams in
organizations, in Borman, W.C.,Ilgen, D.R. and Klimoski, R.J.
(Eds), Comprehensive Handbook of Psychology, Industrial
andOrganizational Psychology, Vol. 12, Wiley, New York, NY, pp.
333-75.
Landis, J.R. and Koch, G.G. (1977), The measurement of observer
agreement for categoricaldata, Biometrics, Vol. 33, pp. 159-74.
Lew, M., Mesch, D., Johnson, D.W. and Johnson, R.T. (1986a),
Positive interdependence,academic and collaborative-skills group
contingencies and isolated students, AmericanEducational Research
Journal, Vol. 23 No. 3, pp. 476-88.
Lew, M., Mesch, D., Johnson, D.W. and Johnson, R.T. (1986b),
Components of cooperativelearning: effects of collaborative skills
and academic group contingencies on achievementand mainstreaming,
Contemporary Educational Psychology, Vol. 11, pp. 229-39.
McCann, J.E. and Ferry, D.L. (1979), An approach for assessing
and managing inter-unitinterdependence, Academy of Management
Review, Vol. 4, pp. 113-9.
McGrath, J.E. (1964), Social Psychology: A Brief Introduction,
Holt, Rinehart and Winston,New York, NY.
Mesch, D., Johnson, D.W. and Johnson, R. (1989), Impact of
positive interdependence and academicgroup contingencies on
achievement, The Journal of Social Psychology, Vol. 128, pp.
345-52.
Miller, L.K. and Hamblin, R.L. (1963), Interdependence,
differential rewarding, andproductivity, American Sociological
Review, Vol. 28, pp. 768-78.
Mitchell, T.R. and Silver, W.S. (1990), Individual and group
goals when workers are interdependent:effects on task strategies
and performance, Journal of Applied Psychology, Vol. 75, pp.
185-93.
Prendergast, C. (1999), The provision of incentives in firms,
Journal of Economic Literature,Vol. 37, pp. 7-63.
Teamperformancemanagement
115
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
Rosenbaum, M.E., Moore, D.L., Cotton, J.L., Cook, M.S., Hiesser,
R.A., Shovar, M.N. and Gray, M.J.(1980), Group productivity and
process: pure and mixed reward structures and taskinterdependence,
Journal of Personality and Social Psychology, Vol. 39, pp.
626-42.
Saavedra, R., Earley, P.C. and Van Dyne, L. (1993), Complex
interdependence in task-performinggroups, Journal of Applied
Psychology, Vol. 78, pp. 61-72.
Shea, G.P. and Guzzo, R.A. (1987), Groups as human resources,
Research in Personnel andHuman Resources Management, Vol. 5, pp.
323-56.
Thierry, H. (2002a), Enhancing performance through pay and
reward systems, in Sonnentag, S.(Ed.), Psychological Management of
Individual Performance, Wiley, New York, NY,pp. 325-47.
Thierry, H. (2002b), Beter Belonen in Organisaties (Better
Rewarding in Organizations),Koninklijke van Gorcum, Assen (in
Dutch).
Thompson, J.D. (1967), Organizations in Action, McGraw-Hill, New
York, NY.
Thurkow, N.M., Bailey, J.S. and Stamper, M.R. (2000), The
effects of group and individualmonetary incentives on productivity
of telephone interviewers, Journal of OrganizationalBehavior
Management, Vol. 20, pp. 3-25.
Van der Vegt, G. and Van de Vliert, E. (2002), Intragroup
interdependence and effectiveness:review and proposed directions
for theory and practice, Journal of Managerial Psychology,Vol. 17
No. 1, pp. 50-67.
Van der Vegt, G., Emans, B. and Van de Vliert, E. (2000), Team
members affective responses topatterns of intragroup
interdependence and job complexity, Journal of Management,Vol. 26,
pp. 633-55.
Vroom, V.H. (1964), Work and Motivation, McGraw-Hill, New York,
NY.
Wageman, R. and Baker, G. (1997), Incentives and cooperation:
the joint effects of task and rewardinterdependence on group
performance, Journal of Organizational Behavior, Vol. 18, pp.
139-58.
Wageman, R. (1995), Interdependence and group effectiveness,
Administrative ScienceQuarterly, Vol. 40, pp. 145-80.
Weldon, E. and Weingart, L.R. (1993), Group goals and group
performance, British Journal ofSocial Psychology, Vol. 32, pp.
307-34.
West, M.A. (1996), Preface: introducing work group psychology,
in West, M.A. (Ed.), Handbookof Work Group Psychology, John Wiley
& Sons, Chichester, pp. xxv-xxxiii.
Yin, R.K. (1994), Case Study Research: Design and Methods, Sage
Publications, Thousand Oaks, CA.
About the authorsHarm van Vijfeijken has a PhD in Industrial
Engineering and Management Science fromTechnische Universiteit
Eindhoven in the Netherlands. He is currently employed as
aneducational consultant at CPS, Amersfoort. His current research
interests are team performancemanagement, team development, and
pay-for-performance plans.
Ad Kleingeld has a PhD in Industrial Engineering and Management
Science from TechnischeUniversiteit Eindhoven in The Netherlands.
He is Assistant Professor at the Department ofTechnology
Management, a subdepartment of Human Performance Management, of
thisuniversity. His current research interests are performance
management of interdependentindividuals and groups, task
strategies, and the relationship between individuals
goalorientation and their use of feedback. Ad Kleingeld is the
corresponding author and can becontacted at:
[email protected]
Harrie van Tuijl has a PhD in experimental psychology from
Nijmegen University. He isAssociate Professor of personnel
management at Technische Universiteit Eindhoven,
PR35,1
116
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
Department of Technology Management, subdepartment of Human
Performance Management.His main research interests are productivity
enhancement, organizational learning, groupproblem solving, task
strategies, and cooperation and competition.
Jen A. Algera has a PhD in social sciences from Leiden
University. He is Professor ofPersonnel Management at Technische
Universiteit Eindhoven, the Netherlands. His currentwork is on goal
setting and feedback in combination with self-management of
individuals andgroups to arrive at high organizational
productivity, especially in new technology environments.
Henk Thierry has a PhD in psychology from the Free University,
Amsterdam. He is EmeritusProfessor in work and organizational
psychology at Tilburg University, The Netherlands. Hisresearch
domain covers pay and compensation at work, work time arrangements
and behavioraleffects, work motivation, and strategic human
resource management.
Teamperformancemanagement
117
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)
-
This article has been cited by:
1. Anat DrachZahavy, Anit Somech. 2013. Linking task and goal
interdependence to quality service. Journalof Service Management
24:2, 151-169. [Abstract] [Full Text] [PDF]
2. Botshabelo Kealesitse, Barry O'Mahony, Beverley Lloyd-Walker,
Michael Jay Polonsky. 2013. Developingcustomer-focused public
sector reward schemes. International Journal of Public Sector
Management 26:1,33-55. [Abstract] [Full Text] [PDF]
3. David J. Glew. 2012. Effects of Interdependence and Social
Interaction-Based Person-Team Fit.Administrative Sciences 2:4,
26-46. [CrossRef]
4. Umit Bititci, Patrizia Garengo, Viktor Drfler, Sai
Nudurupati. 2012. Performance Measurement:Challenges for Tomorrow*.
International Journal of Management Reviews 14:3, 305-327.
[CrossRef]
Dow
nloa
ded
by U
NIV
ERSI
TY O
F LE
ICES
TER
At 0
0:50
08
Mar
ch 2
015
(PT)