The Interaction Effect of Coworkers’ Pacing Styles on
Coworker Satisfaction and Dyadic Job Performance: Can
You Trust Your Coworker?
Joëlle Zander (10074880)
Master thesis Business Studies: Leadership and Management
University of Amsterdam
Faculty of Economics and Business
Amsterdam, June 27, 2014
Supervisor: Dr. W. van Eerde
1 Table of contents Table of contents ... 1 Abstract ... 3 1. Introduction ... 4 2. Literature review ... 7
2.1 Defining pacing style ... 7
2.2 Diversity research ... 9
2.3 Pacing style and related research ... 10
2.4 Trust ... 13
3. Conceptual framework ... 15
3.1 Similar pacing style in relation to coworker satisfaction ... 17
3.2 Similar pacing style in relation to dyadic job performance ... 18
3.3 Dissimilar pacing style in relation to coworker satisfaction and dyadic job performance ... 21
3.4 Trust and pacing style ... 24
3.5 Trust and dyadic job performance ... 27
3.6 Trust and coworker satisfaction ... 28
3.7 Mediated Moderation ... 28 4. Method ... 29 4.1 Participants ... 29 4.2 Research design ... 30 4.3 Sampling ... 31 4.4 Measures ... 32 4.5 Analysis ... 35 5. Results ... 36 5.1 Participants ... 36 5.2 Reliability ... 36 5.3 Correlations ... 38 5.4 PROCESS ... 39 5.4.1 Coworker Satisfaction ... 39
5.4.2 Dyadic Job Performance ... 42
5.4.3 Coworker trust ... 44
2
5.5 Additional findings ... 51
6. Discussion ... 52
6.1 Summary, interpretation, and theoretical contributions ... 52
6.2 Managerial implications ... 55
6.3 Limitations and future research ... 57
6.4 Conclusion ... 59 References ... 61 Appendices ... 70 Appendix A: Survey ... 70 Appendix B: Mail ... 73 Appendix C: Syntax ... 74
Appendix D: Exploratory Factor Analysis for Trust Propensity ... 82
Appendix E: Non-significant interaction effects ... 83
List of Tables and Figures Table 1……….…..………...………....16 2..……….………...37 3..……….……….………..………..…….40 4..……….………..…..……..42 5..……….………..……..……..46 6..………..…..……...47 7..……….………....………..47 8..……….………....………..48 9..………....………...51 Figure 1.……….………....………...16 2.……….………....………...40 3-5.………...41 6-7.………..…………..…...………..43 8-9………...………..…...……..44 10-11.………..……...………..45 12………..………..……...….46
3
Abstract
The concept of pacing style, which is the way individuals generally use their time under
deadline conditions, is regarded as a potentially important temporal topic, but is neglected in
the literature. Therefore, this research focuses on the moderating effect of the pacing style of
the coworker on the relationship between the pacing style of the employee and both coworker
satisfaction and dyadic job performance, which is mediated by coworker trust. This research
uses PROCESS to test model 1 for moderation and model 8 for mediated moderation. The
hypotheses are tested using a sample of coworker dyads consisting of employees employed in
the Netherlands in different industries (N = 90). For dyads that consisted of both members
having a steady action style and dyads that consisted of an employee with a deadline action
style and a coworker with a procrastination style mediated moderation has been found. As a
result of the high coworker trust that is evoked by the coworker with a high steady action
style, the employee has high coworker satisfaction and the dyadic job performance is high as
well. In addition, as a result of the low coworker trust that is evoked by the coworker with a
high procrastination style, the employee with a high early action style has low coworker
satisfaction and the dyadic job performance is low as well.
Keywords: dyadic job performance, coworker satisfaction, coworker trust, moderation,
4
1. Introduction
Time is important in contemporary business, as it is inescapable, restricted, and an
inherent part of daily life (Gevers, Mohammed, & Baytalskaya, 2013). Organizations, and
consequently employees, have to cope with the demand for shorter waiting times and
speeding up of delivery of services and products (Mohammed & Harrison, 2013). As a result,
temporal challenges have arisen. Short deadlines and complex coordination of several projects
in parallel require careful time management (Mohammed & Nadkarni, 2011). However,
aspects of time remain perhaps the most neglected and as illustrated critical issue in research
(Kozlowski & Bell, 2003, p. 364).
Working together inevitable means collaborating with someone who is to some degree
different from yourself. Differences in how individuals think about and value time (i.e.,
temporal differences) greatly influence team performance, either positively or negatively
(Mohammed & Harrison, 2013). Diversity research has mostly been conducted on task-based
types of diversity. Diversity researchers have urged that more research should focus on types
of diversity that are more related to the identities of the team members. Temporal differences
are more about the self-identity of the person because they are deeply rooted (Mohammed &
Harrison, 2013). Research on temporal individual differences is on the rise. However, it
scantly investigates what pacing style individuals possess, which refers to the way individuals
generally use their time under deadline conditions (Gevers, Rutte, & van Eerde, 2006, p. 55),
in spite of the notion that it is characterized as a meaningful and potentially important
construct that warrants further attention (Gevers et al., 2013). Investigating pacing style in
relation to individual, team, and organizational outcomes can be helpful to predict and
enhance employee well-being (e.g., satisfaction) and job fit (Gevers et al., 2013). Satisfaction
of employees is essential for team effectiveness (Gevers & Peeters, 2009). Furthermore,
5 meeting deadlines, which is relevant for practice and theory (Gevers, van Eerde, & Rutte,
2001). Not being able to meet deadlines can result in financial losses. It is claimed that
regularly failing to meet deadlines of clients can damage the competitive position of an
organization (Gevers, van Eerde, & Rutte, 2009).
Research on outcomes of temporal differences, such as pacing style and related
constructs, has been scarcely conducted. Research on teams found that having similar pacing
styles resulted in meeting deadlines more effectively (Gevers et al., 2006). In addition, being
similar has a positive effect on team satisfaction (Gevers & Peeters, 2009). However, a
limitation of the existing research in this area is that is it mostly conducted on students and in
the context of planned projects groups (e.g., Gevers et al., 2006; Gevers et al., 2009; Gevers &
Peeters, 2009). Students are able to delay activities by deciding not to study for an exam. In
contrast, employees have less possibility for this delay and may experience more negative
consequences for not completing promised tasks (Claessens, van Eerde, Rutte, & Roe, 2007).
Therefore, there is no confirmation that the student work environments results are
generalizable to the everyday professional work environment. The question arises: how do
different combinations of pacing styles relate to organizationally relevant outcomes, such as
collaboration between coworkers?
Trust is seen as a vital component of every effective and high quality working
relationship (Colquitt, Scott, & LePine, 2007; Tan & Lim, 2009). Consequences are
organizational effectiveness and as a result building a competitive advantage (Tan & Lim,
2009). Unfortunately, in research, interpersonal supervisor trust is mostly considered, leaving
coworker trust neglected (Knoll & Gill, 2011; Tan & Lim, 2009). Therefore, in this research
the role of coworker trust is investigated. Coworkers hold relatively equal power, interact
during a workday, and report to the same supervisor (Sherony & Green, 2002; Tan & Lim,
6 organization (Tan & Lim, 2009). In general, the more traditional dyadic level of exchange
among coworkers, coworker exchange, has been relatively unpopular in empirical research
(Sherony & Green, 2002). Only at this dyadic level can accurately be investigated what the
consequences are of collaborating with that particular coworker that has a (dis)similar pacing
style compared to oneself. This is in contrast to consequences of differences and average
pacing style at the team level (Gevers & Peeters, 2009; Gevers et al., 2006). What
consequences does it have to collaborate with a procrastinating coworker that constantly
delays its activities and is known for being less likely to meet deadlines (Van Eerde, 2003a)?
What is the probability of performing well together and will you be pleased to collaborate
with this person? What is the process through which this moderated effect is produced?
Would you trust this coworker?
In summary, the goal of this study is to contribute to theory of diversity, coworker
trust, and temporal research. This is done by investigating the consequences of coworker
collaboration with regard to their pacing styles on coworker dyadic outcomes. Furthermore,
coworker trust is incorporated as a mediator in this relationship, because it is characterized as
a source of competitive advantage for an organization and vital to effective relationships (Tan
& Lim, 2009). The current research examines the possible relationships between pacing style,
coworker trust, coworker satisfaction and dyadic job performance, using questionnaires
distributed among employees in multiple organizations.
Chapter 2 reviews the existing literature on pacing style and trust. Chapter 3 presents
the conceptual model and hypotheses. The research method, including participants, design,
sampling, measures, and data analysis, is described in chapter 4. Chapter 5 presents the results
of the data analysis. Finally, chapter 6 discusses the research and limitations and provides
7
2. Literature review
This chapter addresses the existing literature related to the research question in order
to emphasize the need for this research and define the research topic more precisely. First, the
literature concerning pacing style and diversity is discussed, followed by the literature on
trust.
2.1 Defining pacing style
The concept of pacing was introduced by Blount and Janicik (2002) and named pacing
preference. This refers to the anticipated momentum and flow regarding how events will
unfold over time. However, the difficulty related to this concept was that what is preferred
may not always correspond with what is really done. As a result, the concept of pacing style
emerged which refers to the way individuals generally use their time under deadline
conditions (Gevers et al., 2006, p. 55). Although temporal research on individual behavior is
relatively new, there are three related but distinct constructs to pacing style. First, time
urgency refers to feeling chronically hurried and viewing time as an enemy (Gevers et al.,
2013; Price, 1982). Time urgent individuals focus on the passage of time and deadlines
(Landy, Rastegary, Thayer, & Colvin, 1991). Attention is given to when work is due. In
contrast, pacing style captures how the resources are allocated towards a deadline
(Mohammed & Nadkarni, 2011). Second, polychronicity refers to a preference to work on
more than one task in parallel (Bluedorn, Kalliath, Strube, & Martin, 1999). Last, time
perspective refers to a bias towards past, present or future timeframes (Zimbardo & Boyd,
1999). Time urgency and time perspective are more general orientations, whereas
polychronicity and pacing style are more narrowly focused on work activity (Mohammed &
Harrison, 2013). The three constructs have in common that they do not capture how
8 After an extensive validity test, four styles have been proposed to be used in research
on pacing style. These are the early, deadline, U-shaped, and steady action style (Gevers et al.,
2013). An early action style refers to a person starting with a task soon after it has been
assigned and finishes it long before the deadline. A deadline action style entails that a person
starts working when there is pressure of the deadline approaching and works until there is no
time left. There is a hybrid form as well, the U-shaped action style, which combines the early
and deadline action styles. More effort is expended at the beginning and the end of the
deadline with a break in between. Additionally, some individuals spread their effort evenly
over time before the deadline which is called the steady action style. The combination of
pacing styles in a team influences when and how individuals work together towards
completing tasks (Mohammed & Nadkarni, 2011).
Apart from the construct of pacing style, there is the act of procrastination. This refers
to the voluntary delay of an intended course of action despite expecting to be worse off for the
delay (Steel, 2007, p. 66). This delay is engaged in, because the procrastinator anticipates
negative consequences of deadlines (Steel, 2007; Van Eerde, 2000). Some conceptualize the
deadline action style as a broader construct that encompasses procrastination (Gevers et al.,
2013). However, in the validation process it is confirmed that there are conceptual differences
between the deadline action style and procrastination. Individuals procrastinate because the
deadline is threatening or unattractive (Gevers et al., 2013; Van Eerde, 2000). In contrast, it is
possible that deadline action style individuals rationally and intentionally delay starting tasks
until close to the deadline, because they expect favorable benefits and/or want to challenge
themselves (Gevers et al., 2013). It is the inability to control one’s attention that is central to procrastination (Van Eerde, 2003b). The procrastinator engages in a pleasant distraction,
however, the distraction action is less important than the intended action. (Van Eerde, 2003b).
9 style and considered a fifth style. It is a specific instance of how actions towards a deadline
are delayed.
2.2 Diversity research
In diversity research broadly two theoretical perspectives exist based on social
categorization process theory and information/decision-making process theory. Social
categorization process theory proposes that individuals prefer to be in sync with others
(Blount & Janicik, 2002). Differences between individuals may elicit stereotypical
perceptions of dissimilar others, subgroup formation, and intergroup biases (in-group and
out-group) (Van Knippenberg & Schippers, 2007). This intergroup bias disrupts group
functioning, because there is more trust, willingness to cooperate, and a better interaction with
in-group than with out-group individuals (Van Knippenberg & Schippers, 2007). There are
less positive attitudes towards the out-group (Van Knippenberg & Schippers, 2007). In
heterogeneous groups there is more conflict (Jehn, Northcraft, & Neale, 1999; Pelled,
Eisenhardt, & Xin, 1999), less satisfaction (Tsui, Egan, & O’Reilly, 1992), and less social
integration (Harrison, Price, Gavin, & Florey, 2002; O’Reilly, Caldwell, & Barnett, 1989).
Information/decision-making process theory posits that differences in information,
knowledge and perspectives can be beneficial to group performance. As a result of the
divergent viewpoints, discussion about the team’s functioning arises and team learning will improve team performance (Van Knippenberg & Schippers, 2007). However, reflecting on
the team can also be considered a sign of disagreement about time, resulting in not meeting
deadlines (Gevers et al., 2009). With regard to temporal diversity, its value becomes salient
when performance characteristics of effectiveness seem opposites of each other such as speed
and accuracy, focus and flexibility, and bursts of effort and sustained action (Mohammed &
10 specific requirements. For example, deadline and early action style individuals may help in a
burst of effort and steady action style individuals to fulfill the sustained effort. Unfortunately,
diversity in pacing style has so far only been investigated in the research of Mohammed and
Harrison (2013) and Mohammed and Nadkarni (2011).
2.3 Pacing style and related research
Individual similarity in pacing style has been shown to be related to consensus on
temporal aspects of task execution in teams (e.g., deadlines, schedules, and task completion
times) (Gevers et al., 2006). In turn, this consensus stimulates the establishment of
coordination of team action (i.e., coordinated action). An optimal working relationship can be
built with intended actions executed in a timely and integrated manner. This results in meeting
deadlines more effectively (Gevers et al., 2009).
Research has been conducted to examine whether conscientiousness relates to the
different pacing styles. Conscientious individuals are orderly, dutifulness, self-disciplined,
responsible, persisting, and have high achievement motivation (Barrick & Mount, 1991). As a
result, conscientiousness is found to be positively related to job performance (Barrick &
Mount, 1991; Liao & Chuang, 2004). Conscientious individuals complete a higher percentage
of planned tasks on a daily basis than those who are less conscientious (Claessens, van Eerde,
Rutte, & Roe, 2010). In addition, conscientious individuals honor the commitments they have
entered into, avoid distractions, and deal effectively with interruptions (Claessens et al.,
2010). Conscientiousness positively relates to the early, steady, and U-shaped action style and
negatively relates to the deadline action and procrastination style (Gevers et al., 2006; Gevers
et al., 2013; Van Eerde, 2003a). With this outcome, research on conscientiousness is used to
11 When dissimilarity in conscientiousness at the individual level exists (i.e., the distance
between an individual member and his or her team mates), this negatively affects team
satisfaction (Gevers & Peeters, 2009). However, it did not affect satisfaction with the team’s
performance. Team composition in terms of personality affects the pleasure individuals derive
from collaborating and the dissimilarity of team members counteracts coordinated action at
the team level (Gevers & Peeters, 2009). This problem of less coordinated action results in
less satisfaction with the team as well as satisfaction with the team’s performance at the team level. As the dyadic level of exchange among coworkers has been largely ignored in empirical
research, the current research incorporates the concept of coworker satisfaction (Sherony &
Green, 2002). It is considered an important element of job satisfaction (Sherony & Green,
2002; Simon, Judge, & Halvorsen-Ganepola, 2010). Job satisfaction is ‘‘a pleasurable or
positive emotional state resulting from an appraisal of one’s job or job experiences” (Locke,
1976, p. 1300). Coworker satisfaction is therefore defined as a reflection of employees’
cognitive evaluations and affective responses towards their coworkers (Smith, Kendall, &
Hulin, 1969). Coworker satisfaction can relate to the coworkers, to the group composition,
and/or to how the coworkers collaborate on a task. It is reflected in the extent to which the
coworkers are willing to repeat collaboration on a future task (Peeters, Rutte, van Tuijl, &
Reymen, 2006).
Sherony and Green (2002) found that when greater diversity in coworker exchange
relationships exists (i.e., the traditional dyadic level of exchanges among coworkers reporting
to the same supervisor), these employees have less organizational commitment. Greater
diversity can create tensions within the group. No relationship with job satisfaction could be
found. Diversity in coworker exchange relationships can also refer to collaboration with
12 Planning is not identical to pacing style, as it focuses on the extent to which
individuals plan their work, not on when the work is actually done (Gevers et al., 2013).
However, planning styles can provide insight into how different styles might (mis)match with
each other. It seems that planning behavior directly influences employee performance and
indirectly influences employee performance, job satisfaction, and work strain through
perceived control of time (Claessens, van Eerde, Rutte, & Roe, 2004). By planning activities,
employees feel they have control over their time. Planning behavior enables individuals to
distribute attention and energy more effectively by scheduling and structuring their activities
in line with resources and opportunities at hand. This increases the likelihood of completing
work as planned, with less work strain, higher performance, and job satisfaction as a result
(Claessens et al, 2004). Research found that early and steady action style individuals plan
most of their activities and attain the highest performance of the different pacing styles
(Claessens, 2004). Early action style individuals are motivated to do whatever is needed to
finish their tasks in time. Steady action style individuals are not easily distracted in executing
tasks and monitor the time left before the deadline. They have to ensure that they spend
enough time on every task to eventually have maximum performance. In contrast, deadline
action style individuals do not plan their activities in detail. They start working right before
the deadline and will likely only focus on this particular task. The procrastination style cannot
be considered planning as the delay is not purposely planned (Van Eerde, 2003a). Then, when
individuals with a pacing style with a tendency to plan (i.e., early and steady action style) are
matched with individuals with a pacing style that do not plan (i.e., deadline action and
procrastination style), this may result in lower performance, job satisfaction, and higher work
strain. In addition, temporal consensus is found to facilitate meeting a deadline in a first
collaboration when, on average, the group members tend towards an early action style. When,
13 deadline (Gevers et al., 2006). This can be explained by the fact that the deadline action style
individuals do not have time to compensate for too optimistic time estimations for completing
tasks and/or correct for mismatches between the work of the individual team members.
These findings of the consequences of (dis)similar teams may also apply to certain
combinations of pacing styles in relation to coworker satisfaction and dyadic job performance.
This research investigates whether the findings of satisfaction with regard to dissimilarity in
conscientiousness can be generalized to pacing styles (Gevers & Peeters, 2009). In addition,
this research investigates whether the findings can be generalized from a student to a
professional work situation. Moreover, it seems plausible that this relationship between the
interaction of the pacing styles and coworker satisfaction and dyadic job performance goes
through coworker trust, as announced in the introduction. This will be elaborated in the
following section and chapter 3.
2.4 Trust
Trust is generally defined as “the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action
important to the trustor, irrespective of the ability to monitor or control that other party”
(Mayer, Davis, & Schoorman, 1995, p. 712). There are different foci of trust; trust in the
organization, trust in the team and interpersonal trust (e.g., subordinates, leaders, and
coworkers) (Fulmer & Gelfand, 2012).
Coworker trust has many consequences. First, it positively relates to organizational
citizenship behavior (OCB) (Mayer & Gavin, 2005). OCB is behavior of employees that goes
above and beyond their role descriptions to help coworkers and achieve organizational goals
(Organ, 1988).When you trust someone, you do not focus on self-protection such as
14 that more time can be spent on other activities, such as OCB. The trustors do not feel they are
taken advantage of and want to reciprocate this trusting relationship by engaging in OCB
(Chattopadhyay, 1999; Mayer & Gavin, 2005).
Second, with regard to job performance, results are mixed, with findings of positive
and non-existent relationships (Mayer & Gavin, 2005). The line of reasoning is practically the
same as for OCB. Less time is wasted on monitoring others and more time is available for
productive efforts.
Third, trust positively relates to coworker satisfaction. As already mentioned, trust is
important for high quality working relationships (Tan & Lim, 2009). Trust makes it easy to
interact with each other and work can be coordinated more effectively (Lount Jr, 2010). As
mentioned, trust makes employees help each other and makes it unnecessary to watch
everything the coworker does (Mayer & Gavin, 2005). This all makes collaborating pleasant.
With regard to trust and being (dis)similar to coworkers, research is scarce.
Chattopadhyay (1999) found that demographic dissimilarity (i.e., age) negatively relates to
OCB through trust. As a result of their dissimilarity, the employees are not able to build
strong coworker relationships. Therefore, they will not engage in helping them beyond what is
expected (Chattopadhyay, 1999). Although pacing style is not a demographic characteristic,
being dissimilar in pacing style can have negative effects on trust and subsequently OCB (cf.
Gevers et al., 2006). In addition, individuals are likely to trust others when expectations about
appropriate behavior are shared and when they can rely on the other to capably complete their
assignments (McAllister, 1995; Husted, 1990). This is consistent with the idea that diversity
in pacing style is related to coworker trust.
Taken together, this leads to the research question; To what extent are dyads that
15 dyadic job performance? Additionally, to what extent does coworker trust mediate this
moderated relationship?
3. Conceptual framework
This chapter first discusses the reasoning with regard to coworker dyads consisting of
members having similar pacing styles in relation to coworker satisfaction and dyadic job
performance. This is followed by coworker dyads consisting of members having dissimilar
pacing styles in relation to coworker satisfaction and dyadic job performance. The last section
discusses the mediating role of coworker trust in the mentioned moderated relationships. An
overview of the hypotheses and the conceptual model are presented in Table 1 and Figure 1.
In this research the U-shaped action style is not included in the hypotheses. The reason
for this is that different interpretations of the motives of this style exist (Claessens, 2004). On
the one hand, the style can be motivated by starting early, give it a rest to think things over or
give priority to other tasks and then finish with new energy. On the other hand, the break
might be an indicator of distraction or not following a planning as well. Individuals with a
U-shaped action style might be frustrated and do not know how to finish and wait until time
pressure builds up near the deadline, hoping this affects performance positively. This makes it
ambiguous how to interpret the behavior.
Unfortunately, little research has been conducted on pacing style. Therefore, in
determining the hypotheses, aside from the pacing style literature, there is drawn upon another
concept as well. This concept is conscientiousness, as it is found to be related to all five
16 Table 1 Hypotheses Interactions (E * C) (a) Coworker Satisfaction (E) (b) Dyadic Job Performance (E) (c)
Coworker Trust (E)
1. Early * Early + + + 2. Deadline * Deadline + - + 3. Steady * Steady + + + 4. Procrastination * Procrastination - - - 5. Early * Deadline - + - 6. Early * Steady + + + 7. Early * Procrastination - - - 8. Deadline * Early - + + 9. Deadline * Steady - - + 10. Deadline * Procrastination - - - 11. Steady * Early + + + 12. Steady * Deadline - - - 13. Steady * Procrastination - - - 14. Procrastination * Early - - + 15. Procrastination * Deadline - - - 16. Procrastination * Steady - - + (a)
Coworker Satisfaction (E)
(b)
Dyadic Job Performance (E)
17. Coworker Trust + +
Note. Early: Early Action Style; Steady: Steady Action Style; Deadline: Deadline Action Style; Procrastination: Procrastination Style.
(E) = Employee, (C) = Coworker.
To illustrate, hypothesis 1a indicates that there is an interaction effect for dyads consisting of both members having an Early Action Style (1. Early * Early) on Coworker Satisfaction (a). When both members have an Early Action Style, Coworker Satisfaction will be high (+).
17 As can be seen from Table 1, hypotheses have been generated with the effect of the
interactions made up of all combinations of pacing styles on the outcome variables. Even
though, it is expected that the pacing style combination sequence (e.g., Early * Deadline or
Deadline * Early), has no impact on (a) coworker satisfaction and (b) dyadic job performance,
hypotheses have been included. Hypothesizing every possible combination separately makes
it possible to specify which exact combination of the two pacing styles is associated with
high/low coworker satisfaction and/or dyadic job performance. When dyads consist of
members having dissimilar pacing styles (e.g., Early * Deadline or Deadline * Early), the
pacing style combination sequence could potentially differently impact (c) coworker trust. For
example, when the employee has an early and the coworker a deadline action style,
hypothesis 5c expects a negative association with coworker trust. However, when the
employee has a deadline and the coworker an early action style, hypothesis 8c expects a
positive association with coworker trust.
3.1 Similar pacing style in relation to coworker satisfaction
As mentioned, social categorization process theory refers to people preferring to be in
sync with others (Blount & Janicik, 2002). On average, individuals are more attracted to
others who are similar to themselves (Byrne, 1971). They foresee that by collaborating with
similar individuals, reinforcement of their own beliefs and values will occur (Gevers &
Peeters, 2009). With regard to pacing style, to build a satisfying working relationship, there
should be consensus on how to approach a deadline. When collaborating with a coworker
with the same style, there will be no arguments about the scheduling, timing, and sequencing
of tasks (Gevers & Peeters, 2009). The coworkers know what to expect from each other. This
creates feelings of positive affect among coworkers (Gevers & Peeters, 2009). As a result,
18 associated with high coworker satisfaction (H1a, H2a, and H3a). More precisely, the
relationship between the pacing style of the employee and coworker satisfaction will be
moderated by the pacing style of the coworker, such that when both are the same, the
relationship will be stronger. For brevity, the short form of this formulation will be used
throughout this thesis.
With regard to coworker dyads consisting of both members having a procrastination
style, the expectations deviate from the other dyads having members with the same style. Although the coworkers are similar in pacing style, procrastinators have too many
characteristics that make collaboration unpleasant. First, consensus on how to work towards a
deadline cannot be established. They do not delay purposefully, which makes it impossible to
accurately indicate when they start working (Van Eerde, 2003a). This works both ways,
because both procrastinators have no clear approach to accomplish their tasks. Second,
procrastinators evaluate other procrastinators more harshly. This is because they are
dissatisfied with their own, similar behavior and think the procrastinator should be punished.
The procrastinators indicated that other procrastinators need to be fired (Ferrari, 1992). Last,
procrastinators can be impulsive (i.e., choosing a less important action that is more pleasant),
which makes them unpredictable (Van Eerde, 2000). Therefore, dyads consisting of both
members having a procrastination style will be associated with low coworker satisfaction,
because arguments will likely arise about when they are going to execute their tasks (H4a).
3.2 Similar pacing style in relation to dyadic job performance
All pacing styles are related to conscientiousness (Gevers et al., 2013; Van Eerde,
2003a). The early and steady action style positively relate to conscientiousness, while the
deadline action and procrastination style are negatively related. Conscientious individuals
19 Mount, 1991; Liao & Chuang, 2004). It appears that low variability in and high mean levels
of conscientiousness result in higher performance at the team level as well (Peeters, van Tuijl,
Rutte, & Reymen, 2006). Reasons are, first, conscientious individuals honor the commitments
they have entered into, avoid distractions, and effectively deal with interruptions (Claessens et
al., 2010). Second, they are more concerned with planning, timeliness, punctuality, and
meeting schedules to eventually achieve high performance (Gevers & Peeters, 2009). Third,
homogenous conscientious groups assume that all members will take similar levels of
responsibility in completing a task (Mohammed & Angell, 2003). Last, high achievement
oriented individuals are more concerned with the success of the team and solve problems
more efficiently (Barrick, Stewart, Neubert, & Mount, 1998). These elements make the team
work more effectively. In contrast, low conscientious individuals do not take these elements
as serious and tolerate more flexibility in their working style (Gevers & Peeters, 2009).
To perform well together, the importance of setting goals, scheduling time, and
striving towards timely completion of tasks becomes apparent (Peeters, Rutte et al., 2006).
These are characteristics that are congruent with the early and steady action style (Gevers et
al., 2013). Early action style individuals have a mindset of doing whatever they have in their
power to complete a task in time (Claessens, 2004). They are committed to everything they
do. In addition, they expect obstacles to arise and prevent inability to execute their task by
starting as soon as possible. When a group of early action style individuals reaches consensus
on how to work towards a deadline, this results in meeting deadlines (Gevers et al., 2006).
They have enough time to compensate for too optimistic time estimations and/or correct for
mismatches between the work of the individual team members. Working at a constant pace
can imply that steady action style individuals are not easily distracted, monitor their time left,
and plan at what time they work on tasks (Claessens, 2004). As conscientiousness positively
20 mean level of conscientiousness (Gevers et al., 2013). In addition, they are found to attain the
highest performance of the pacing styles (Claessens, 2004) Therefore, it is expected that
dyads consisting of members having both an early or both a steady action style will be
associated with high dyadic job performance (H1b and H3b).
In contrast to early and steady action style individuals, individuals with a deadline
action style start near the deadline. Time pressure can result in increased stimulation and
motivation, because of the challenge created to eventually succeed given the short period of
time (Van Eerde, 2003a). However, time pressure can decrease accuracy and punctuality as
well, because less time is left than when started earlier (Van Eerde, 2003a). Starting near the
deadline results in having no time to compensate for inaccurate time estimations and/or
mismatch correction, which the early action style has (Gevers et al., 2006). Having consensus
on starting late may be satisfying in collaborating. However, when coworkers both
underestimate the time needed or agree that the deadline is not important, this sharing of
pacing style detracts from their ability to meet the deadline (Gevers et al., 2006). The way of
approaching the deadline should be functional in meeting the task demands, which does not
seem to be the case for two deadline action or procrastination style individuals collaborating.
What has to be noted is that in the research of Gevers and colleagues (2006) the time frame of
the projects used was short. Therefore, they argue that the teams may have had no chance to
adapt their activities that are required to meet the deadline. However, this may always be the
case as that is how the deadline action style individuals operate. They start near the deadline,
which makes learning from feedback less likely. Early in time they have no work on which
feedback can be provided. In addition, the mean level of conscientiousness is low. Therefore,
dyads consisting of both members having a deadline action style will be associated with low
21 Procrastinators do not purposefully plan their delay, distinguishing them from the
deadline action style (Van Eerde, 2003a). As a result of the delay of tasks, they have new
tasks to be performed on top of the delayed tasks. This makes procrastinators more likely to
be dysfunctional in effective performance than deadline action style individuals, who can
deliberately delay. The delay of procrastinators can have positive effects, because the
procrastinator has time to help others (Van Eerde, 2000). However, in the case of deadlines,
help should not be expected, because helping a coworker still contributes to meeting the
deadline. Other characteristics that impair their performance are first, procrastination is
positively related to missing deadlines (Van Eerde, 2003a). Second, procrastinators lack
self-control and volition (Van Eerde, 2000; 2003b). They cannot self-control themselves from
preferring immediate positive outcomes that are often less important (Van Eerde, 2003b; Van
Eerde, 2000). Third, they handicap themselves by creating activities that impede their chance
of successful performance (Urdan & Midgley, 2001; Van Eerde, 2003a). Last, they have a
planning fallacy, which mean they tend to be too optimistic about the time that is left (Van
Eerde, 2000). Therefore, dyads consisting of both members having a procrastination style will
be associated with low dyadic job performance as there is no control over their behavior
because both lack it (H4b).
3.3 Dissimilar pacing style in relation to coworker satisfaction and dyadic job performance
Differences in how coworkers approach deadlines and goals put the effectiveness of
the team at stake (Peeters, Rutte et al., 2006). There will be disagreement, which can easily
evolve into dissatisfaction. When individuals are unreliable and do not feel responsible for
their work, there will be less communication between coworkers, efforts will be withhold, and
the cooperation suffers (Molleman, Nauta, & Jehn, 2004). This is not beneficial to pleasant
22 characteristics such as being reliable and accountable (i.e., high mean level of
conscientiousness), everyone will want to cooperate. This is applicable to the conscientious
pacing styles (early and steady action style) as they can be considered reliable (Molleman et
al., 2004). When similarity in conscientiousness exists, this positively affects team
satisfaction, because there is coordination of action (Gevers & Peeters, 2009). For both the
early and steady action style it is important to start early and plan in advance how to approach
the deadline (Claessens, 2004). They both want to avoid working under deadline pressure.
Therefore, dyads consisting of members having an early and steady action style will be
associated with high coworker satisfaction and dyadic job performance (H6a-b and H11a-b).
Combining different pacing styles can be beneficial. For example, early action style
individuals start the work. Then, deadline action style individuals can anticipate on changes
that arise after the early action style has finished (Mohammed & Nadkarni, 2011). However,
being unreliable and not feeling responsible for your work (i.e., low conscientiousness) is not
beneficial to pleasant collaboration and performance (Molleman et al., 2004). In addition,
being different from coworkers can result in less social integration and identification (Gevers
& Peeters, 2009; Harrison et al., 2002). This negative influence of diversity seems to be
stronger. Individuals with a deadline action and procrastination style feel different from their
coworker. They may even feel threatened because the highly conscientious coworker, in this
case the early and steady action style, maintains higher standards of performance (Gevers &
Peeters, 2009). As threat is the main reason to procrastinate, this may result in procrastinators
procrastinating even more (Gevers et al., 2013). It may also be the case that the
procrastinators and deadline action style individuals feel rushed and not appreciated. It seems
likely that the early and steady action style individuals want to change the way of working of
the other styles to ensure order, predictability, and no need to rush (Gevers et al., 2013). This
23 individuals. Similarly, the behavior of the deadline action and procrastination style individuals
can result in coworker dissatisfaction for the early and steady action style individuals as well.
For the early and steady action style individuals it may be frustrating to observe that they are
working and the deadline action and procrastination style individuals have not done anything
so far. When there is dissimilarity in and a low mean level of conscientiousness, uneven
quality or even no work done by coworkers can occur, resulting in low performance
(Mohammed & Angell, 2003; Peeters, van Tuijl et al., 2006). The early and the steady action
style individuals may feel they have to perform or redo the work of the deadline action and
procrastination style individuals, resulting in feelings of inequity (Mohammed & Angell,
2003). Feelings of inequity counteract the process of effective team performance. Work
processes that are effective, resulting in high performance, are an important determinant of the
satisfaction of a team member (Peeters, Rutte et al., 2006). Early and steady action style
individuals can perceive the behavior of deadline action style individuals as irresponsible and
a sign of no commitment to the team (Mohammed & Nadkarni, 2011). However, deadline
action style individuals perceive their own style as most efficient because when changes
occur, they can still render their tasks. Therefore, dyads consisting of members having a
deadline and steady action style will be associated with low coworker satisfaction and dyadic
job performance (H9a-b and H12a-b).
The relationship between a dyad consisting of members having an early and deadline
action style and dyadic job performance is different. These coworker dyads will be associated
with high dyadic job performance (H5b and H8b). The early action style is the highest on
conscientiousness and is determined to meet the deadline and wants to invest whatever they
can to do so (Claessens, 2004). This however likely results in very low satisfaction when
24 With regard to the procrastination style, as mentioned, procrastinators seemingly have
too many characteristics that make properly performing impossible. The delaying can
frustrate coworkers. They have to wait for them and lose valuable time in reminding them of
the deadline (Van Eerde, 2000). For the early and the steady action style individuals it is
important to start early and plan in advance how they will approach the deadline. Waiting and
losing time increases the chances of working under pressure near the deadline, which they
both do not want (Gevers et al., 2013). For deadline action style individuals it is crucial to
work as efficiently as possible as they do not have much time to lose. Then, waiting for a
procrastinator, with the possibility of further delay, will detract from the possibility to have
high performance. In the worst case, when the avoided goals are work-oriented and the
distracting goals are private-oriented this can have negative effects. For example, they arrive
late at work or are maybe even absent or withdraw for longer periods (Van Eerde, 2000). This
makes it difficult to perform highly with procrastinators and makes the working relationship
unpleasant. It is not possible to rely on them. Furthermore, there is dissimilarity in and a lower
mean level of conscientiousness. To conclude, dyads consisting of a member having a
procrastination style will be associated with low coworker satisfaction and dyadic job
performance (H7a-b, H10a-b, H13a-b, H14a-b, H15a-b, and H16a-b).
3.4 Trust and pacing style
Illustrated in the literature review, trust is an important aspect of high quality
relationship (Tan & Lim, 2009). In the reasoning in the section above there are indicators of
the importance of trust. For example, being reliable and responsible in completing a task
(Mohammed & Angell, 2003; Molleman et al., 2004). In addition, having consensus on
25 coworkers and makes them more likely to trust each other (Chattopadhyay, 1999; Klimoski &
Mohammed, 1994; Krebs, Hobman, & Bordia, 2006; Tsui et al., 1992).
Coworkers who share the approach to work towards deadlines are likely to trust each
other. They know what to expect from each other as they work similarly. Sharing expectations
about appropriate behavior is important for trust (Husted, 1990). Therefore, dyads consisting
of both members having an early, deadline or steady action style will be associated with high
coworker trust (H1c, H2c, and H3c).
Early and steady action style individuals are expected to be considered trustworthy.
Conscientiousness positively relates to both styles and to trustworthiness (Evans & Revelle,
2008). Conscientiousness individuals are disciplined, organized, and persistent which seem
characteristics that make these individuals capable of completing tasks well. As a result, these
two pacing styles are likely to be consistent, predictable, reliable, and not impulsive (Evans &
Revelle, 2008; Molleman et al., 2004). Early and steady action style individuals are known for
planning, which makes it possible to check whether they are on schedule and what this
schedule looks like (Claessens, 2004). Therefore, dyads consisting of a coworker having an
early or steady action style will be associated with high coworker trust (H6c, H8c, H9c, H11c,
H14c, and H16c).
Considering the trustworthiness of the deadline action style, they do not plan their
activities in detail, but simply start when the deadline is near (Claessens, 2004). Although
they are described as delaying intentionally to challenge themselves, an aspect of threat of the
deadline is still present (Gevers et al., 2013). This makes them seem less dependable, as the
delay can be too far stretched in time, exceeding a deadline. However, this way of working
does not necessarily mean that they want to take advantage of coworkers starting early. The
majority of their delays are deliberate to challenge themselves. Nevertheless, the other pacing
26 and steady action style individuals start the work, while the deadline action style does
nothing. In addition, adherence to a set of principles that the trustor thinks are acceptable is
important in the development of coworker trust (Mayer et al., 1995). Early and steady action
style individuals likely perceive the deadline action style individuals as not adhering to their
set of principles. Both the early and the steady action style individuals want to avoid rushing
near the deadline (Gevers et al., 2013). Although procrastination style individuals delay, it is
difficult for them to fully trust the deadline action style individuals. Therefore, except for
dyads consisting of both members having a deadline action style, dyads consisting of a
coworker having a deadline action style will be associated with low coworker trust (H5c,
H12c and H15c).
With regard to the trustworthiness of procrastinators, they are impulsive, lack
self-control, handicap their selves, and are positively related to missing deadlines (Van Eerde,
2000; 2003a; 2003b). As a result, it is difficult to depend on them and predict their behavior
which makes the trustor more vulnerable. In addition, procrastinators do not delay
purposefully, which makes it impossible to accurately indicate when they start working (Van
Eerde, 2003a). As with the deadline action style, such an attitude can be seen as free riding.
The coworkers feel they are taken advantage of, as they trusted the procrastinator to complete
the tasks. As a result, to meet the deadline they have to do the work themselves. This makes
them less likely to support the trustee (Ferres, Connell, & Travaglione, 2004). Furthermore, to
trust someone, this person has to act with volition, which procrastinators do not do (Mayer &
Davis, 1999; Van Eerde, 2000). The coworkers with an early and steady action style have to
wait and see whether their coworker with a procrastination style will in the end act on their
promises. They cannot expect with full confidence that the procrastination style individuals
will perform the tasks to meet the deadline. This does not correspond with the definition of
27 procrastination style as well. Therefore, dyads consisting of a coworker having a
procrastination style will be associated with low coworker trust (H4c, H7c, H10c, and H13c).
3.5 Trust and dyadic job performance
As already mentioned in the literature review, trust improves performance because less
time is lost in monitoring others and more time is available for productive efforts (Mayer &
Gavin, 2005). Trust makes it more likely that cooperation is improved and a motivation arises
to collaborate (Dirks, 1999). In addition, coworkers who trust each other will help one
another, as they expect that their coworkers will not take advantage of them (Dirks, 1999).
Furthermore, individuals who trust each other are more likely to share their resources (e.g.,
knowledge) (Dirks & Skarlicki, 2009; Van Wijk, Jansen, & Lyles, 2008). This is a social
exchange. What is important is that the exchange is reciprocated by the coworker who should
also trust the initial exchanger (Blau, 1964; De Jong & Dirks, 2012). Trustworthiness can be a
signal helping to create a social exchange. It gives the motivation to return something
(Colquitt et al., 2007). This all leads to better performance and a bigger chance of meeting
deadlines.
In contrast, when employees do not trust their coworkers this will detract from
performance. If employees are afraid they are taken advantage of by coworkers or view them
as undependable it is likely that they will focus their attention towards tasks and goals on
which they do not rely on that person (Dirks, 1999). This makes it more likely that the mutual
goals are not granted attention and will not be met. Therefore, coworker trust will be
28
3.6 Trust and coworker satisfaction
Building on this knowledge, trust will improve coworker satisfaction as well. Trust
makes it easy to interact with each other and work can be coordinated more effectively (Lount
Jr, 2010). As mentioned, trust makes employees help each other and makes it unnecessary to
watch everything the coworker does (Mayer & Gavin, 2005). An environment of trust makes
employees feel safe and think more positively about their coworkers (Matzler & Renzl, 2006;
Paille, Grima, & Dufour, 2012). This all makes collaboration more pleasant. Where there is
no trust, psychological distress occurs because one cannot depend on anybody (Matzler &
Renzl, 2006). Therefore, coworker trust will be associated with high coworker satisfaction
(H17a).
3.7 Mediated Moderation
As evident in the hypotheses above, it is expected that the effect of the pacing style of
the employee on coworker satisfaction/dyadic job performance will depend on whether the
coworker is high or low on a pacing style. Further, this moderation will be mediated by
coworker trust. Trust has been identified as a mediator between dissimilar coworkers and
OCB (Chattopadhyay, 1999). As trust positively relates to coworker satisfaction and
performance (Mayer & Gavin, 2005; Matzler & Renzl, 2006), it is proposed that coworker
trust mediates the relationship between the interaction of coworkers’ pacing styles and
coworker satisfaction/dyadic job performance. That is, the difference in coworker satisfaction
(H1d – H16d) and dyadic job performance (H1e – H16e) for different combinations of pacing
styles is due to the level of coworker trust. The pacing style of the coworker will influence
whether the employee has high or low coworker trust. For example, dyads consisting of both
members having a high early action style will be associated with high coworker trust (H1c).
29 performance is high. However, if the coworker is perceived as untrustworthy, coworker
satisfaction/dyadic job performance is low. This implies that coworker trust, which is evoked
by the different combinations of pacing styles, is essential in the workplace to result in high or
low coworker satisfaction and dyadic job performance.
4. Method
In the previous chapters, the research question has been highlighted by the literature
review and the conceptual framework including hypotheses has been developed. This chapter
discusses the research design and method through which the hypotheses were tested. The first
section discuses the participants, research design, sampling, and measures used in the survey.
The chapter ends with the proposed analysis of the data.
4.1 Participants
The sample consisted of 90 matched coworker dyads employed in the Dutch
workplace. There were incomplete dyads, because one of the participants of the dyad did not
fill out the questionnaire. These participants were excluded from the sample. In the sample,
45% of the participants were male, the average age was 39.39 years (SD = 12.62). On average
they worked for their current employer for 9.39 years (SD = 9.68) and had a 34.3 hour work
week (SD = 9.22). The highest level of education attained was 5.6% high school, 21.7% trade
school, 43.3% bachelors, 28.9% university bachelor and master and .6% attained another
education level. The participants on average collaborated 19.30 hours per week (SD = 12.03)
for 4.33 years (SD = 4.44). The average team size was 12.47 members (SD = 13.07). 31.7%
30
4.2 Research design
This research used a cross-sectional survey design, because relationships between
variables were to be established and the answers given were standardized and allowed easy
comparison. The survey made it easier to reach individuals in a quick and affordable manner.
More specifically, a Dutch and an English self-administered questionnaire was used taken in
the form of matched dyads of two coworkers.
The survey (see Appendix A: Survey) contained a cover letter, which explained the
purpose of the questionnaire, why the participants’ opinion was important, and assurance of
anonymity and confidentiality. The questions were split up in three parts to give an overview
of what the participant could expect. The questionnaire started with questions that concerned
the participant followed by questions concerned with the coworker and their collaboration.
Finally, general questions were asked. Items with the same response scale were grouped as
much as possible to reduce cognitive load and make it visually look shorter. The questionnaire
closed with a short text in which the participants were thanked for their contribution and
provided the opportunity to enter their email address to receive the results of the study. A pilot
study was conducted to indicate the time needed for the questionnaire and whether the
questions were clear. Filling out the questionnaire took approximately fifteen minutes. The
supervisor checked the questionnaire as well. To link the coworker’s responses, codes were
used that had to be entered in the first question. The timeframe of the data collection was
approximately eight weeks. Approximately two weeks and five weeks after the first
invitation, a reminder email was sent to estimate who would participate and request the
participants to fill out the questionnaire.
Besides the advantages, there are limitations to the survey design. First, the limited
number of questions to be asked, as individuals are generally not willing to fill out a long
31 questions. Second, with a questionnaire, there is only one chance to collect the data.
Therefore, it must be designed properly from the beginning. The questions cannot be adapted,
as is the case with qualitative research. This is something that was kept in mind while
designing the questionnaire. This was a conscious and well-thought process using for example
reliable and validated scales. Third, it is not possible to get in depth insight into why
participants choose their answers. However, at this point, in the research on pacing style any
finding is of added value because not much is known. Last, it is a cross-sectional design,
which means causal relationships cannot be established. However, due to time limits this was
the best option.
4.3 Sampling
An online questionnaire was developed using uvafeb.qualtrics.com. Participants were
recruited using two non-probability sampling methods. First, purposive sampling as the
judgment of the researchers was used to select participants that were typical and suitable for
the questionnaire as questions included working towards deadlines and coworker
collaboration. Second, snowball sampling was used by requesting participants to search for
other persons who could fill out the questionnaire as well. The participants were preferably
approached face-to-face to explain the need for their participation and build compassion. If
this was not possible, it was done via telephone or e-mail (see Appendix B: Mail). The
limitation of the purposive and snowball sampling is the reduced variety in participants,
because they are all in the network of the researchers and participants. To ensure a broad
representative sample, dyads originating from various organizations in different industries
were approached. The target number of participants was 100 coworker dyads. The
prerequisites for participation were that the participants collaborated and had some form of
32 manager and the participants could not be interns or have the job as a part time student job.
Furthermore, not more than one coworker dyad was collected in one team to ensure
independence of the measurement points. As a result of these prerequisites it turned out to be
more difficult to find appropriate participants than was thought beforehand. Individuals
apparently worked more independently, did not work under the same manager or only
collaborated with their manager.
4.4 Measures
All variables, except for coworker exchange and closeness, were measured at the
employee level using a 5-point Likert scale (1= strongly disagree, 5= strongly agree). The
coworker trust, coworker satisfaction and dyadic job performance scales have been reworded
to reflect the coworker dyad level. Except for the PACED scale, all scales had to be translated
to English or Dutch. The technique of back-translating was used. To illustrate, the item “If I
got into difficulties at work I know my workmates would try and help me out” was translated into “Als ik problemen heb op mijn werk weet ik dat mijn collega mij zal proberen te helpen”. The item “Onze prestaties zijn meestal beter dan die van anderen” was translated into “Our
performance is usually better than the performance of other coworkers”.
Pacing style (9 items). Pacing style was assessed by means of the PACED scale
(Gevers et al., 2013). Each style, deadline, steady, and U-shaped action style, was measured
with three items. The early and the deadline action style were measured with the same three
items, indicating that a high score was the deadline action style and a low score was the early
action style. They define pacing style as behavioral tendencies regarding the distribution of
33 tasks in a relatively short time before the deadline” (+). A high score indicated a high degree
of the pacing style.
Procrastination (9 items, employee α = .81; coworker α = .79). Procrastination was
assessed by means of the General Procrastination (GP) scale (Lay, 1986). Lay defines
procrastination as the tendency to postpone that which is necessary to reach some goal. An
example item is “I am continually saying I’ll do it tomorrow” (+). Items 5, 7, and 9 were counter indicative. A high score indicated a high degree of procrastination.
Coworker trust (6 items, α = .90). Coworker trust was assesses by means of the
Interpersonal Trust at work scale (Cook & Wall, 1980). They define interpersonal trust at
work as the faith in the trustworthy intentions of others and the confidence in the ability of
others. An example item is “I have full confidence in the skills of my coworker” (+). A high score indicated a high degree of coworker trust.
Coworker satisfaction (3 items, α = .89). Coworker satisfaction was assessed by means
of the Individual Satisfaction with the team scale (Peeters, Rutte et al., 2006). They define it
as experiencing both the team (mates) and the cooperation within the team as pleasant. An
example item is “Taken as a whole, I am satisfied with our cooperation” (+). A high score
indicated a high degree of coworker satisfaction.
Coworker exchange (6 items, α = .77). Coworker exchange was assessed by means of
the Coworker Exchange scale (Sherony & Green, 2002). They define it as exchanges among
coworkers who report to the same supervisor. It was measured using several 5-point Likert
scales ranging from rarely - very often to not a bit – a great deal. An example item is “How
well does your coworker understand your work problems and needs?” (+). A high score indicated a positive coworker exchange.
34
Dyadic job performance (7 items, α = .67). Dyadic job performance was assessed by
means of the Job Performance scale (Roe, Zinovieva, Dienes, & Ten Horn, 2000). An
example item is “Compared to the standards we usually get good results from our work” (+). Items 5 and 6 were counter indicative. A high score indicated high dyadic job performance.
Coordinated action (9 items, α = .82). Coordinated action was assessed by means of
the Coordinated Action scale (Gevers, 2004; Heath & Staudenmayer, 2000; Janicik & Bartel,
2003; Salas, Burke, & Cannon-Bowers, 2000). It is defined as the establishment of integrated
and synchronized action within the team. An example item is: “Our work activities are well coordinated” (+). Items 3, 4, 5, 6, 7, 8, and 9 were counter indicative. A high score indicated a high degree of coordinated action.
Control variables. As potential control variables, age, gender, level of education,
industry, organizational tenure, working hours per week, collaboration tenure, working hours
together, self-assignment to dyad, team size and nationality were included in the
questionnaire. Trust Propensity1 (8 items, α = .58). Trust propensity was assessed by means of
the Trust Propensity scale (Mayer & Davis, 1999). They define trust propensity as a general
willingness to trust others. It was considered a potential control variable because it influences
how much a person will trust others before information on those other persons is available
(Mayer et al., 1995). An example item is “Most people can be counted on to do what they say
they will do” (+). Items 1, 4, and 6 were counter indicative. A high score indicated a high
degree of trust propensity. Task interdependence (3 items, α = .73). Task interdependence was
assessed by means of the Task Interdependence scale (Pearce & Gregersen, 1991). They
define task interdependence as the degree to which employees work continuously with other
1 Trust propensity was not included in the data analysis, because an exploratory factor analysis with coworker trust and trust propensity showed that the items loaded on three components indicating that the scale is not coherent (see Appendix D: Exploratory Factor Analysis for Trust Propensity). It was tried to make a coherent scale with items 3, 5, 7 and 8 (α = .59). This scale still did not correlate with coworker trust (r = .12, ns).
35 employees who depend on them. An example item is “I frequently must coordinate my efforts with my colleague” (+). A high score indicated a high degree task interdependence. Closeness was assessed at the employee level with one item using a 5-point Likert scale (1 = not at all
close, 5 = very close) (Cummings, Butler, & Kraut, 2002). The item was “How close do you
feel to your coworker?”.
4.5 Analysis
For the data analysis, the statistical analysis program SPSS for Windows (Version 21)
was used. For a summary of the syntax, see Appendix C: Syntax. To test the hypotheses,
model 1 (simple moderation) and 8 (mediated moderation) in PROCESS were used with
various control variables (Hayes, 2012). In the model 1 testing, the pacing style of the
employee was the independent variable (X), the pacing style of the coworker was the
moderator (M) and coworker satisfaction consecutively dyadic job performance and coworker
trust rated by the employee were the dependent variable (Y).
A boot-strapped mediated moderation analysis (Model 8) (Hayes, 2012) was used to
assess whether the relation between the pacing style of the employee and coworker
satisfaction/dyadic job performance, moderated by the pacing style of the coworker, goes
through coworker trust. In the model 8 testing, the pacing style of the employee is the
independent variable (X), the pacing style of the coworker the moderator (W), coworker trust
of the employee is the mediator (M) and dyadic job performance rated by the employee was
the dependent variable (Y). In a second step of testing coworker satisfaction of the employee
36
5. Results
This chapter discusses the results of the data analysis on the collected data. First,
descriptive statistics are presented and the reliability of the scales is shown by calculating
Cronbach’s Alpha. Secondly, the results of the PROCESS analyses are presented to test the hypotheses.
5.1 Participants
For the data cleaning, some scales were adjusted to numeric answers only, in case they
contained words. For example, for the tenure of working together, participants answered five
years or six months. These answers were adjusted to 5 and 0.5. Cases in which sentences were
written such as “The hours I collaborate with my coworker depends on the week, sometimes it is 20 other weeks 5” were considered missing values. Dealing with missing values of the variables important for the hypothesis testing was not necessary, because the responses were
forced to ensure that all questions were filled out. Only completed questionnaires and
complete dyads were included in the analysis. This means that the analysis was conducted
with 90 coworker dyads.
5.2 Reliability
Several items of the procrastination and dyadic job performance scale had to be
recoded as they contained counter-indicative items. For the procrastination scale these were
items 5, 7 and 9 and for the dyadic job performance scale items 5 and 6. After recoding the
items, the reliabilities for the scales were calculated. Reliabilities, scale means and
correlations are depicted in Table 2. Except for the steady action style of the employee (α = .51), all reliabilities are above the agreed cut-off point (with the supervisor) of a Cronbach’s
