Citation
Rupert, J. (2010, April 14). Diversity faultlines and team learning. Kurt Lewin Institute Dissertation Series. Retrieved from
https://hdl.handle.net/1887/15223
Version: Not Applicable (or Unknown)
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the
University of Leiden
Downloaded from: https://hdl.handle.net/1887/15223
Note: To cite this publication please use the final published version (if applicable).
Measuring a Typology
of
Team Learning
Task, Process, and Social Learning
1This chapter is based on Rupert, J. & Jehn, K.A. (2009a) and is therefore written in the first-person plural.
I
n recent years, organizations have structurally changed (e.g., Ellis, Hollenbeck, Ilgen, Porter, West & Moon, 2003). Management layers have been eliminated, jobs have become more complex (Cannon-Bowers, Oser & Flanagan, 1992; Ilgen, 1994), and teams have become important building blocks for learning and innovation in organizations. In order to develop innovative products and to adapt rapidly to changing environments, organizations increasingly depend on teamwork and learning processes in teams (e.g., Wilson, Goodman & Cronin, 2008). Team learning can be defined as the process of reflection and interaction through which team members actively acquire, process, and share knowledge and information, resulting in changes in knowledge, behaviors, and/or actions (based on Argote, Gruenfeld & Naquin, 2001; Sarin & McDermott, 2003; for an overview of definitions see Chapter 1, Table 1). Some researchers have distinguished actual team learning behaviors associated with these concepts, such as asking questions, evaluatingalternatives, seeking feedback, reflecting on results, detecting, discussing, and correcting errors (e.g., Agryris & Schön, 1978; Edmondson, 1999; Gibson &
Vermeulen, 2003; Van der Vegt & Bunderson, 2005) and a few have given examples of what learning can be about (e.g., Hinsz, Tindale & Vollrath, 1997;
Tjosvold, Yu & Chun, 2003; West, Garrod & Carletta, 1997). However, in past definitions and measurements of team learning, no explicit distinction has been made between different types of team learning based on the topic that team learning can be about.
As a response to this gap in research on team learning, Jehn and Rupert (2007) have recently introduced a typology of team learning based on three topics. They proposed that teams can learn about the task they perform (task learning), about work routines and processes (process learning), and about team members’ personalities and other non-work related topics (social learning), which may differently affect group outcomes. For instance, a team can have a high level of information exchange and generate many ideas about the task at hand (which means task learning is high), but if they do not make clear agreements of who is going to do what and how these ideas are being realized, the team still will suffer from ineffective processes (which means process learning is low). Vice versa, a team that has learnt from the past about effective processes to deal with certain problems can still be ineffective when the content of the problem is not completely clear yet for everyone. When team members know each other’s personalities well (which means social learning is high), this can help to improve coordination and interactions between members. However, when team members start to exchange more information
about social matters than about work, for instance, task learning and process learning may suffer.
Until now, the different types of team learning have only theoretically been distinguished (Jehn & Rupert, 2007). No measurements have been developed yet to measure this typology of team learning and its dynamics. The aim of this study is to develop a measurement scale to measure the team learning types in organizations. Based on interviews with managers in
organizations we developed items and conducted a pilot test to validate these items on a sample of individuals reporting about team learning types in their teams. A second aim of this study is to explore the relationships between the team learning types and other related constructs, such as (dimensions of) transactive memory and team innovation. Before elaborating on the pilot study we will first describe the team learning typology in more detail.
A Typology of Team Learning
Other literatures have distinguished different topics that (inter)actions and organizational phenomena can be about, for instance literatures on conflict (Amason, 1996; Jehn, 1995; 1997; Pelled, 1996), team interdependence
(Baumeister & Leary, 1995; Lindenberg, 1997) leadership (Bales, 1958; Burns, 1987), and team mental models (e.g., Klimoski & Mohammed, 1994; Mathieu, Heffner, Goodwin, Salas, & Cannon-Bowers, 2000). In these literatures it is argued and found that the dynamics of these organizational phenomena are different depending on the topic that they are about. For instance, research has shown that different types of conflict (Jehn, 1995; 1997) and different sorts of leaders (Bales, 1958; Burns, 1987) have different effects on followers, group processes, and performance (for meta-analyses see De Dreu & Weingart, 2003, Judge & Piccolo, 2004). Jehn and Rupert (2007) build on these literatures to argue that three topics can be distinguished that teams can learn about: task, process, and social learning. Below we will shortly describe the different types of team learning. As part of the scale development process to measure these different types of team learning, we conducted interviews with eight male and female managers from a multinational bank and a health care organization. We
will illustrate each type of learning with some interview quotes from these managers.
The first type of team learning that can be distinguished is task learning, which is the process of improving the team’s understanding of the content of the task by sharing and reflecting upon knowledge, ideas, and insights through interaction with each other in order to improve team performance. A male health care manager gave the following example of task learning: “The topics (of learning) often have to do with health care. So how do I mobilize someone who is handicapped? These topics are about the content of the work.” Another female manager from the multinational bank said:
“Everyone has his own field of expertise and can contribute to the team’s learning experience by sharing each other’s discoveries. Learning occurs by exchanging what I have discovered, what I did, and what I want to do”. Other managers illustrated that team members increased their professional knowledge as a team “by sharing best practices,” (one manager) and “by sharing ideas and insights about professional strategies in our team and thereby work on strategic issues and vision in meetings (another manager).”
These are examples of task learning, since the team members enhance their understanding of the task by sharing each other's knowledge and expertise about the content of the task and they identify the knowledge and skills needed to accomplish the task and take further action.
Secondly, team members can learn about the process of working together with each other. Process learning can be defined as the pattern of interaction through which team members create effective work routines and develop work procedures in order to increase performance. This can, for instance, be about clarifying team member’s roles (Levine & Moreland, 1999), group objectives and strategies (West et al., 1997) or identifying strengths and weaknesses in their own efforts (Tjosvold et al., 2004). When we asked managers what the topics were that teams learn about, a few different managers
mentioned process learning activities such as “improving communication with each other,” “making records,” “learning about how to work with each other,”
“asking how other people solved things to improve their own performance,”
and “giving each other criticism and suggestions for improvement.” One female health care manager suggested that the team she supervised should have separate meetings to discuss the content of the work and to discuss the process of working together, since “the cooperation with each other as a team is a topic in itself to learn about.” A male manager from the multinational bank
acknowledged the importance, but also the difficulty of process learning, and
said that “improving cooperation and communication is more difficult than learning about task-related topics.” Learning about communication and cooperation, thereby learning how to work with each other, are examples of process learning.
Although in many topic areas it has been acknowledged that
emotional and relational features exist in organizations (Bales, 1958; Borgatta &
Bales, 1953; Chemers, 1997; Lindenberg, 1997), the idea that members can learn about social relationships is an overlooked aspect in team learning research (cf.
Jehn & Rupert, 2007). Social learning is the process through which team members get to know each other better and learn to interpret each other’s behavior in the context of their personal lives and personalities. When we asked managers what the topics were that teams learn about they usually came up with examples of task and process learning first. However, some of them also mentioned examples, such as learning about “each other’s personal lives and problems,” and “how to deal with emotions at work.” One male manager from the multinational corporation mentioned that there is a personal aspect of working together and employees often have a need to share emotions and life situations or personal problems at work. He said: “Service employees are often very people-oriented and emotionally driven. Sharing their emotions and problems with each other helps them in performing their work. In this way, group members learned about each other’s personalities and personal situations, even outside of work”. Also in the health care sector, one male health care manager said: “I believe the profession of a nurse or caretaker also means that you have to learn to deal with your own and other’s emotions.
Knowing what your own pitfalls are, given your personality, and how other colleagues with a different personality deal with emotions can be very helpful in performing health care work.” Through sharing these personal experiences, members get to know each other better as individuals and are better able to interpret each other’s behavior at work, which can positively influence group effectiveness.
Based on the examples from these interviews and past team learning definitions, we developed scales to measure the three types of team learning.
As the second aim of this study was to explore the relationships between the team learning types and related constructs, such as (dimensions of) transactive memory and team innovation, we will first shortly elaborate upon these related constructs. Following that, we will describe the pilot test and the scale
development procedure with final scales.
Related Constructs
Transactive Memory
Transactive memory can be defined as a system in which knowledge from particular group members is combined with an awareness of who knows what (Liang et al., 1995; Wegner, 1987). A transactive memory develops based on a person’s belief about the expertise other team members have combined with the accessibility of that expertise. It can be classified as meta-knowledge:
knowledge about the knowledge of others (Lewis, 2003). A transactive memory system exists when two or more team members use a transactive memory to be able to encode, retrieve, and store information. Three aspects of transactive memory have been distinguished (Liang et al., 1995; Moreland & Myaskovsky, 2000): specialization, credibility, and coordination. The degree of specialization reflects the extent to which team members carry unique expertise or
knowledge. Credibility refers to the extent to which team members will rely on other member’s expertise, which depends on their beliefs about the reliability of this knowledge. Finally, effective and organized knowledge processing is needed to have team members actually exchange information and knowledge, thereby actively using the transactive memory system (coordination). These manifestations of a transactive memory system are likely to be related to the types of team learning. First, specialization is likely to be related to task and process learning since the different expertise team members will rely on will come to surface when the team is learning about the task. Specialization can, however, also come to surface when team members are exchanging
information about non-work related issues. The degree of specialization is likely to be associated with process learning as well, as it will influence the division of work. Secondly, the extent to which team members feel that the knowledge, information, and expertise of their team members can be trusted (credibility) will likely be associated with the extent to which team members learn about the task and about processes. When trust is low, team members are less likely to consult each other’s expertise, which can prevent task learning to occur. As a result, team processes will be less likely to be reconsidered and adjusted (process learning). Vice versa, when team members know each other better through social learning they are more likely to trust the expertise and
information other team members have. We expect coordination to be related to
all team learning types, as effective processes and smooth interactions will be likely to facilitate sharing and reflection upon different topics.
Team Innovation
Innovation is “the intentional introduction and application within a role, group, or organization of ideas, processes, products or procedures, new to the relevant unit of adoption, designed to significantly benefit the individual, the group, the organization, or wider society” (West & Farr, 1990: 9). When for instance a new technology is discovered and implemented this can be regarded as team innovation (adapted from Amabile, 1988). The concept of team
innovation is closely related to outcome definitions of team learning that focus on changes in knowledge resulting from interaction with each other or the implementation of new technologies, work procedures, or ideas (e.g., Argote et al., 2001; Ellis et al., 2003). Team learning and innovation are closely related to each other and we therefore expect positive relationships between these constructs, especially for task and process learning, since these are more work- related types of learning. However, as argued above, the exchange of
information and ideas about non-work related matters can also promote the exchange of other types of knowledge, ideas, and information, which can also positively influence team innovation. We therefore expect that all types of team learning will be positively related to team innovation.
Method
Sample and Procedure
We used a snowball sampling technique (Goodman, 1961) to collect a sample of 255 individuals who filled in an online survey with the original items.
In the survey, we explained to participants what we defined as a team, which was a group of people usually between 3 to 8 members, who work
interdependently, have common goals, and see themselves and are seen by others as a team (Hackman, 1987). The types of organizational teams
represented ranged from teams in health care / social services (26%), voluntary work teams / committees (32%), marketing and sales teams (16%), sports teams (12%), student work groups (10%) government / research teams (4%). The average team tenure was two-and-a-half years (SD = 3.5 years).
Furthermore, sixty-one percent of the participants were female, with an overall average age of 30, ranging from 18 to 74 years old. Eighty-three percent of the participants were Dutch, one participant was Belgian, and one German (16% missing). The sample represented a broad variety of educational levels, university (27%), vocational / technical college (15 %), and high school (41%;
17% missing).
We used our personal networks to send participants an email providing the link to the online survey. Participants were asked to fill in the survey and to forward the link to others they thought would be willing to participate. In addition, we recruited students in the university building (23% of the sample) and asked them to fill in a paper-and-pencil survey. If the students had work experience or experience with voluntary / committee work, they filled this in. If they had no work experience yet, they filled in the questions about their student work group (10% of the sample). Both groups were equally rewarded with a monetary payment. In the introduction to the survey we explained that the survey was anonymous and confidential and that the goal of the survey was to pretest a measurement scale.
Scale Development Procedure and Final Team Learning Scales
Items were written to cover the three constructs of task, process, and social learning. The items were formulated based on examples from the interviews and the definitions of task, process, and social learning. In line with past team learning definitions (Argote et al., 2001; Hinsz et al., 1997; Tjosvold et al., 2004), we made sure that the following aspects were consistent over the three learning types: sharing and reflecting upon knowledge, information, and/or ideas and improving team performance as a result of these processes.
We had two team learning experts review our definitions and items at different stages of our scale development process for validation. Based on these
consisted of 10 items to measure task learning, 7 items to measure process learning, and 9 items to measure social learning. Participants were asked to respond on a 1-7 point likert scale (1 = strongly disagree to 7 = strongly agree).
All items are listed in Table 1.
Task learning. To measure task learning we developed items aimed at measuring the extent to which team members felt that they shared knowledge, expertise, and opinions about the task and reflected upon insights regarding the task (e.g., “As a team we share our ideas about the task”). Additionally, we asked to what extent this learning about the task improved team performance (e.g., “We improve our performance on the task by sharing task-related knowledge with each other”).
Process learning. To measure process learning we developed items focused on measuring the extent to which team members learn about and improve work procedures and routines with the goal to improve their team performance. Sample items were: “We regularly reflect on our work procedures to see how we can improve them” and “As a team, we create work routines that help us to improve our work.”
Social Learning. To measure social learning we developed items measuring the extent to which team members knew each other well and how this influences team performance (e.g., “When you know each other personally, it makes work easier.”). As sharing non-work related experiences often take place outside work, we also added items asking about social hours and lunches as well (e.g., “During lunch or drinks we get to know each other personally”).
Other Measures
Transactive memory. We measured transactive memory using the scale by Lewis (2003), consisting of three sub-constructs, specialization, credibility, and coordination, which were measured each with 5 items. Sample items of the scale were “I know which team members have expertise in specific areas,” (specialization); “I was confident relying on the information that other team members brought to the discussion” (credibility), and “Our team worked together in a well-coordinated fashion” (coordination). Since we did not aim to investigate project teams in particular but instead a broad range of teams, we slightly changed the wording of some items that referred to knowledge or aspects of the “project” into knowledge or aspects of the “work”. The appendix lists all items. Factor analysis showed that the three subscales loaded on three
different factors. We therefore decided to create 3 different subscales reflecting the subdimensions of transactive memory (for specialization α = .86, credibility α = .69, and coordination α = .69). We removed one (reversed coded) item from the credibility scale (“When other members gave information, I wanted to double-check it for myself”), since reliability increased substantially (scale if item deleted α = .76). We therefore created a subscale of credibility based on the remaining 4 items.
Innovation. We measured innovation using an adaptation of the scale by De Dreu (2006) consisting of 4 items. As the original scale was developed for supervisors ratings of team innovation, we slightly adjusted the wording to enable team members to fill it in about their own team. Sample items were: “As a team, we often implement new ideas to improve the quality of our products and services”, and “My team is an innovative team”, α = .82. The appendix lists all items.
Controls. We controlled for team tenure in years (“How long have you been a member of this team?”), group size and team type (health care / social services, voluntary / committee work, marketing and sales teams, sports teams, student work groups, and government / research teams).
Results
Factor Analysis
We did an exploratory factor analysis, using an oblique factor solution.
As the team learning types are likely to be correlated rather than independent, the use of an oblique factor solution (using Kaiser Oblimin Rotation) is more appropriate than using an orthogonal one. The initial solution showed 5 underlying factors with an eigenvalue above 1. However, using the “scree criterion” (Catell & Vogelmann, 1977), 3 factors could be extracted, explaining 51 percent of the total variance. The three factors reflected the types of team learning: task, process, and social learning (see Table 1 for factors with item loadings).
We used several criteria to assess the reliability and internal consistency of the scales (e.g., Schippers et al., 2007; Den Hartog, Van Muijen & Koopman,
Table 1. Factor Loadings on Items of Task, Process, and Social Learning
Factor 1 Factor 2 Factor 3
Task Learning
TL1 As a team, we learn about the task at hand. .62 .06 -.17
TL2 We improve our performance on the task by sharing task-related knowledge with each
other. .62 .01 -.16
TL3 During task performance, we try to share our expertise as much as possible. .53 .02 -.19 TL4 Every team member makes sure he or she understands the content of the task. .55 .02 -.04 TL5 We improve our performance by reflecting upon the ideas of all team members about the
task. .71 -.03 -.05
TL6 As a team we share our ideas about the task. .74 .08 -.01
TL7 By reflecting upon knowledge about the task we improve our performance on the task. .72 .10 -.07 TL8 We all share our knowledge in order to increase our understanding of the task. .81 -.05 -.04 TL9 In this team all team members offer their specific expertise to improve our performance. .10 -.06 -.33 TL10 As a team we improve our performance by learning about the task. .38 -.08 -.36 Process learning
PL1 We regularly reflect on our work procedures to see how we can improve them. .08 .02 -.74 PL2 We learn about how we can improve our work processes. .13 .07 -.72 PL3 As a team, we create work routines that help us to improve our work. .03 .10 -.68
PL4 We often make agreements about who should do what. .33 .16 -.15
PL5 In our team, we learn about different ways to do our work. .15 -.04 -.65 PL6 We improve our work processes by reflecting upon the way we do our work .18 -.06 -.72
PL7 We often discuss how to do our work. .25 . 15 -.52
Social Learning
SL1 In our team we usually sympathize with someone’s personal situation when this
influences work. .19 .52 -.07
SL2 We learn about each other’s personalities. .11 .66 -.03
SL3 I can better understand the actions of my team members when I know more about their
personal lives. .14 .64 .26
SL4 When you know each other personally, it makes work easier. .19 .52 .24 SL5 As a team, we know each other´s personal life situations pretty well. -.06 .77 -.12 SL6 We often lunch or have drinks together, so that we get to know each other better. -.35 .74 -.32 SL7 During lunch or drinks we get the know each other personally. -.34 .73 -.29 SL8 Knowing each other’s personalities helps us in doing our work. .15 .74 .12 SL9 Because we know each other´s personalities, we learn to interpret each other’s behavior. .01 .79 .00
Eigenvalue 7.52 3.76 1.73
Explained variance 28.9% 14.5% 6.7%
Note. N = 255, PCA with Kaiser Oblimin Rotation; Explained variance (cumulative): 50%. Items shown in italics are discarded.
1997). First, the Cronbach’s alpha should be above .70 (Nunnaly, 1976), secondly factor loadings should be above .40, and the difference between factor loadings should be larger than .20. Third, the item-rest correlations should be above .20 (Kline, 1986). Based on these criteria two items were discarded from the analyses since they did not meet the criteria. One task learning item (item 10) had the highest loading on the process learning scale, but still did not meet the second criterion and was therefore discarded. One process learning item (item 4) had the highest loading on task learning, but was below .40 as well, and was therefore discarded as well. The final scales consisted of 8 items for task learning (α = .87), 6 items for process learning (α = .85) and 9 items for social learning (α = .86) (see Table 1).
Correlations and Discriminant Validity
Table 2 includes the means, standard deviations, and correlations among the variables in this study. We found significant correlations between team tenure and process learning. The longer individuals were part of the team, the more they reported process learning. We also looked at the effects of team type on the types of team learning. We found differences in task learning between government / research teams and the other types of groups.
Individuals in government / research teams reported more task learning (M = 6.37; SD = .25) than individuals in teams in the area of voluntary / committee work, health care, marketing and sales, student work and sports (average M = 5.70; SD = .80). For process and social learning we did not find differences between the different team types. Nor did we find significant correlations between demographic variables such as gender, age, educational level, and work experience and task, process, and social learning. We controlled for the data collection procedure: a T-test revealed that there were no significant differences between the group of participants who filled in the survey online and the group who did the paper and pencil test and their scores on team learning.
To investigate the construct validity of the team learning typology, we examined the correlations between the team learning typology and the three subdimensions of transactive memory (specialization, credibility, and
cooperation) and team innovation (see Table 2). The three team learning types were significantly correlated with each other. The subdimensions of transactive memory were also significantly but differently correlated with the three types of team learning. As we expected, the subdimension of specialization was
significantly positively associated with process learning and social learning, however, not with task learning. The dimension of credibility, in turn, was significantly related to task learning and social learning, but not to process learning. The correlations support the expectation that team coordination was significantly related to all types of team learning. In line with what we expected, all three types of team learning were significantly correlated to team
innovation, with stronger correlations for process and task learning than for social learning.
To assess the discriminant validity of the team learning types we factor analyzed the team learning types using the same factor analysis procedure, but now with the other related constructs included as well. The initial solution showed 10 underlying factors with eigenvalues over 1. However, using the
“scree criterion”, 7 underlying factors could be extracted with an eigenvalue above 1, explaining 58 percent of the total variance. The 7 factors reflected the three types of team learning and distinguished the three different
subdimensions of transactive memory and innovation (see Table 3 for factors
M SD 1 2 3 4 5 6 7 8 9 1. Group Size 7.12 4.41 -
2. Team Tenure 2.51 3.52 .22** - 3. Task Learning 5.57 .81 -.06 .04 - 4. Process Learning 4.90 1.09 .03 .15* .60*** - 5. Social Learning 5.46 .89 .06 .05 .22*** .25*** - 6. Specialization 4.75 .67 .05 .27*** .02 .17** .15* - 7. Credibility 5.71 .78 -.04 -.03 .19** .11 .20** .13 - 8. Coordination 5.12 1.00 -.05 .03 . 36*** .27*** 27*** -.02 .43*** - 9. Team Innovation 4.63 1.19 .01 .07 .39*** .53*** .22*** .19** .14* .28*** - Note. * p <.05, ** p <.01, *** p <.001, N ranges from 211 to 255.
Table 2. Means, Standard Deviations, and Correlations between the Variables
Table 3. Factor Loadings of Team Learning Types with Related Constructs
1 2 3 4 5 6 7
Task Learning
TL1 learn about task .57 .05 -.06 -.01 .04 -.06 -.27
TL2 improve performance by sharing knowledge .56 .01 -.02 -.03 -.06 -.07 -.21
TL3 sharing expertise .51 .03 .02 -.09 .08 -.13 -.26
TL4 makes sure to understand content of task .52 -.01 .03 .14 -.04 -.16 .04 TL5 improve performance by reflect upon ideas .74 -.03 .09 .09 -.13 .07 .06
TL6 sharing ideas .72 .08 -.09 .00 -.03 -.02 -.04
TL7 reflect upon knowledge and improve
performance .68 .11 -.05 .01 -.13 -.02 -.06
TL8 share knowledge to increase understanding
task .76 -.05 -.04 -.00 -.08 -.08 -.04
TL9 members offer expertise to improve
performance .25 -.11 .66 .07 -.15 .07 .05
TL10 improve performance by learning about task .39 -.10 .09 .09 -.01 -.04 -.32
Process Learning
PL1 reflect upon procedures to improve them .08 -.01 .01 -.08 -.12 -.09 -.68 PL2 learn about improving work processes .17 .02 .17 -.08 -.11 -.13 -.58 PL3 create work routines that improve work .08 .04 .20 -.17 -.15 -.19 -.48
PL4 make agreements who does what .36 .14 .20 -.10 .06 -.09 -.13
PL5 learn about different ways to do work .14 -.05 .03 .12 -.29 .17 -.57 PL6 improve processes through reflection .13 -.07 -.07 .08 -.19 .05 -.74 PL7 discuss how to do the work .23 .16 -.12 .07 -.01 .08 -.63 Social Learning
SL1 sympathize with personal situation .15 .48 .03 .18 .05 .02 -.18
SL2 learn about personalities .09 .62 .09 -.08 .02 -.16 -.03
SL3 increased understanding knowing personal
lives .08 .67 -.19 .24 .10 .15 .04
SL4 knowing each other makes work easier .16 .54 -.20 -.07 -.04 -.03 .17 SL5 knowing each other's personal life situation -.05 .71 .21 .02 -.02 -.14 -.02 SL6 have lunch or drinks together -.34 .69 .03 .05 -.15 -.09 -.19 SL7 get to know each other during lunch / drinks -.29 .70 .08 -.05 -.05 -.07 -.19 SL8 knowing personalities helps doing work .19 .74 .13 .00 -.06 .03 .20 SL9 learn interpret behavior by knowing
personality .02 .79 .04 -.04 -.07 -.01 .02
Team Innovation
IN1 implement new ideas -.04 .02 -.01 -.03 -.82 -.07 -.12
IN2 consideration new methods /procedures (R) -.03 -.03 -.06 .10 -.67 -.04 -.16 IN3 produce new services, methods, procedures .04 .04 .09 -.16 -.76 -.01 -.02
IN4 my team is innovative .07 .10 .01 .12 -.82 -.00 .10
with item loadings). Based on the criteria mentioned earlier, the same two task learning items were discarded from the analyses, since they did not meet the criteria. Task learning item 10 double loaded with items from the process learning scale and had loadings under .40 and was therefore discarded. Again, process learning item 4 loaded on task learning, but was below .40, and was therefore discarded as well. Task learning item 9 loaded on the specialization items, with a loading above .40. Inspection of the content of this item revealed that this item actually measured the degree of specialization and was therefore discarded for theoretical reasons. The final scales for the team learning types were consistent with Table 1.
Table 3. (continued)
1 2 3 4 5 6 7 Transactive Memory: Specialization
SP1 team member’s specialized knowledge .04 .01 .79 .09 -.08 .12 .02 SP2 knowledge is specialized -.08 -.07 .67 -.10 -.14 .02 .14 SP3 responsibility different expertise -.03 .08 .88 -.00 .09 .01 .03 SP4 specialized knowledge needed to perform -.02 .04 .86 .01 .11 .02 -.11 SP5 knowing each other's expertise -.08 .17 .75 .16 .07 -.06 -.11 Transactive Memory: Credibility
CB1 acceptance procedural suggestions others .01 -.17 .12 .54 -.04 -.19 .08 CB2 trust other members’ knowledge .18 .07 .05 .82 -.03 .01 .11 CB3 rely on other's information -.01 .08 .03 .82 -.01 -.03 -.05 CB5 faith in other's expertise (R) -.15 .08 -.02 .71 -.01 -.18 -.14 Transactive Memory: Coordination
CO1 well coordinated work style .11 .06 -.06 -.02 -.13 -.71 -.06
CO2 few misunderstandings .07 .01 -.04 .10 -.07 -.79 .23
CO3 need to backtrack and start over (R) .01 -.05 -.10 .17 .09 -.72 -.06 CO4 smooth and efficient task accomplishment .17 .04 -.06 -.03 -.08 -.72 .03 CO5 confusion about task accomplishment (R) .11 -.05 -.04 -.9 -.02 .71 .04
Eigenvalue 9.70 4.09 3.90 2.90 2.06 1.60 1.46
Explained variance 22.0% 9.3% 8.9% 6.6% 4.7% 3.6% 3.3%
Notes. N = 255, PCA with Kaiser Oblimin Rotation; Explained variance (cumulative): 58.4%. Items shown in italics are discarded. Factor Loadings above .40 are bold. Short versions of items are displayed. Full items can be found in Table 1 and the appendix. (R) = reversed coded.
Discussion
Past definitions and measurements of team learning have not
explicitly distinguished different types of team learning based on the topic that team learning can be about. As a response to this, Jehn and Rupert (2007) have theoretically proposed a typology of team learning specifying three types of team learning: task, process, and social learning. They proposed that teams can learn about the task they perform (task learning), about work routines and processes (process learning), and about team members´ personalities and other non-work related topics (social learning), which may differently affect group outcomes. In this study we developed an instrument to measure the three types of team learning, which may help initiate future research on this topic. Based on interviews with some managers in organizations and past team learning definitions, we developed items to create a survey useful for different sorts of teams with different tasks. We conducted a pilot test to validate the items on a sample of individuals reporting about a wide variety of teams. A second aim of this study was to explore the relationships between the team learning types and other related constructs, such as (dimensions of) transactive memory and team innovation.
Factor analysis showed that three underlying factors could be distinguished, representing three different types of team learning. The task learning items measured the extent to which team members shared and reflected upon knowledge, expertise, and opinions with each other regarding the task in order to improve team performance. Items of process learning were aimed at measuring the extent to which team members learnt about and improved their work procedures and routines with the goal to improve their team performance. To measure social learning we developed items measuring the extent to which team members knew each other’s personal life situations and personalities well and how this affected team performance. The three team learning types were internally consistent and reliable.
Although the team learning types were significantly correlated, with relatively high correlations between task and process learning in particular, they showed high internal consistency and were differently related to other related constructs. Our exploratory factor analysis conducted to assess discriminant validity showed that the team learning types loaded on different factors than the subdimensions of transactive memory and team innovation. Furthermore, the correlations showed that the team learning types were positively related to
subdimensions of transactive memory (specialization, credibility, and coordination) and innovation.
The extent to which team members had different specializations was associated with process and social learning, but not with task learning, as we expected. The correlation table also showed that specialization occurred more in teams with higher team tenure. It might be the case that when team members work together for a longer period, they know each other’s expertise well and have built an extensive transactive memory system, so that they need less reflection upon the content of the task. They might therefore spend more time dividing up the work according to each other’s expertise and getting to know each other in other non-work related areas. Future research should therefore examine team dynamics in multiple stages of development to see how team learning and transactive memory can change over time. The second factor of transactive memory was credibility, which is the extent to which team members trusted each other’s expertise. This factor was associated with task and social learning. When team members trust each other’s expertise they are more likely to learn as a team about the content of the task, and when they know each other well personally, they might be more likely to trust each other’s expertise as well (or vice versa). Future research is needed to determine the causality. Finally, the extent to which teams worked together in a well- coordinated fashion (coordination) was positively associated with all types of team learning, which supports our expectations.
In line with what we expected, all types of team learning were related to team innovation, although task and process learning were more strongly associated with innovation than social learning. This could be an interesting direction for future research. It might be the case that, since task and process learning are more work-related types of learning, they are more strongly associated with work-related outcomes, such as performance and innovation.
Social learning might be more strongly associated with affect-based team outcomes, such as satisfaction and cohesion. It might also facilitate the other, more work-related types of team learning, such as task and process learning and/or other aspects of the team process, such as specialization. Thus, the team learning types may sometimes act as predictors of other team outcomes or some of them might act as moderators, facilitating other processes and/or outcomes. Future research should therefore examine the dynamics underlying these types of team learning. The contribution of this study was to measure the different types of team learning and to show that they are distinct from other related constructs.
Limitations and Future Research Directions
An important limitation of this study is that we did not conduct the pilot study on full teams, in which all team members report about the level of team learning they perceive in their team. Therefore, future research should replicate our findings, preferably using multi-method field research on full teams in organizations and in different settings. Additional interviews and observations should be conducted to further examine and validate the typology. This examination might reveal additional types or subtypes of team learning as well.
Future research should also investigate the antecedents and
consequences of different types of team learning. How the team is composed in terms of social category and informational diversity (e.g. Jehn, Northcraft &
Neale, 1999; Milliken & Martins, 1996; Webber & Donahue, 2001), can be an important predictor of task, process, and social learning. Future research should therefore examine how different types of diversity and diversity faultlines (Lau &
Murnighan, 1998) influence these different types of learning (Rupert & Jehn, 2008). Furthermore, it would be interesting to investigate how these types of learning influence different types of team outcomes, such as performance, innovation, satisfaction, commitment, and social cohesion, and to examine what factors could moderate or mediate these relationships. It might be that some types of learning distract more from the task than others, depending on the team’s learning phase and learning needs. Therefore, future research should also measure other aspects of team learning, such as team learning distraction, phase, needs, and the relative importance of the different types.
Conclusion
In this paper we developed an instrument to measure three types of team learning depending on the topics that teams can learn about: task, process, and social learning. Based on interviews with managers from a health care organization and a multinational bank, we developed items which we validated using a pilot sample of 260 individuals working in teams. Results showed that three factors could be distinguished and that the factors were different from other related constructs. This instrument should aid both researchers and practitioners to measure these types of team learning in
organizations and examine the dynamics underlying the typology and how they relate to team characteristics, processes, and other team outcomes.
Appendix:
Survey Scales of Related Concepts
Transactive Memory
Specialization
1. Each team member has specialized knowledge of some aspect of our work.
2. I have knowledge about an aspect of the work that no other team member has.
3. Different team members are responsible for expertise in different areas.
4. The specialized knowledge of several different team members was needed to complete the work deliverables.
5. I know which team members have expertise in specific areas.
Credibility
1. I was comfortable accepting procedural suggestions from other team members.
2. I trusted that other members’ knowledge about the work was credible.
3. I was confident relying on the information that other team members brought to the discussion.
4. When other members gave information, I wanted to double-check it for myself.
(reversed)
5. I did not have much faith in other members’ “expertise.” (reversed)
Coordination
1. Our team worked together in a well-coordinated fashion.
2. Our team had very few misunderstandings about what to do.
3. Our team needed to backtrack and start over a lot. (reversed) 4. We accomplished the task smoothly and efficiently.
5. There was much confusion about how we would accomplish the task.
(reversed)
Note. To improve reliability item 4 of the credibility scale was discarded.
Team Innovation
1. As a team we often implement new ideas to improve the quality of our products and services.
2. My team gives little consideration to new and alternative methods and procedures for doing their work. (reversed)
3. As a team we often produce new services, methods, or procedures.
4. My team is an innovative team.
