The influence of virtualness on faultline perception in a
change context
MSc Master Thesis
MSc BA: Change Management
April 2014
JAN WILLEM DOUMA
Student: 1705598 Damsterdiep 17 9711 SG Groningen E-mail: jwillem@gmail.com Phone: +31630987558 Rijksuniversiteit Groningen
Faculty of Economics & Business
Supervisor: J. Rupert
1
ABSTRACT
This study examines the influence of virtualness on the perception of faultlines in teams currently exposed to change, by analyzing the data of 254 respondents. During this study I will examine to what extent the degree of virtualness influences the way subgroups will be formed in teams that are currently exposed to change. I will investigate if faultlines get perceived differently when teams operate in a different way and to what extent perceived faultlines influence a team’s performance. No significant effects of virtualness could be discovered during this study. On the other hand, using linear regression analyses, a negative influence of both faultlines bases on team performance is discovered. The effects become out shadowed however, as soon goal similarity is added as a control variable. Goal similarity unexpectedly shows to have a negative relationship with faultline perception and a positive relationship with team performance.
2
1. INTRODUCTION
The introduction of IT technology, almost instant electronic communication and an increased globalized (labour) market; introduced a new phase to today’s world that causes more and more organizations to make a shift from face-to-face teams to virtual teams. There is still little consensus about how to define virtual teams, most researchers do however agree that one of the key features of ‘virtualness’ is the relative absence of face-to-face contact (Griffith, Sawyer & Neal, 2003). Virtual interactions reduce the emphasis on visible and tangible dimensions that traditionally defines a team, instead it emphasizes togetherness based on members’ perceptions of belonging (Wiesenfeld, Raghuram & Garud, 2001). Identification in virtual teams thus promotes a sense of togetherness despite a relative lack of physical contact (Pratt, 2001). Various degrees of virtualness can be experienced on a continuum between just face-to-face to pure virtual teams (Townsend, DeMarie & Hendrickson, 1998). Fiol & O’Connor (2005) define virtualness as the extent of face-to-face contact among team members (encompassing amount as well as frequency of contact) and suggest that technological support and dispersion represent tendencies, rather than definitional attributes of virtual teams.
An important facet within virtual teams is the increased range in which teams can operate, because geographical boundaries have become smaller. Mainly because communication can still take place even when members are geographically dispersed (Peters & Karren, 2009). Without these geographical boundaries, chances of diversity and thereby heterogeneity within teams increase. The definition for diversity used here is; “The differences between individuals on any attribute that may lead to the perception that another is different from themself” (van Knippenberg, Homan & De Dreu, 2004: p1008).
3
(2005). They also argue that these differences can improve creativity and problem solving within teams. In virtual teams however, surface-level observable characteristics will often be less observable at the moment a team is formed. Hence this could influence the group dynamics in teams. It will be examined which categories are of most influence when dealing with diversity in teams with a certain degree of virtualness during this research.
Departing from prior diversity research, Lau & Murnighan (1998) came up with so called faultlines within groups. Instead of looking at a group from a single attribute perspective, groups are being looked at as a complex bundle of demographics, where each person in a group belongs to many subgroups such as gender, race, education and age (Gibson & Vermeulen, 2003; Bezrukova, Jehn, Zanutto, & Thather, 2009). A faultline can be defined as “A hypothetical dividing line that splits a group into relatively homogeneous subgroups based on group members’ demographic alignment among different attributes” (Lau & Murnighan, 1998, p.328). When more attributes are aligned with each other in the same way, the strength of a faultline will increase. During this research I will examine to what extent the degree of virtualness in teams plays a role in which and how faultlines get activated and will possibly form subgroups.
So far most literature around faultlines has focused on the forming of subgroups in traditional teams, or at least did not take the degree of virtualness into account (eg. Lau & Murnighan, 1998; Bezrukova et al., 2009). According to previous faultline research subgroups can be formed by demography, professional and organizational affiliations, and psychological factors such as personality and values. Recently Polzer, Crisp, Jarvenpaa & Kim (2006) added geographic location as a dimension of diversity. One of their findings was that geographic distance between subgroups evoked less trust and more conflict when the members of the subgroup were homogeneous in nationality (Polzer et al., 2006).
4
2003; Bezrukova et al., 2009), are also valid for teams with a certain degree of virtualness. According to Griffith & Neale (2001) virtualness can be measured on two different dimensions namely; the level of technology support and the time team members spent apart from each other. If the degree on one or both dimensions increases, the virtualness of a team will also increase.
During this study; organizations that just recently went through a change or are still busy implementing change will be used to examine the effects of virtualness on the faultline activation process. This change could be to one of the dimensions mentioned by Griffith & Neale (2001), but is not necessarily. Basically the whole context of my research will happen in an environment exposed to some sort of change. The reason that this research has to happen in a changing environment is because most previous research does not take the constantly changing environment into account. Nowadays it’s simply not possible to work in a stable / static environment, so this way the research will match more closely to situations organizations and specifically their teams have to deal with.
The main theoretical purpose of this study is to extend faultline literature which is still mainly focused on traditional teams. Fiol & O’Connor (2005) talk about the explosive growth in the use of virtual teams that begs for a deeper understanding of identification process in virtual teams, both hybrid and pure. Staples & Zhao (2006) thereby mention that it could be valuable to examine the different degrees of heterogeneity to see if and when faultlines develop, in their research about the effects of cultural diversity in virtual teams vs. face-to-face teams. In order to make a contribution the following research question will be answered:
“How will the degree of virtualness influence faultline perceptions and team performance for teams exposed to change?”
5
2. THEORETICAL FRAMEWORK
This section is written to give a literature review of the concepts used during this study. First the change context is shortly introduced, followed by a literature review of the concepts virtualness, faultlines, and team performance resulting into four hypotheses and a conceptual model, see also figure 2, p11.
Change Context
“For society at large, and organizations in particular, the magnitude, speed, impact, and especially the unpredictability of change are greater than ever before” (Burnes, 2009). According to McKinsey & Company (2008) organizations need to change constantly. So the context of organization change is frequently experienced by most organizations that exist nowadays. For research purposes it’s relevant to expand upon what is known about team behavior during change. So during this research; change will be the context that teams have to operate in. Change can influence the degree of virtualness by introducing new ways of working for a team. If for example a change makes teams rely more on electronic communication, the virtualness could increase. On a side note; the degree of virtualness before a change can also influence how team members perceive a change, due to the fact team members don’t see each other very often or not at all. Team members could cope differently with the change and start behaving in contradicting ways to each other, which could eventually affect a team’s performance. Therefore it’s important to take the scope of change into account to see in what extent change affects a team member’s ability to work effectively (Balogun & Hailey, 2008).
Virtualness
6
frequency of contact) and thereby suggests that technological support and dispersion represent tendencies, rather than definitional attributes of virtual teams. Griffith & Neale (2001) thereby argue virtualness is influenced by two dimensions, namely: the time team members spent working apart and the level of technology support they employ, see also figure 1.
FIGURE 1.
In addition Berry (2011) argues that nowadays almost all teams use technology to some degree, but for virtualness to increase; reliance on electronic communication has to increase. Pazos, Chung & Micari (2013) even argue that technology is taking a leading role in supporting communication and task completion in organizations. The environment is demanding organizations to transform from traditional ways of working to becoming more flexible and adaptive (Nyström & Asproth, 2013). These transformations go paired with an increased level of technological support in most teams, which increases the chances of team members to interact with other team members in some sort of virtual way. The level of technology support can best be described as the amount and frequency team members use different kind of ways to interact and perform as a team, relying on electronic dependent systems, for instance; E-mail, instant messaging or Skype.
Some systems of technology can be used in a synchronous (instant) way eg. Video-chat, while other systems will be asynchronous (time-delayed), for example E-mail (McGrath & Berdahl, 1998). Both ways have their own advantages and disadvantages, for instance asynchronous communication enables flexibility, since team members can receive information when it’s
7
most convenient for them. Synchronous communication on the other hand makes real-time conversations possible, either face-to-face or in a virtual way. The speed of interactions will be influenced based on the communication system that is used. Speed of interaction of the communication system will also influence the degree that teams still have to rely on face-to-face meetings to be able to function. As soon as the need for face-to-face-to-face-to-face meetings decreases, the chances become high also the frequency and time of team members spending apart will increase.
More and more authors write about the influence of face-to-face contact, or the lack of it, on how teams will behave. This can be more-or-less directly compared to the degree of virtualness, since important factors in virtualness are the amount and frequency of face-to-face contact. Griffith, Sawyer & Neale (2003) state that teams who never met face-to-face-to-face-to-face are different in a non-lineair way than those who do meet, even if only occasionally. According to Watson, Kumar & Michaelsen (1993) different national backgrounds, each with its own set of values, orientations, and priorities, can detract from effective internal communication. Cramton, (2001) states geographically separated team members lack "mutual knowledge" of each other's situations, increasing coordination problems in acquiring knowledge and resources. Finally Staples & Zhao (2006) say that by minimizing the salience of surface level diversity, by avoiding face-to-face meetings in early life of a team, may reduce potential negative influence of diversity.
Faultlines
To define a faultline according to Lau & Murnighan (1998:p.328): “A hypothetical dividing line that splits a group into relatively homogeneous subgroups based on group members’ demographic alignment among different attributes”. Instead of looking at a group from a single attribute perspective, groups are being looked at as a complex bundle of demographics, where each person in a group belongs to many subgroups such as gender, race, education and age (Gibson & Vermeulen 2003; Bezrukova, Jehn, Zanutto, & Thather 2009).
8
dormant state to an active one, individual attributes have to align with other team members and form a homogeneous subgroup, which is actively perceived by the team members (Lau & Murnighan, 1998). Most of the time in order for a faultline to become activated some sort of trigger is needed for team members to realize their team is split into several homogeneous subgroups. A trigger is defined by Chrobot-Mason, Ruderman, Weber & Ernst (2009) as: “An event involving two or more people from different social identity groups that ignite a replication of societal-based identity threat in an organization”. A great variety of triggers can cause dormant faultlines to become activated within a team. During this research I will examine to what extent the degree of virtualness will trigger faultlines to become activated while implementing change.
Many different types of objective faultlines can exist within teams. First of all social category faultlines: race/ethnic background, nationality, sex, and age (Lau & Murnighan, 1998); secondly information based faultlines: work/education and experience/expertise (Jackson, Joshi & Erhardt, 2003); finally non-demographic faultlines like: personal values and personality (Molleman, 2005). All these faultlines can exist within a team, without it being necessary that these faultlines become activated. How strong an activated faultline will be is based on three factors, namely: the degree of diversity in a team, the strength and the distance of the faultlines (Lau & Murnighan, 1998). Here strength could best be described as the more attributes that align themselves in the same way the higher the strength will be (Lau & Murnighan, 1998). As Lau & Murnighan (1998) give a clear example to illustrate this; when in a group all women are 60 years old and all men are 30 years old, both age and gender align and form one single stronger faultline. In this example also distance is illustrated; since distance is the degree that attributes diverge from each other, in this case 30 years and 60 years are further away from each other than for example 30 and 35 years.
9
used can have an influence on how visible cues can be, that may highlight subgroup faultlines. If however observability is low, chances of social category based faultlines to become activated will possibly decrease as well. This could mean that the team members will rely more on information based attributes like education, experience and expertise. So I suggest that in teams with a higher degree of virtualness, chances increase that subgroups will be formed based on information based faultlines instead of social category faultlines. I thereby propose the following hypotheses:
Hypothesis 1: The higher the degree of perceived virtualness, the higher the strength of
information based faultlines will be perceived
Hypothesis 2: The higher the degree of perceived virtualness, the lower the strength of social
category faultlines will be perceived. Team performance
Bezrukova et al., (2009) found that groups with activated faultlines were more likely to form coalitions and bring more conflict compared to groups with dormant faultlines. Conflict can interfere with team performance and reduce satisfaction because it can provide tension, antagonism and distract team members from performing their tasks (De Dreu & Weingart, 2003). Conflicts can cause negative process problems such as a lack of coordination, cooperation and cohesion (Brewer, 1995) and thus also negatively influence team performance.
10
extreme strong contrast between subgroups, so that there are no commonalities between team members, will cause overall team performance to decrease.
Carton & Cummings (2013) speak of two different types of subgroups, namely: identity based subgroups that are based on social category characteristics and knowledge based subgroups that are formed based on information based characteristics. To compare this with faultline bases, identity based subgroups are formed by perceived social category faultlines and knowledge based subgroups will be formed by information based faultlines. Carton & Cummings (2013) also argue that knowledge-based subgroups can increase performance, because there will be a broader base of knowledge that a team can draw from before it’s integrated and decided by the whole team (Carton & Cummings, 2013). Identity-based subgroups do however generally negatively impact team performance because it leaves little ambiguity to who is in-group and who is out-group as soon as subgroups are based on a clearly observable social characteristic. This can then cause subgroups to develop a hostile attitude to the other subgroup by seeing them as an out-group.
According to Bezrukova et al. (2009) strength of social category faultlines is negatively related to team performance. However, there was no evidence that information-based faultlines were negatively related to performance. To quote Gibson & Vermeulen (2003): “Faultline strength may cause damage in groups with social category faultlines, but it may also promote healthy competition, stimulate information elaboration, and be beneficial for the groups with information-based faultlines”. Based on the similarities between the effects of social category faultlines to identity-based subgroups and information based faultlines to knowledge-based subgroups, I propose the following hypotheses:
Hypothesis 3: Information based faultlines will be positively related to a team’s performance. Hypothesis 4: Social category faultlines will be negatively related to a team’s performance.
12
3. METHODS
The following section is written to give a clear overview how this study is conducted. First the data collection procedure will be explained, followed by an indication of the data sample, thirdly the measures will be explained and finally the factor analyses of both main variables and control variables will be presented.
Data collection procedure
13
The questionnaire team members have to fill in, is made out of 35 7-point Likert scale items, see also appendix 3, and additional questions for demographics such as: age, gender, education, team tenure, duration of change. The range of each item will be between (1) completely disagree to (7) completely agree. This way it’s possible to measure the intensity and direction of each individual’s attitudinal composition regarding each factor that’s measured within the questionnaire (Matell & Jacoby, 1971). To prevent a non-response bias; the questionnaire is anonymous and individual results will not be linked back to managers, only overall findings will. Next in order to prevent common method bias a separate questionnaire for the team leader of each team will be used. Common method bias can best be described according to Podsakoff, MacKenzie & Lee, (2003) as: “Common method biases arise from having a common rater, a common measurement context, a common item context, or from the characteristics of the items themselves”. Because the questionnaire for the team leader will contain different items, a common rater is prevented in this study.
Sample
In total 508 respondents working in 37 different organizations distributed between 94 teams filled in the questionnaire, after performing a missing values analysis, 225 respondents got deleted due to missing values on key items such as virtualness and faultline items. Next due to low response rate on a team level (< 60%), another 29 respondents got removed from the data set. So in the end a total of 254 respondents working across 15 organizations spread between 36 different teams are used to make further analyses. To give a small impression of the final sample group; 47 percent of the respondents is male and 53 percent is female. The average age is 43 years old (SD = 11.3), normally distributed between the age of 17 and 62. The majority of the respondents have a Dutch nationality (97%), while 92 percent of the respondents are born in the Netherlands as well. The highest percentage of the sample group has an MBO degree, namely 39% of the respondents. Other education degrees are represented as following: HBO 28% and University 27% and finally 6% got educated on a different degree. In total 52% of the respondents has a full-time contract, while 48% has a part-time contract.
14
leaders is 47 years old (SD = 10.4). Every team leader is born and raised in the Netherlands. 80 percent of the team leaders followed higher level of education (HBO) or higher, with an average work experience of 17.6 years (SD = 11.1). Every team leader works in a full time position with an average team tenure of 36 months where 40 percent of the leaders works for the team a year or shorter. The sample size is too small to perform any further statistical measurements with the data of team leaders, since it doesn’t meet Pallant’s (2010) first criterion for a factor analysis; that sample size should be bigger than 150 and one should have at least five cases for each item.
On a team level data of 36 teams spread around 15 different organizations was collected. In these teams the average team tenure was 3.83 years (SD = 5.58) ranging between less than a year to 37 years of experience in the team. On average team members come together with their (whole) team 2.2 times a month (SD = 2.75) or 16.4 times a year (SD = 24.17). This is distributed somewhere between 0 times a month and 20 times a month with a skewness to the right of 4.0. This high skewness is due to the fact that some respondents meet 20 times a month and more than 50% of the respondents only meet with their whole team 1 time a month or less.
15
TABLE 1. Reasons for change
Reasons for change Frequency Percentage
1. To maintain continuity of organization. 114 21%
2. It was obligatory (due to laws, external parties). 43 8%
3. To become more adaptable to the external environment. 64 12%
4. Cost reduction. 74 13%
5. To become (more) competitive. 33 6%
6. Improve reputation of organization 29 5%
7. A similar change was successful in a similar organization. 11 2%
8. Economic crisis. 25 5%
9. For growth. 41 7%
10. To improve internal communication / team work. 85 15%
11. Different… 30 6%
Total 549 100%
Measures
Virtualness
In order to measure virtualness, both dimensions of Griffith & Neale (2001) will be measured in the questionnaire, namely: the time team members spent apart from each other and the level of technology team members’ use. The level of technology is measured based on a 7-point Likert scale ranging between make no use to very often use of technology, using items like; eg. “How many times do you have face-to-face contact with your team members?” and “How much do you make use of real-time online discussions, like chat or instant messaging services?” These items were adapted from an 18-item scale measuring virtuality, or here virtualness, used by Chudoba et al. (2005). The time team members spent apart from each other is already pre-tested during a similar study last year.
16
frequency items and the use of a different scale (a month / a year), items are recoded into one ordinal scale. A 4-point scale is developed between 1 = zero interaction and 4 = much interaction, see also appendix 3 for the exact scale. Due to a lot of missing values N is only 158 for interaction frequency, this scale will not be used during hypotheses testing because too many valuable respondents in other scales will be lost. During this study respondents somehow seemed to have an issue recalling the frequency they interact with their whole team. The perceived reliance on electronic communication, an important factor to virtualness (Berry, 2011), will also be measured on a 7-point Likert scale ranging between no perceived reliance to very high reliance on electronics. Questions will be more in terms off; “How many times is your team dependent on electronic communication systems, to be able to function?” Here electronic communication systems can best be described as any communication system that is dependent on electronics, eg. Skype, e-mail, instant chat services, social media and so on. The items are adapted from Chudoba’s et al. (2005) study.
Since all other variables used in this study were pre-tested for reliability and validity, in a similar study last year, I’ve decided not to do a pre-test for just the four items measuring virtualness, due to time constraints this study is facing.
Objective faultlines
Objective faultline bases will be obtained using a managerial grid, which can be found in appendix 1, combined with questions about age, gender, nationality, education and work experience in the questionnaire. The demographic characteristics of every team member must first be known in order to calculate objective faultline scores.
17
subgroups by calculating the ratio of the between group sum of squares to the total sum of squares” (Thatcher et al., 2003. p225):
( ∑ ∑ ( ̅ ̅ ) ∑ ∑ ∑ ( ̅ ̅ )
)
In this formula indicates the value of the characteristic of the member of subgroup . The overall group mean of characteristic of the is presented in ̅ .and the mean of characteristic in subgroup is denoted by ̅ . Finally, presents the number of members of the subgroup under split . The strength of faultlines division presented by , is then calculated as the maximum value of over all possible splits . For small group sizes it is possible to calculate by enumerating all possible splits and calculating the maximum .” (Thatcher et al., 2003. p225). The faultline strength can take on values between zero and one, with larger values indicating greater strength (Thatcher et al., 2003).
Perceived faultlines
18
Team performance
Team performance will be measured based on individual perceptions that team members have on the effectiveness and performance of their team. Next to team member perception, also team leader perception about team performance will be measured, in order to get a more trustworthy view of a team’s performance than based on individual team member perceptions on themselves. There will be several items that measure the perceived team performance for each individual based on the study of Jehn & Northcraft (1999). To measure performance, respondents have to answer the following items according to their perceptions: “My team performs well” and “My team works effective”. Both measured on a 7-point Likert scale ranging between totally disagree to totally agree. Team performance perception is necessary in order to see if teams with activated social category based faultlines perform better or worse compared to teams with activated information based faultlines.
Control variables
In order to make sure that the team definition used during this study is met by respondents I make use of the following control variable scales: task type, team interdependence and goal similarity. Also team size and team tenure and individual demographics are measured to create a clear profile of each team. Task type is a 4-item scale based on Jehn’s (1995) scale with α = .88. An example of the items used is: “How much variety is there in your job?” The scale is measured on a 7-point Likert scale ranging from not often to very often.
Task interdependence is a 3-item scale, best described as the way team members interact and depend on one another to accomplish the work (Campion, Medsker & Higgs, 1993). An example of the items used is: “Within my team, jobs performed by team members are related to one another”. For both task interdependence and goal similarity a 7-point Likert scale ranging between totally disagree to totally agree is used. Goal similarity is based on a 3-item scale pre-tested by Jehn (1995) that had α of .83 in that study. An example of the items used here is: “As a work unit, we have similar goals”.
7-19
point Likert scale that’s ranging between totally disagrees to totally agree. All items used can be found in appendix 3.
Factor analyses
A factor analysis will be conducted in order to see if the items of the scales indeed load on the same factor. In this case a rotated Varimax factor analysis will be conducted, see also table 2. Extraction is based on eigenvalue above 1. The minimal loading criteria for each factor is 0.45 and whenever double loadings happen at least 0.2 in-between difference has to be there in order for a double loading to be acceptable (Reise, Waller and Comrey, 2000).The factor analysis has to meet several criteria in order to be useful for further analyses. According to Pallant (2010) a good overall sample used for a factor analysis should contain more than 150 cases (respondents) and at least 5 cases for each item used during the analysis. The factor analysis contains 20 items, so at least 150 respondents should be present (Pallant, 2010); since there are 254 respondents present, the first criterion is met. It’s also important to assess the factorability of the data by looking at the Barlett’s test of sphericity (Barlett, 1954) and the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy (Kaiser, 1970). In this case Barlett’s test of sphericity is .00, which is significant since p is < .05. The KMO index is .75 which is higher than .6 and thereby proves the significance of this factor analysis combined with the other criteria that were met.
TABLE 2. Factor analysis: main variables
Component 1 2 3 4 5 6 Perceived subgroups FAU1 .061 .898 .038 -.090 .133 .050 FAU2 .038 .901 -.017 -.061 .165 -.006 FAU3 .001 .756 .256 -.020 -.127 -.003 FAU4 .003 .833 .082 -.117 .238 -.019
Social category faultlines
20
Since the prior criteria were met, the extracted components can now construct the scales for further analyses. In total seven components with an Eigenvalue above one, covering a total variance of 74.2% was extracted by the Varimax rotation. The first component is containing the 4-item scale for perceived subgroups, that has high loadings and α = .88. The second component is responsible for the scale construction of the 4-item scale for social category faultlines that also has high loadings and α = .84. Thirdly information based faultlines forms a scale based on its high loadings of both items FAU10 and FAU11, with α = .80. The original scale consisted of 3-items, but FAU9 (education) did not load in the initial factor analysis, see also appendix 2, thus got disposed in the final factor analysis. The fourth scale constructed by
FAU6 (age) .053 .099 .737 -.069 .350 .138
FAU7 (culture) -.009 .127 .818 -.060 .144 .056
FAU8 (nationality) -.075 .089 .869 -.021 -.004 -.035
Information based faultlines
FAU10 (expertise) .083 .215 .216 -.168 .816 -.039
FAU11 (work experience) .050 .156 .357 .002 .803 -.024
21
the factor analysis is team performance, which has α = .85 and high loadings on the same component. As fifth the control variable scope of change forms a scale with α of .91 and high loadings on the same component. The sixth scale is containing two items from the 4-item scale virtualness, because the other two items did not load correctly in the initial factor analysis, see also appendix 2. A critical side-note has to be made, that also the final two items, that do have high loadings on the same component, have a combined α = .35, which is not reliable. The full scale has α .05 and if virtual1 is removed α becomes .36, both still unreliable. Ideally the Cronbach alpha of a scale should be above .7, but everything above .6 is reliable (DeVellis, 2011). This phenomenon can be theoretically explained, because the original scale contained 18-items and now there are only two items left, which can decrease the scale’s reliability with a significant amount. Thereby according to Pallant (2010) it can be difficult to get a decent Cronbach alpha with scales smaller than 10 items. The virtualness scale (virtual2, 3) will still be used for further analyses based on its theoretical significance and high factor loadings, but I will continue about this reliability issue in the discussion section.
To see if respondents indeed match the definition of a team that is used during this study, control variables were used. Another rotated Varimax factor analysis for team control variables was conducted, see also table 3.
22
Four components got extracted with eigenvalues exceeding 1, explaining 31.5%, 19.9%, 14.2% and 11% of the variance respectively. The Kaiser-Meyer-Olkin value was .71, which is higher than .6 (Kaiser, 1970) and also Bartlett’s Test of Sphericity (Bartlett, 1954) reached statistical significance, this supports the factorability of the correlation matrix.
Task type 1 and 2 form a scale based on the factor analysis and task type 3 and 4 form a scale. Task type 1 and 2 have α of .71 which is >.6, with other words the scale is reliable. On the other hand Task type items 3 and 4 have a combined α of .21 which is way below .6, so not reliable and thereby disposed, which is contradicting to the theoretical significance (α = .88) of the total scale (Jehn, 1995). The combined α of the four task type items is .28, this could be due to the fact items three and four measure totally different constructs within the task type scale. Item three measures how often someone can predict how long a task will take and the fourth item measures how much a job includes problem-solving, while items one and two both measure the degree of variety there is in a team.
Task interdependence can be combined into one scale, since all items load on the same component and α of .82. Also goal similarity can be combined as a 3-item scale since the factor loadings are all above >.8 while loading on the same component with α of .89.
Taskinterdep3 .227 .780 -.127 -.010
Goal similarity
Goalsim1 .915 .112 .098 .038
Goalsim2 .908 .066 .122 -.045
23
4. RESULTS
This section gives an overview of analyses on an individual level. First the descriptive statistics will be described, followed with correlations between the different variables and finally the results of hypotheses testing will be displayed.
Descriptive statistics
The correlation table can be found in table 4, p25.
Both faultline bases have low means; social category faultlines (M = 1.76, SD = 1.11) and information based faultlines (M = 2.79, SD = 1.69). Which indicates faultlines to be present, were information based faultlines are slightly stronger perceived than social category faultlines.The high mean of team performance (M = 5.14, SD = 1.17) indicates a high amount of highly performing teams. The means for team control variables are; task type (M = 5.44, SD = 1.28) team interdependence (M = 4.75, SD = 1.38) and goal similarity (M = 5.42, SD = 1.17) indicating the existence of real teams in the sample group. Also the high mean (M = 4.56, SD = 1.59) of scope of change indicates that the perceived change is relevant enough for team members to be influenced by it and thereby confirms the change context of this study. Finally the above average mean (M = 4.12, SD = 1.69) of virtualness indicates that; there are teams with different degrees of virtualness present in the sample of this study.
Correlations
The first important correlations are the correlations between perceived faultlines with team performance. Both correlations seem to be negatively related to team performance, respectively: social category faultlines (r = -.24, p <.01) and information based faultlines (r = -.28, p <.01). These correlations indicate a negative relationship between different categories perceived faultlines and team performance. So far this study indicates a higher negative influence of information based faultlines when compared to social category faultlines. To give an indication of faultline perceptions; 40 respondents perceive strong faultlines based on gender (Likert scale > 4), 50 on age, 32 on cultural background, 29 based on nationality, while 107 respondents perceive faultlines based on expertise level and finally 77 respondents perceive faultlines based on work experience.
24
25
TABLE 4. Descriptive statistics and correlation matrix.
M SD N 1 2 3 4 5 6 7 8 1. Goal similarity (CV) 5.42 1.17 241 - 2. Task interdependence (CV) 4.75 1.38 247 .16* - 3. Task type (CV) 5.44 1.28 252 .33** .16* - 4. Scope of change (CV) 4.56 1.59 251 -.03 .07 .01 - 5. Virtualness (IV) 4.12 1.69 254 .07 .22** .21** .26** -
6. Information based fau. (B) 2.79 1.69 253 -.30** .08 -.02 .11 .00 -
7. Social category fau. (B) 1.76 1.11 247 -.28** .03 -.03 .05 .03 .82** -
8. Team performance (DV) 5.14 1.17 249 .60** .04 .32** -.14* -.06 -.28** -.24** -
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
26
Hypotheses testing
The hypotheses were tested on an individual level using both linear (Model 3) as multiple regression analysis (Model 1 and 2). Regression analyses allow one to assess the relationship between one dependent variable and several independent variables (Pallant, 2010). In this case the first hypothesis: “The higher the degree of perceived virtualness, the higher the strength of information based faultlines will be perceived”; was tested. The output of the first regression analyses is displayed in table 5.
TABLE 5. Regression analyses: Hypothesis 1
Model 1 Model 2 Model 3
DV: Information based faultlines
Independent variables β Sig. β Sig. β Sig.
Task type .09 .22 .10 .16 Goal similarity -.34 .00 -.34 .00 Task interdependence .12 .08 .13 .16 Scope of change .08 .24 .09 .05 Virtualness -.07 .29 .00 .95 Adjusted R2 .10 .10 -.00 R2 .11 .12 .00 F 6.83 5.68 .00
27
present since there are significant correlations between the different variables, see also table 4, p25. The only significant outcome of the regression analysis is goal similarity with a β of -.34 when p < .001, which shows a negative relationship between goal similarity and information based faultlines. The adjusted R2 grew from 9.6% to 9.7% when adding virtualness, which is a very minimum amount of growth. In the linear regression analysis used in Model 3 virtualness neither proved to have a significant relationship with information based faultlines (β = .00, p = .95). The first hypothesis will thereby be rejected based on the p-value of virtualness (p = .29) and (p = .95) depending on which model was tested during the analysis.
The second hypothesis: “The higher the degree of perceived virtualness, the lower the strength of social category faultlines will be perceived”; was tested using multiple regression analyses (Model 1 and 2) and using linear regression analysis (Model 3), see also table 6.
TABLE 6. Regression analyses: Hypothesis 2
Model 1 Model 2 Model 3
DV: Social category faultlines
Independent variables β Sig. β Sig. β Sig.
28
During the second regression analysis social category faultlines formed the dependent variable, while virtualness was the independent variable controlled by task type, goal similarity, task interdependence and the scope of change. The only significant outcome was goal similarity with a β of -.32 when p < .001, which shows a negative relationship between goal similarity and social category faultlines. When adding virtualness to the model the adjusted R2 decreases from 9.6% to 6.7%. So in this case adding virtualness to the regression analysis makes the total explained variance decrease. When conducting a linear regression analysis, see also Model 3, the relationship between virtualness and social category faultlines did not prove to be significant either (β = .03, p = .63). The second hypothesis will thereby be rejected because the p-value of virtualness is not significant (p = .89) and (p = .63) depending on which model is tested.
The third hypothesis: “Information based faultlines will be positively related to team performance”; was tested using multiple regression analyses (Model 1 and 2) and using linear regression analysis (Model 3), see also table 7.
TABLE 7. Regression analyses: Hypothesis 3
Model 1 Model 2 Model 3
DV: Team performance
Independent variables β Sig. β Sig. β Sig.
Task type .12 .04 .13 .03
Goal similarity .57 .00 .53 .00
Task interdependence -.08 .14 -.07 .02
Scope of change -.13 .02 -.13 .12
Information based faultlines -.11 .06 -.28 .00
Adjusted R2 .38 .39 .08
R2 .39 .41 .08
29
In the third regression analysis team performance was used as the dependent variable and information based faultlines was added as independent variable controlled by task type, goal similarity, task interdependence and the scope of change. The same principles were used; first a regression analysis was conducted with only the control variables present, see also Model 1. In Model 2 information based faultlines are added to the analysis, this makes total explained variance increase from 38.4% to 39.3%. Model 3 tests the linear relationship between information based faultlines and team performance to see if the control variables out shadow the effects of information based faultlines. Which indeed seems to be the case here, now information based faultlines have a significant relationship of -.28 with team performance. The only significant outcome in the multiple regression analysis is goal similarity with a β of .53 when p < .001, which shows a positive relationship between goal similarity and team performance. Based on the multiple regression analysis the third hypothesis will be rejected, because the p-value of information based faultlines is not significant (p = .056 > .05). Based on the linear regression analysis tested in Model 3 however; the relationship proves to be significant (β = -.28, p = .00) but contradicting. The results show a negative relationship between information based faultlines and team performance, while the hypothesis indicates a positive relationship. So the hypothesis will be rejected.
The fourth hypothesis: “Social category faultlines will be negatively related to team performance”; was tested using multiple regression analyses (Model 1 and 2) and using linear regression analysis (Model 3), see also table 8, p30.
30
TABLE 8. Regression analyses: Hypothesis 4
Model 1 Model 2 Model 3
DV: Team performance
Independent variables β Sig. β Sig. β Sig.
Task type .12 .04 .12 .04
Goal similarity .57 .00 .55 .00
Task interdependence -.08 .14 -.08 .16
Scope of change -.13 .02 -.14 .01
Social category faultlines -.08 .18 -.24 .00
Adjusted R2 .38 .40 .01
R2 .39 .41 .06
F 36.32 29.03 13.43
31
5. DISCUSSION
This section is written to give my interpretation on key findings, implications, limitations and further research. In this study the effects of virtualness are examined by investigating the following question: “How will the degree of virtualness influence faultline perceptions and team performance for teams exposed to change?” To be able to answer this question hypotheses have been tested, which will now be further discussed.
Key findings
The first important finding is that both social category faultlines and information based faultlines have negative correlations with team performance. A linear regression analysis shows a significant negative relationship between social category faultlines and team performance. This confirms previous literature that indicates the same relationship between social category faultlines and team performance (eg. Jehn & Bezrukova, 2010; Gibson & Vermeulen, 2003). In the multiple regression analysis, however the effects of social category faultlines were out shadowed by the effects of the control variable goal similarity. This means that the higher goal similarity is in teams, the lower the perceived strength of social category faultlines will be. So there seems to be a relationship between social category faultlines and team performance, but the relationship is not so strong and measurable as previous literature claimed it to be.
32
An unexpected finding is that goal similarity (Jehn, 1995) has a significant relationship with all dependent variables used in this study during the multiple regression analyses. First of all goal similarity has a negative relationship with information based faultlines and also with social category faultlines. This indicates the higher goal similarity is within teams, the less faultlines will be perceived by individual team members. The high mean of goal similarity will cause faultline strength to be perceived less and can thus explain the fact that both faultline bases got out shadowed during all multiple regression analyses. Besides this fact the low means on both faultlines bases could also be due to the fact not every team member perceives the same type of faultline at once. For example one team member can perceive a faultline based on age while not perceiving a faultline based on gender, still both faultlines are of the same (social) category, resulting into mixed scores within this category.
Secondly goal similarity has a positive relationship with team performance, in other words the higher goal similarity is the higher a team’s performance will get. These findings prove both the relevance of having control variables, but also show an unexpected role that goal similarity has on both faultline perception and team performance. So the higher the agreeability and similarity of goals among team members the higher team performance will become.
33
Objective faultlines did not seem to show significant outcomes during this study, therefore were excluded from the results section, perceived faultlines did however show significant outcomes. This shows the importance of taking perceived faultlines into account, besides taking objective faultlines into account. This study also confirms perceived faultlines can actually have an influence on team performance, as was already shown in previous studies for instance; Jehn & Bezrukova (2010) and Gibson & Vermeulen (2003).
During this study the context is change, therefore scope of change is being controlled during all multiple regression analyses. The average respondent perceives scope of change above average, there is no significant relationship with any of the dependent variables. So no direct effects of change are measured on the faultline perception or team performance among team members.
Theoretical implications
The main theoretical contribution of this study is that both faultline bases seem to have a negative relationship with team performance. So first it can be confirmed social category based faultlines do have a negative influence on team performance, as was already argued by (Jehn & Bezrukova, 2010; Gibson & Vermeulen, 2003). The strength of this relationship should be taken into account though, since effects are in this case easily out shadowed by the control variables that were used in this study. Secondly the negative relationship between information based faultlines and team performance is contradicting with previous literature that claims this relationship is positive (Gibson & Vermeulen, 2003).
Another theoretical contribution of this study is that it takes change into account. Since organizations often deal with a turbulent environment, it’s important to control the effects of such an environment. In this case no significant relationship between changes and both faultlines and team performance could be proven. Since the generalizability of this finding is restricted to the used data sample, the influence of change should be further explored during future research.
34
means that teams with high goal similarity will experience less faultline strength, so also the negative effects that activated faultlines have on team performance (Jehn & Bezrukova, 2010) could be perceived less.
Managerial implications
When starting to implement a new type of change into an organization, characteristics that are now proven to influence faultline perception can be taken into account before reorganizing or merger teams. Independently from the phase of change a team is facing, team leaders should always try to increase goal similarity in teams. First by creating a working environment that will increase goal similarity within teams, negative influences of perceived faultlines will be reduced. This can be accomplished by increasing the similarity of goals, but also by increasing the degree team members agree upon what’s important to the group. So a team leader can test to what extent team members agree with each other but also how similar current goals are within teams. Secondly, according to the outcomes of the multiple regression analyses testing hypotheses three and four, overall team performance will be increased the higher goal similarity will get.
By using the algorithm to calculate objective faultlines (Thatcher et al., 2003), team leaders could flag potential harmful team compositions, based on their demographic characteristics beforehand. That way teams can be formed based on the right amount of diversity while maximizing effectiveness. But since the algorithm is quite complicated to start using, managers could now also quickly identify faultline perception by re-using the items that were used during this study. It will be important to identify strong perceived faultlines that could potentially cause this harm. A team leader could also try to prevent the emergence of subgroups by identifying team members that are highly similar on the same attributes, for example by looking at demographics obtained through the managerial grid, see also appendix 1. So an example what one could prevent when compiling a new team is the composition of two 30 year old women and two 60 year old men or similar corresponding faultlines within groups.
Research limitations and further research
35
not to do another pre-test just for the four extra items of virtualness was made due to time constraints. A limitation to this study is that virtualness is used as a scale in the analyses, but both the factor analysis and the reliability analyses show that the scale is not statistically reliable. Two items of the virtualness had to be removed due to wrong loadings in the initial factor analysis, see also appendix 2 and secondly Cronbach alpha between any of the items was never above .6. The relationship between interaction frequency and faultlines was tested but not put into the results section, because only 62% of the respondents gave an answer to their interaction frequency. Thereby no significant beta or adjusted R2 values were found when a regression analysis was actually performed. So for further research purposes a valid scale to be able to appropriately measure the degree of virtualness should be developed, which is first extensively tested for both reliability and validity.
Another limitation is that most teams used in the sample group actually have a low degree of virtualness, which could cover up the effects that virtualness has for teams with a higher degree. Since the current technology wave keeps increasing the use of virtual teams or teams that use a combination between face-to-face contact and virtual contact. Because reliance and use of technology both keeps increasing, it could become easier to obtain data from teams with a higher degree of virtualness during future research.
The other questionnaire that was used for team leaders did not form a great enough sample size to further perform statistical analyses on. Since the ratio team member: team leader is often too high to obtain a reliable sample size, it could be an improvement to obtain leadership data using another research method. This way there will also be triangulation of research methods which will improve the reliability of the study. Triangulation is the combination of multiple sources of evidence that can prevent shortcomings and biases of these instruments by complementing and correcting each other (van Aken, Berends & van der Bij, 2012).
36
37
REFERENCES
Balogun, J., & Hailey, V. H. (Eds.). (2008). Exploring strategic change. Pearson Education. Bartlett, M. S. (1954). A note on the multiplying factors for various χ 2 approximations. Journal of the Royal Statistical Society. Series B (Methodological), 296-298.
Berry, G. R. (2011). Enhancing Effectiveness on Virtual Teams Understanding Why Traditional Team Skills Are Insufficient. Journal of Business Communication, 48(2),
186-206.
Bezrukova, K., Jehn, K. A., Zanutto, E. L., & Thatcher, S. M. (2009). Do workgroup faultlines help or hurt? A moderated model of faultlines, team identification, and group performance. Organization Science, 20(1), 35-50.
Burnes, B. (2009). Management Change: A Strategic Approach to Organisational Dynamics. Prentice Hall-Financial Times.
Campion, M. A., Medsker, G. J., & Higgs, A. C. (1993). Relations between work group characteristics and effectiveness: Implications for designing effective work groups. Personnel
psychology, 46(4), 823-847.
Chrobot-Mason, D., Ruderman, M. N., Weber, T. J., & Ernst, C. (2009). The challenge of leading on unstable ground: Triggers that activate social identity faultlines. Human
Relations, 62(11), 1763-1794.
Chudoba, K. M., Wynn, E., Lu, M., & Watson‐Manheim, M. B. (2005). How virtual are we? Measuring virtuality and understanding its impact in a global organization. Information
systems journal, 15(4), 279-306.
Cramton, C. D. (2001). The mutual knowledge problem and its consequences for dispersed
collaboration. Organization science, 12(3), 346-371.
38
DeVellis, R. F. (2011). Scale development: Theory and applications (Vol. 26). Sage Publications.
Fiol, C. M., & O'Connor, E. J. (2005). Identification in face-to-face, hybrid, and pure virtual teams: Untangling the contradictions. Organization science, 16(1), 19-32.
Gibson, C., & Vermeulen, F. (2003). A healthy divide: Subgroups as a stimulus for team
learning behavior. Administrative Science Quarterly, 48(2), 202-239.
Griffith, T. L., & Neale, M. A. (2001). 8. Information processing in traditional, hybrid, and virtual teams: From nascent knowledge to transactive memory.Research in organizational
behavior, 23, 379-421.
Griffith, T. L., Sawyer, J. E., & Neale, M. A. (2003). Virtualness and knowledge in teams: Managing the love triangle of organizations, individuals, and information technology. MIS
quarterly, 265-287.
Harrison, D. A., Price, K. H., & Bell, M. P. (1998). Beyond relational demography: Time and the effects of surface-and deep-level diversity on work group cohesion. Academy of
management journal, 41(1), 96-107.
Higgs, M., & Rowland, D. (2005). All changes great and small: Exploring approaches to change and its leadership. Journal of Change Management,5(2), 121-151.
Jackson, S. E., Joshi, A., & Erhardt, N. L. (2003). Recent research on team and organizational diversity: SWOT analysis and implications. Journal of management, 29(6), 801-830.
Jehn, K. A. (1995). A multimethod examination of the benefits and detriments of intragroup conflict. Administrative science quarterly, 256-282.
Jehn, K. A., & Bezrukova, K. (2010). The faultline activation process and the effects of activated faultlines on coalition formation, conflict, and group outcomes. Organizational
39
Jehn, K. A., Northcraft, G. B., & Neale, M. A. (1999). Why differences make a difference: A field study of diversity, conflict and performance in workgroups. Administrative science
quarterly, 44(4), 741-763.
Kaiser, H. F. (1970). A second generation little jiffy. Psychometrika, 35(4), 401-415.
Lau, D. C., & Murnighan, J. K. (1998). Demographic diversity and faultlines: The compositional dynamics of organizational groups. Academy of Management Review, 23(2),
325-340.
Mannix, E., & Neale, M. A. (2005). What differences make a difference? The promise and reality of diverse teams in organizations. Psychological science in the public interest, 6(2),
31-55.
Matell, M. S., & Jacoby, J. (1971). Is there an optimal number of alternatives for Likert scale items? I. Reliability and validity. Educational and psychological measurement.
Maznevski, M. L., & Chudoba, K. M. (2000). Bridging space over time: Global virtual team dynamics and effectiveness. Organization science, 11(5), 473-492.
McGrath, J. E., & Berdahl, J. L. (2002). Groups, technology, and time. In Theory and
research on small groups (pp. 205-228). Springer US.
McKinsey & Company. (2008). “Creating organizational transformations”. The McKinsey Quaterly, July, 1-7, available at www.mckinseyquarterly.com
Molleman, E. (2005). Diversity in demographic characteristics, abilities and personality traits: do faultlines affect team functioning?. Group decision and Negotiation, 14(3), 173-193.
Nyström, C. A., & Asproth, V. (2013). Virtual Teams-Support for Technical
Communication?. Journal of Organisational Transformation & Social Change,10(1), 64-80.
Pallant, J. (2010). SPSS survival manual: A step by step guide to data analysis using SPSS. McGraw-Hill International.
Pazos, P., Chung, J. M., & Micari, M. (2013). Instant messaging as a task-support tool in
information technology organizations. Journal of Business Communication, 50(1), 68-86.
Peters, L., & Karren, R. J. (2009). An examination of the roles of trust and functional diversity on virtual team performance ratings. Group & Organization Management, 34(4),
40
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: a critical review of the literature and recommended
remedies. Journal of applied psychology, 88(5), 879.
Pratt, M. G. (2001). Social identity dynamics in modern organizations: An organizational psychology/organizational behavior perspective. Social identity processes in organizational
contexts, 13-30.
Reise, S. P., Waller, N. G., & Comrey, A. L. (2000). Factor analysis and scale revision. Psychological Assessment, 12(3), 287.
Staples, D. S., & Zhao, L. (2006). The effects of cultural diversity in virtual teams versus
face-to-face teams. Group Decision and Negotiation, 15(4), 389-406.
Thatcher, S. M., Jehn, K. A., & Zanutto, E. (2003). Cracks in diversity research: The effects of diversity faultlines on conflict and performance. Group Decision and Negotiation, 12(3), 217-241.
Townsend, A. M., DeMarie, S. M., & Hendrickson, A. R. (1998). Virtual teams: Technology and the workplace of the future. The Academy of Management Executive, 12(3), 17-29.
Van Aken, J., Berends, H., & Van der Bij, H. (2012). Problem Solving in Organizations: A
Methodological Handbook for Business and Management Students. Cambridge University
Press
Van Knippenberg, D., De Dreu, C. K., & Homan, A. C. (2004). Work group diversity and group performance: an integrative model and research agenda. Journal of applied
psychology, 89(6), 1008.
Walsh, J. P. (1988). Selectivity and selective perception: An investigation of managers' belief
structures and information processing. Academy of Management Journal, 31(4), 873-896.
Watson, W. E., Kumar, K., & Michaelsen, L. K. (1993). Cultural diversity's impact on interaction process and performance: Comparing homogeneous and diverse task
41
42
APPENDICES
43
APPENDIX 2. Initial factor analysis (main variables) Component 1 2 3 4 5 6 7 Perceived subgroups FAU1 .069 .050 .898 -.095 .127 .021 .027 FAU2 .043 .001 .902 -.064 .167 -.006 -.024 FAU3 -.006 .260 .748 -.003 -.132 -.062 .109 FAU4 .005 .104 .831 -.117 .233 -.033 .018
Social category faultlines
FAU5 (gender) .017 .770 .003 -.012 .091 -.095 -.088
FAU6 (age) .079 .722 .086 -.080 .308 .037 .080
FAU7 (culture) -.003 .809 .112 -.058 .125 .058 .092
FAU8 (nationality) -.080 .872 .073 -.012 -.018 .007 .070
Information based faultlines
FAU9 (education) -.128 .619 .244 .004 .415 -.047 .061 FAU10 (expertise) .088 .252 .208 -.174 .802 -.067 .023 FAU11 (work experience) .058 .380 .147 -.004 .784 -.044 .020
44
APPENDIX 3. Questionnaire items: team members
Scale Items Reference
Perceived subgroups Adaption of:
Perceived subgroups 1 Perceived subgroups 2 Perceived subgroups 3 Perceived subgroups 4
1. During my job my team often splits into subgroups. 2. During my work my team divides itself into different parts. 3. During team meetings subgroups often sit together. 4. During work subgroups arise within the whole team
Jehn, K.A. and Bezrukova, K. 2010.
Social category faultlines Adaption of:
Social category faultlines 1 Social category faultlines 2 Social category faultlines 3 Social category faultlines 4
1. Subgroups perceived based on gender 2. Subgroups perceived based on age
3. Subgroups perceived based on cultural background 4. Subgroups perceived based on nationality
45
Information based faultlines Adaption of
Information based faultlines 1* Information based faultlines 2 Information based faultlines 3
1. Subgroups perceived based on education (disposed: double loadings) 2. Subgroups perceived based on expertise
3. Subgroups perceived based on work experience
Jehn, K.A. and Bezrukova, K. 2010.
Team performance
Performance 1 Performance 2
1. My team performs well. 2. My team works effective.
Jehn, K. A., Northcraft, G. B., and Neale, M. A., 1999.
Interaction frequency Based on:
Interaction frequency 1 Interaction frequency 2
1. As a team we meet [___] a month
Recoded into 1= zero interaction, 2 = some (1-2), 3 = regularly (3-10), 4= often (10+)
2. As a team we meet [___] a year
Recoded into 1 = zero interaction, 2= some (1-24), 3= regularly (25-120), 4= often (120+)
Griffith, Terri L., and Margaret A. Neale, 2001.
Virtualness Adaption/combination of:
Virtualness 1 R* Virtualness 2 Virtualness 3 Virtualness 4*
1. How many times do you have face-to-face contact with your team members? (reverse) 2. How many times is your team dependent on electronic communication systems, to be able to
function?
3. How many times do you and team members work at different physical locations?
4. How many times do you make use of real-time online discussions, like chat or instant messaging services?
Berry, G. R., 2011.
Chudoba, K. M., Wynn, E., Lu, M., and Watson‐Manheim, M. B. 2005.
46 Task type Task type 1 Task type 2 R Task type 3* Task type 4*
1. How much variety is there in your job? 2. How often is your job boring? (reverse)
3. How often can you predict how long a task will take? 4. How much does your job include problem-solving?
Jehn, K.A. 1995.
Task interdependence
Task interdependence 1 Task interdependence 2 Task interdependence 3
1. I cannot accomplish my tasks without information or materials from other members of my team.
2. Other members of my team depend on me for information or materials needed to perform their tasks.
3. Within my team, jobs performed by team members are related to one another.
Campion, M. A., Medsker, G. J. and Higgs, A. C. 1993.
Goal similarity
Goal similarity 1 Goal similarity 2 Goal similarity 3
1. As a work unit, we have similar goals.
2. The main goals of my work unit are the same for all members in my work unit. 3. We (my work unit) all agree on what is important to our group.
47 Scope of change Scope 1 Scope 2 Scope 3 Scope 4 Scope 5 Scope 6
1. Because of the change the way of working has really changed.
2. Because of the change the responsibilities of team members have really changed. 3. Because of the change the way team members worked together really has really changed. 4. Because of the change different team members now work with each other.
5. Because of the change the team accomplishes their goals in a different way. 6. Team members work in different structure compared to before the change.
Balogun, J., and Hailey, V. H., 2008.
*= removed due to low reliability or low/double loadings during the initial factor analysis.
