1
AN INPUT-PROCESS-OUTPUT MODEL AND
MEASUREMENT INSTRUMENT OF TEAM
EFFECTIVENESS TESTED IN A FIELD
SETTING
Master thesis, MscBA, specialization HRM/OB
University of Groningen, Faculty of Management and Organization
July 4, 2012
FRANKA REUVERS
Studentnumber: 1776142
Jozef Israëlsstraat 101a
9718 GH Groningen
tel.: + 31 (0)6-23617897
e-mail:
f.g.h.reuvers@student.rug.nl
Supervisor/university
prof. Dr. G.S. van der Vegt
Acknowledgment: Helpful comments on earlier drafts of this thesis were given
by my supervisor prof. dr. G.S. van der Vegt. I thank prof. dr. G.S. van der Vegt
for supporting me in writing this thesis, the cooperating organization for
2
AN INPUT-PROCESS-OUTPUT MODEL
AND MEASUREMENT INSTRUMENT OF
TEAM EFFECTIVENESS TESTED IN A
FIELD SETTING
Abstract
The importance of teams for organizations has resulted in a great deal of research examining the characteristics of effective teams. Numerous input and process characteristics have been found to be related to team effectiveness. Building on this literature, I examined the relationship between various input and process characteristics and team effectiveness in a large healthcare organization. Team composition, task design and context formed the input characteristics and transition, action and interpersonal processes formed the process characteristics. In total 25 characteristics were assessed. Team effectiveness characteristics used were team satisfaction, team performance judged by team managers, and financial performance of teams. Data were collected from 768 team members, 44 team managers and from the management information system. Consistent with existing input-process-output models of team effectiveness, results showed that the input characteristics were most strongly related to the process characteristics, whereas the process characteristics were related to team satisfaction and team performance. Moreover, this study provided and validated a measurement instrument for team effectiveness. This instrument can be used by organizations to diagnose problems and identify areas for improvement in teams. Finally, implications and suggestions for designing more effective teams were discussed. So, findings of this study contribute to the understanding of team effectiveness and its antecedents in a field setting.
Keywords: Teams, input, process, output characteristics, team effectiveness, measurement
3
INTRODUCTION
The popularity of work groups and teams is still increasing, both in research and in practice. Many academics and practitioners still put it on best practice lists (Pfeffer, 1998). Indeed, teams are widely used in organizations. They are assumed to produce a variety of advantages for organizations and its employees. For instance, teams provide a way of pooling ideas and for that reason yield a competitive advantage. Moreover, teams foster direct participation and empowerment, which generates better decision-making and solutions that are more creative (Kang, Yang & Rowley, 2006). As a result, teams are an integral part of business success (Cohen & Bailey, 1997).
The importance of teams for organizations has resulted in a great deal of research about the characteristics of effective teams (Campion, Medsker & Higgs, 1993; Hyatt & Ruddy, 1997; Marks, Mathieu & Zaccaro, 2001; Mathieu, Maynard, Rapp & Gilson, 2008). Ilgen, Hollenbeck, Johnson and Jundt (2005) found that research on teams showed a movement over the years. Prior to 1996, most empirical research concentrated on team performance and viability. This type of research focused on team input characteristics, such as team composition, structure and resource allocation (Ilgen et al., 2005). After 1996, more attention was paid to mediating processes that explain why certain input characteristics affect team effectiveness outcomes. Kozlowski and Ilgen (2006) expressed this in their so-called input-process-output model (i.e. I-P-O model). Input refers to various resources (e.g. composition and team design) available to the team. Processes represent mechanisms (e.g. communication and cooperation) that inhibit or enable the ability of team members to combine their capabilities and behavior. Output refers to criteria that assess the effectiveness (e.g. performance and satisfaction) of teams’ actions (Kozlowski & Bell, 2001). The I-P-O model has exerted a powerful influence on team research, because most studies explicitly or implicitly invoke the I-P-O model (Ilgen et al., 2005). Moreover, the I-P-O framework provides structure to many models of team effectiveness, and creates a substantial degree of similarity across models (Kozlowski & Bell, 2001).
4 this research will contribute by developing and validating a measurement instrument to assess the elements of the I-P-O model in order to better predict team effectiveness in field settings. Such a measurement instrument can be used by organizations to diagnose problems and identify areas for improvement. Taken together, the following research question will be answered:
5
THEORETICAL BACKGROUND
According to Kozlowski and Ilgen (2006) and Cohen and Bailey (1997) a team (1) consists of two or more individuals, (2) who socially interact, (3) possess one or more common goals, (4) are brought together to perform organizationally relevant tasks, (5) exhibit interdependencies with respect to workflow, goals and outcomes, (6) have different roles and responsibilities, (7) and are together embedded in an encompassing organizational system, with boundaries and linkages to the broader system and task environment. This definition of a team is provided in order to elaborate further on the theoretical background of the I-P-O model.
Team effectiveness framework
This study will focus on the I-P-O model because research has shown that the I-P-O model has had a great influence on recent team research (Mathieu et al., 2008). Almost fifty years ago, McGrath (1964) was one of the first researchers that came up with an I-P-O model for examining team effectiveness. Since this model was presented, several other modified and extended I-P-O models of team effectiveness followed (Cohen & Bailey, 1997; Gladstein, 1984; Hackman & Morris, 1975; Kozlowski & Ilgen, 2006; Mathieu et al., 2008). According to Mathieu et al. (2008) most of the adaptations to the I-P-O model have either placed it in a larger context, emphasized a temporal element, or rediscovered more subtle aspects of the model that have gone overlooked. However, all these models recognized the difference between input and process factors that influence team effectiveness. There were also many other important studies that only focused on parts of the I-P-O model, either the input or the process characteristics for predicting team effectiveness (Campion et al., 1993; Cohen, Ledford & Spreitzer, 1996; Marks et al., 2001; LePine, Piccolo, Jackson, Mathieu & Saul, 2008). So, the I-P-O model is definitely not new in researching team effectiveness, but there is no study that examined the I-P-O model altogether in a field setting.
6
FIGURE 1
The input-process-output model of team effectiveness
INPUT Team composition - Team size - Functional diversity - Team fluidity Task Design - Team autonomy - Task interdependence - Interteam interdependence - Clear goals - Challenging goals Context - Team training - Organizational resources - Organizational appreciation - Transformational leadership - Transactional leadership PROCES Transition processes - Team reflexivity - Temporal planning Action processes - Team effort - Team monitoring - Backup behavior
- Team coordination (problems) - Team communication
7
Output
Output refers to team effectiveness characteristics and results from team processes and team input characteristics. Team effectiveness comprises team satisfaction, team
performance, and financial performance. Many studies use attitudes and performance as the
two major indicators when assessing team effectiveness, with attitudes mostly measured in terms of the degree of work satisfaction (Wang, Lin & Hsu, 2010). Work satisfaction refers to one's response, either to one's job or to certain aspects of one's job (Mowday, Porter & Steers, 1982). Since this study does research at the team level, team satisfaction refers to the average work satisfaction in a team, which is based on work satisfaction of individual team members. Performance is generally assessed as subjective performance or in terms of objective, quantifiable standards. Concerning subjective performance, Mathieu et al. (2008) discusses the distinction between performance behaviors and performance outcomes. In this study, team
performance refers to both performance behaviors and outcomes, where behaviors are the
actions that are relevant for achieving goals, while the outcomes are the consequences or results of performance behaviors, such as quality of work (Mathieu et al., 2008). Finally, objective performance in terms of quantifiable standards refers in this study to the financial performance of teams.
Processes
Since team effectiveness results from team processes, processes are described next. Processes are the activities that explain why input factors affect team effectiveness outcomes (Hackman & Morris, 1975). These are the activities that team members engage in, which are mostly dynamic due to team members’ interactions and development (Kozlowski & Ilgen 2006). During the past decade, processes received increasing attention in research (Mathieu et al., 2008). Therefore, this section will consider the most important processes identified in the literature and their relationship with team effectiveness. Following Marks et al. (2001), three types of processes can be distinguished based on the period of time in which work is performed and evaluated. It covers transition, action and interpersonal processes, which are all processes that are validated by LePine et al. (2008). These three processes including their characteristics are successively discussed.
8 al., 2008). Nonetheless, two important transition processes are included in this study, namely team reflexivity and temporal planning.
The first transition process characteristic is team reflexivity, also called mission analysis by Marks et al. (2001). It is the tendency to overtly reflect upon the group’s objectives, strategies, and processes and adapt them to current or anticipated circumstances (De Dreu, 2002). A study of Carter and West (1998) showed that team reflexivity was positively related to participation in decision making, clarity of team objectives, and affective well-being (e.g. satisfaction). Moreover, task reflexivity predicted team effectiveness (Carter & West, 1998).
Another transition process is temporal planning, also called strategy formulation and planning by Marks et al. (2001), and is about developing courses of action and plans, or adjusting to changes (Marks et al., 2001). When a team plans, it actively scans and assesses task requirements, including resources needed and possible restraints (Janicik & Bartel, 2003). Moreover, planning facilitates the formation of group norms that regulate members’ responses to temporal issues, such as deadlines. Therefore, temporal planning at the beginning of a task results in an effective coordination and successful performance in teams (Janicik & Bartel, 2003). For team satisfaction, Gevers and Peeters (2009) found that temporal consensus, the extent to which the team agreed upon their planning, predicted individual satisfaction. However, the relationship between temporal planning itself and team satisfaction was not examined before.
Action processes. Action processes are the activities that take place when the team works towards the accomplishment of its goals (Marks et al., 2001). In contrast to transition processes, action processes received considerable attention in research. Therefore, there are many characteristics covered, namely team effort, team monitoring, backup behavior, team coordination, team communication and interteam coordination and cooperation.
Team effort is defined as the extent to which team members devote their resources
(i.e., energy, attention, time) to executing team tasks (Yeo & Neal, 2004). Research of de Jong and Elfring (2010) suggested that team effort positively influences team performance, because if team members see other team members work hard, they tend to set higher goals for themselves. Effort is closely linked to motivation processes (Yeo & Neal, 2004), and therefore expected to also relate with satisfaction.
9 to Marks and Panzer (2004) team monitoring benefits overall team effectiveness, so both team performance and satisfaction, because it enhances team coordination and the provision of feedback. Furthermore, team monitoring is related to backup behavior (Marks et al., 2001).
Backup behavior means that team members assist each other in the performance of
their tasks. This factor shows similarities with workload sharing found by Campion et al. (1993) and Ilgen et al. (2005), and is considered as an important characteristic in predicting team effectiveness. Results show that backup behavior related positively with team satisfaction and performance (Campion et al., 1993; Ilgen et al., 2005).
Another action process is team coordination. It refers to aligning activities of the team members with respect to sequencing and timing (Marks et al. 2001). It was shown that high levels of coordination result in contribution of all team members to performance. In accordance, it was demonstrated that coordination has a positive impact on all team outcomes (Steiner, 1972; Wilke & Meertens, 1994).
Team communication refers to the information exchange within teams and is
considered as an important process characteristic for team effectiveness (Barrick, Stewart, Neubert & Mount, 1998; Kozlowski & Ilgen, 2006;). Barrick et al. (1998) showed in their study that communication was positively related with team performance. Communication supports task work by exchanging information and developing solutions for problems. It also supports teamwork by establishing a higher quality of interaction (Kozwlowski & Ilgen, 2006). More effective task work due to communication is expected to relate with team performance, whereas a higher quality of interaction thanks to communication is expected to relate with satisfaction.
A final characteristic is interteam coordination and cooperation, and it represents the alignment of activities and communication between groups (Marks et al., 2001). As teams oftentimes depend on other teams, the work of one team has implications for the work in other teams. Therefore, coordination and cooperation between teams is needed (Hoegl, Weinkauf & Gemuenden, 2004). Besides, coordination and cooperation between teams positively influences the performance of teams because it offers an opportunity to exchange experts and perspectives between teams (Hoegl et al., 2004). Interteam coordination is also a critical component in delivering customer satisfaction (Ambrosini, Bowman & Burton-Taylor, 2007), but effects on individual job satisfaction are not found yet.
10 interpersonal relationships, which is an on-going activity at all times (Marks et al., 2001). The interpersonal processes included in this study are positive and negative affect, collective efficacy and psychological safety.
Two interpersonal characteristics of effective teams are whether they have positive and/or negative affect. Positive affect is defined as the extent to which a person feels enthusiastic, active and alert. On the other hand, negative affect reflects subjective distress and unpleasant engagement that subsumes a variety of aversive mood states, including anger, disgust, guilt and fear (Watson & Clark, 1984). Affect has been found to influence team effectiveness (Kozlowski & Ilgen, 2006). For example, George (1990) found that positive affect related negatively with absences in teams and negative affect related negatively to the level of prosocial behavior in teams. Moreover, positive affect has been found to be positively related to emotional well-being and higher satisfaction within a team (Barrick et al., 1998).
Another interpersonal process characteristic is collective efficacy, which represents the belief in one’s team and in the group’s collective capability to organize and execute courses of action required to produce given levels of goal attainment (Kozlowski & Ilgen, 2006). Across multiple studies, collective efficacy showed a positive relation with both team performance and team satisfaction (Ilgen et al., 2005; Kozlowski & Bell, 2001; Lestler, Meglino & Korsgaard, 2002). It is related to these team effectiveness criteria, since collective efficacy affects what a group chooses to do (i.e. goal setting), but also how much effort it will exert, and persistence in the face of failure (Bandura, 1997).
Finally, psychological safety has been found to affect team effectiveness. It is defined as a shared belief held by team members that the team is safe for interpersonal risk taking (Edmondson, 1999). It describes a team climate characterized by interpersonal trust and mutual respect in which people are comfortable being themselves. Edmondson (1999) revealed that psychological safety has a positive influence on team performance through learning behavior, such as feedback seeking, sharing information, asking for help, and discussing errors.
Input
11 study cover team composition, task design and context characteristics. In the next section these characteristics are elaborated.
Team composition. Team composition is incorporated in nearly all models of team effectiveness (Mathieu et al., 2008). It refers to the nature and attributes of group members (Guzzo & Dickson, 1996), and contains in this research team size, functional diversity, and team fluidity. Since team composition is expected to influence team processes (Kozlowski & Bell, 2001), and in order to select and construct more effective teams, organizations need to have a better understanding of the effects of team composition characteristics. So, these different characteristics are discussed below.
One characteristic of team composition is team size. It refers to the number of people in a team. In general, research suggests that teams should be as small as possible given the work to be accomplished (Wageman et al., 2005). Moreover, LePine et al. (2008) recently demonstrated that larger teams have more linkages among members than do smaller ones, and therefore face greater coordination challenges. Accordingly, social loafing theory also suggest that if team size increases, individual effort decreases (Buelens, Sinding & Waldstrom, 2011). Likewise, larger teams will have more motivation and coordination losses (LePine et al., 2008). So, team size is expected to have a negative relationship with team processes.
Secondly, functional diversity is considered to be an important team composition characteristic. It is the extent to which teams vary in the number of team members with a different functional background (Ilgen et al., 2005). Teams require a higher diversity in competences, because teams need different skills and abilities than individuals (Mathieu et al., 2008). Likewise, diversity in competences has been found to have a positive effect on performance, especially when teams need to perform diverse tasks (Guzzo & Shea, 1992).
12 Task design. Task design refers to the characteristics of the work performed by team members. A task is about what teams have to do, but also functions as a source of goals, roles and task-based exchanges (Kozlowski & Ilgen, 2006). In accordance, a task must be well-aligned with the team’s purpose (Hackman, Wageman, Ruddy & Ray, 2000). Therefore, in this study task design captures team autonomy, intra-team task interdependence, interteam task interdependence, and goals. These characteristics will be explained in the following paragraphs.
The first characteristic, team autonomy is interchangeably used with self-management. According to Molleman (2000) team autonomy refers to “freedom in decision making within a unit with respect to both the transactions (output) it wants to realize and the way it organizes its transformation processes to achieve these transactions”. Team autonomy has been used a lot in work organizations (Cohen et al., 1996; Yan & Guy, 2010) and is therefore considered as an important variable. Moreover, team autonomy was found to be positively related to team performance (Cohen & Bailey, 1997).
Another characteristic of importance is task interdependence. Task interdependence is defined as “the extent to which an individual team member depends on other members of the team for being able to carry out his or her job” (Van der Vegt & Janssen, 2003). It results from the work that organizations divide among work teams, that is further divided among individuals occupying different jobs or roles within their team. It enhances the sense of responsibility for other’s work (Kiggundu, 1983) and the reward value of group accomplishments (Shea & Guzzo, 1987). Therefore, task interdependence has shown a positive relationship with team effectiveness. Additionally, research reveals that if task interdependence increases, the requirements for coordination, communication and cooperation also increase in order to perform well (Galbraith, 1978). Since interdependence derives from the work divided by the organization, task interdependence is not only present within teams, but also between teams.
13 Meuse & Futrell, 1990). However, research has not extensively tested this characteristic yet, so there is little evidence about the relationship between interteam interdependence and team effectiveness (Campion et al., 1993).
The final characteristics of task design are about the teams’ goals. The importance of goals was already recognized in previous studies that included providing a compelling direction, which refers to whether the teams’ direction is clear, challenging, and consequential (Wageman, Hackman & Lehman, 2005). Therefore, two characteristics are distinguished,
clear goals and challenging goals. First, clear goals in a team context refer to a shared
understanding of the goals and objectives that are set and followed by a team (De Dreu, De Vries, Fransson & Altink, 2000). Clear goals imply that a team can focus their efforts, and can easier complete their tasks (Hu & Liden, 2011), which results in a higher performance. Secondly, challenging goals are neither too demanding nor too easy, and therefore are a challenge (Hackman et al., 2000). They are related to team effectiveness since these goals motivate teams to succeed in their tasks (Hackman et al., 2000).
Context. Since teams are nested within organizations and larger environments, context is considered an important input characteristic. Context characteristics are sources of influence from outside or inside the organization that influence team functioning (Mathieu et al., 2008). Specifically, it refers to the teams’ environment and organizational level resources, like influences from the team manager, organization and environmental surroundings. These contexts where teams operate in hinder or facilitate their functioning (Mathieu et al., 2008), and in that way organizational context is important in predicting team effectiveness (Cohen & Bailey, 1997). In this study, context characteristics covered are team training, organizational support, and managerial support.
The first resource teams need is team training, which is a systematic strategy that targets for improvement of specific objectives or competencies designed to enhance teamwork and influences effectiveness (Kozlowski & Ilgen, 2006). There has been a lot of research about training, but there are still key questions about what, how and when to train (Mathieu et al., 2008). Nevertheless, a recent team training meta-analysis conducted by Salas, Nichols, and Driskell (2007) confirmed a small to moderate relation for the overall influence of team training on improvements in both objective and supervisor ratings of team performance.
14 1993; Mathieu et al., 2008). Likewise, an organizational climate of openness in terms of a well-developed social structure and sociopolitical support has shown a positive relationship with team empowerment and team outcomes (Kirkman & Rosen, 1999). In this study, organizational support consists of organizational resources and organizational appreciation.
Organizational resources include, inter alia, information provided by the organization. So, the
organization must provide the team with whatever data the team needs for carrying out the work (Wageman et al., 2005). In addition, organizational appreciation is an important context characteristic for teams. It refers to the reward system of the organization, and according to Wageman et al. (2005) the organization should provide positive consequences for excellent team performance.
A final context characteristic is managerial support. This refers to leadership or coaching (Wageman et al., 2005) in terms of helping and supporting the team in order to accomplish team tasks and goals. Team coaching involves direct interaction with a team intended to help members make appropriate use of their collective resources in accomplishing the team’s work (Kozlowski & Ilgen, 2006). Moreover, managerial support is needed in order to function effectively as a team, because managers control resources, for instance information and rewards (Guzzo & Shea, 1992). A distinction is made between
transformational leadership and transactional leadership styles. Transformational leadership
15
METHOD
Data collection and procedure
This study examined team effectiveness and its antecedents in a field setting. A survey study was conducted and participants were recruited at an Addiction Healthcare Institute in the Northern part of the Netherlands. This organization employs around 1000 employees. The study was conducted in almost every team of the organization, including health care teams and staff function teams. Division teams including team managers, as well as the management team were participating as teams in this study. A total of 85 teams were contacted to participate in this research project, ranging in size from 3 to 31 employees. The activities of the teams were very diverse and varied from providing drugs and dealing with addicts, to making schedules and recruiting new employees. The researcher, supported by the communication department of the organization, contacted all team managers and their team members. An announcement and introduction letter on intranet followed by a presentation for each team during a team meeting or a specific arranged information meeting informed the teams about the study. It was explained that data would be collected from team members and their managers and that for data matching procedures the researchers had insight in names and personal data. A privacy protocol ensured complete confidentiality. Participation was voluntary, but every team member was encouraged by the researcher, the organization, and the team manager to participate in this research project. The online questionnaires were distributed among all team members, whether they agreed to participate or not.
16 questions about their team, participants also answered questions about themselves, their team manager and the organization.
In order to make reliable statements, a sufficiently large number of employees from each team must be included. In this study, the minimum number of employees for each team had to be six team members, meaning three team members who answered the input questions and three team members who answered the process questions.
The data were collected in April and May 2012. In total 1003 questionnaires were distributed across 85 teams and a total of 724 fully completed questionnaires and 44 not-fully completed questionnaires across 85 teams were returned, resulting in a response rate of 72% respectively 77% at the individual level and 100% at the team level. The average age of all participants was 44 years (SD=10.81), ranging from 23 to 65 years. About a third (35%) of all respondents were male and nearly all respondents were of Dutch nationality (99%). Average tenure was 6.58 years (SD = 5.67), more than half (54%) had an additional higher education (HBO), and even 11% had a university degree (WO).
Team managers received an additional questionnaire including measures about their teams’ performance. For these questionnaires, a total of 83 questionnaires were distributed among 44 team managers. Twenty-seven of these team managers supervised multiple teams, ranging from one to five teams per team manager. Therefore, they filled in multiple questionnaires, one for each team. A total of 82 fully completed questionnaires were returned, resulting in a response rate of 99.8%. These questionnaires were distributed about a week and a half later than the questionnaires for employees. For two teams there was not one single supervisor due to reorganization. Therefore, these teams did not participate in the study. So, in total we collected outcome measures for 82 teams.
Finally, to obtain data about demographics of all respondents, and other objective data, the management information system of the organization was used.
Measures
17
Output
Team satisfaction. Team satisfaction was measured with an individual scale of job satisfaction. The individual scale was aggregated to the team level in order to make statements about the effectiveness of teams. Individual job satisfaction was measured with a scale developed by Van der Vegt, Emans and van der Vliert (2001). An example item was ‘I enjoy my work’.
Team performance. Team performance was measured by asking team managers to compare their teams’ performance to other teams when assessing their team(s) on seven different items. Items used were for instance about achieving goals, quality of the work, productivity and effectiveness of the team. The scale used was developed by Van der Vegt and Bunderson (2005). The items were assessed on a 7-points scale ranging from far below average to way above average. Measures were obtained from 81 out 82 teams due to one missing case.
18
Transition processes
Team reflexivity. Team reflexivity was measured with a scale of de Jong and Elfring (2010). An example item was ‘In my team we often discuss whether we are working effectively as a team’.
Temporal planning. Temporal planning was measured with a scale developed by Janicek and Bartel (2003). An item used was for instance ‘This team regularly speaks about deadlines.’
Action processes
Team effort. Effort was assessed with a scale of de Jong and Elfring (2010). An item used was for example ‘Most of my team members put a lot of effort in achieving a good performance’.
Team monitoring. Team monitoring was assessed with a scale of de Jong and Elfring (2010). An item used was ‘In this team we check whether everybody complies with the obligations’.
Backup behavior. The items used for backup behavior were from Campion et al. (1993). An item used was for example ‘Team members help each other, even if this does not directly belong to their tasks’.
Team coordination. Team coordination was assessed with a coordination problems scale of Janicek and Bartel (2003) and combined with a scale of Lewis (2003). An item used was ‘My team cooperates in a well-coordinated way’.
Team communication. Team communication was measured with an information exchange within teams scale of de Dreu (2003). An item used was ‘In my team there are communication problems’.
Interteam coordination and cooperation. Interteam coordination and cooperation was measured with a scale of Hoegl and Gemuenden (2001) and Faraj and Yan (2009). An example item was ‘We often have problems in cooperating with other teams’.
Interpersonal processes
19 Items used were for instance ‘In general, in my work I feel inspired’, or ‘In general, in my work I feel angry’.
Collective efficacy. Collective efficacy was measured with a scale of Riggs, Warka, Babasa and Hooker (1994). An item used was for instance ‘In my team there is a lot of team spirit’.
Psychological safety. Psychological safety was measured with a scale developed by Edmondson (1999). An item used was ‘Team members sometimes exclude other team members for being different’.
Team composition
Size. Team size was measured objectively by counting the number of people that were in the team. This information was provided by the management information system and checked by team managers.
Functional diversity. Information on function of each participant was provided by the management information system of the organization. Different functions were coded and duplicate functions were removed. This resulted in a total of 69 different job functions. These functions were all highly relevant in order to show functional diversity in the teams. Therefore, functional diversity was computed on these different job functions using Blau’s (1977) index, with the formula 1 – ∑p²i, where p is the proportion of a group in the ith category. The index
score, ranging from 0 to 1, indicated functional diversity in a team, where 0 is no diversity and 1 is very high functional diversity among team members.
Team fluidity. Team fluidity was measured with a scale of de Jong and Elfring (2010). One item used was ‘People from my team continuously leave to other teams or organizations’.
Task design
Team autonomy. Team autonomy was assessed with a scale developed by Molleman, Nauta and Jehn (2004). An item used was for instance ‘We can autonomously decide what the final outcome of an assignment will be’.
20 interdependence scale of Campion et al. (1993). An example item was ‘The members of my team are highly interdependent in performing their tasks’.
Interteam interdependence. Interteam interdependence was measured with an unpublished and self-created scale based on the intra-team interdependence scale of Van der Vegt et al. (2003). An item used was for example ‘Performing our team task requires much communication and coordination with other teams’.
Goals. Goals were measured with two variables, clear goals and challenging goals. Both variables were measured with a scale developed by Wageman et al. (2005). An item used for clear goals was ‘The goals of my team are clear’, and an item used for challenging goals was ‘The goals of my team are challenging’.
Context
Team training. Training and education was assessed with a scale of Campion et al. (1993), and added with a self-developed item: ‘Team members can make suggestions about training and education themselves’.
Organizational support. Organizational support included two variables. The first variable measured was organizational resources and was assessed with a scale developed by Wageman et al. (2005). An item used was ‘My team has sufficient resources to perform tasks properly’. The second variable measured was organizational appreciation and was also assessed with a scale of Wageman et al. (2005). An item used was ‘Teams that perform well are rewarded by the organization’.
Managerial support. Managerial support, refering to leadership, was assessed with a combined transformational and transactional leadership scale developed by Podsakoff, MacKenzie, Moorman and Fetter (1990). An item used for transformational leadership was for example ‘Our team manager has a clear vision’. For transactional leadership for instance ‘Our team manager tells us what to do to be rewarded for our efforts’ was used.
Control variables
21 the team effectiveness characteristics. Measures were obtained from almost all respondents. For all continuous variables, like age, tenure and gross salary, the mean and standard deviation from these teams was used. For gender and education, I used the Blau index (Blau, 1977). Finally, for type of employment (fixed or temporary) I used the percentage of temporary workers in each team.
Data Analysis
One of the objectives of this study was to develop a reliable measurement instrument to operationalize the different elements of the I-P-O model. In order to test the measurement instrument in general, an exploratory factor analyses with Direct Oblimin Rotation and a scale reliability analysis were conducted. In total seven factor analyses were performed, for both output variables separately, for transition process, action process and interpersonal process variables, for the team composition and task design variables (i.e. input), and for the context variables. Furthermore, the scale reliability, also called internal consistency of the items, was examined to confirm whether the items could be grouped into one variable. Factor analyses and internal consistency analyses both checked whether the constructs were measured properly.
The next step in the data analysis was to justify aggregation of the variables to team variables, in order to test which variables do relate with team effectiveness. There are multiple requirements for checking whether aggregation is justified. First, measures must refer to the level of interest. In this study, this was the case since almost all input and process items refer to the team level, except for affect and satisfaction. Another requirement to aggregate the variables to the team level is that the variables must show a significant F-test (i.e. One-Way ANOVA) to check whether the within group variance is lower than the between group variance. Finally, a number of reliability statistics, except for internal consistency, must be examined. Therefore, IRR (interrater reliability) and IRA (interrater agreement) statistics were calculated.
22 interpreted as a reliability statistic, that ideally must be higher than .70 (LeBreton & Senter, 2008). It assesses the reliability of average ratings rather than the reliability of a single rating. For example, if another random sample of raters rate the same subjects, ICC(2) is aproximately the correlation between the averaged ratings from the two sets of raters. So, it is a reliability statistic of the group mean. The ICC(2) must in any case be greater than or equal to ICC(l). Both interrater reliability statistics indicate whether there is relative consistency in ratings provided by multiple judges of multiple targets. So, it checked whether the respondents rated relatively consistent with other targets.
23 Lastly, in order to test the research framework of team effectiveness, correlation analyses were performed. Correlation was used instead of regression due to the large number of variables in the model. Regression would not yield any satisfying results, since there was too much multicollinearity between the variables, which means that there are high correlations among the latent exogenous constructs (Grewal, Cote & Baumgartner, 2004). This would result in unstable estimates. Results from the correlation analyses are presented in the next section.
Sample
After aggregating the variables to the team level, teams were selected for the correlation analyses for testing the team effectiveness framework. The final analyses were only based on teams that meet three criteria. First of all, the teams must consist of six team members or more. Secondly, teams must comply with the team definition of this research. Thirdly, teams must have sufficient interrater agreement scores. Therefore, 6 teams were deleted from the 85 teams, because they did not meet the number of team members criteria. Another 6 teams were deleted, since they did not comply with my definition of teams. For example, these teams did not depend on each other at all or team members did not even know each other yet. This also included the teams that did not have one single supervisor. Finally, 11 teams did not have appropriate values of interrater agreement. These teams were all deleted for the final analysis of the team effectiveness framework. This resulted in a final study sample of 647 people across 62 teams. Average age in this team sample is 43 years (SD = 10.89) with 35.5% of the respondents being male and almost all respondents (99.8%) from Dutch origin.
24
RESULTS
Preliminary Analysis
25 variables loaded perfectly on their underlying hypothesized factors, whereas some variables had crossloadings. In general, these were the factors about leadership and affect because these factors contained many items.
Interrater agreement and reliability analyses. Several types of reliability were examined (Table 1). First, internal consistency reliabilities (Cronbach’s alpha’s) were inspected. Table 1 shows that all the Cronbach’s alpha’s were at least higher than .60, which is sufficient. Moreover, even 21 out of 24 variables had a Cronbach’s alpha higher than .70. In particular, the team process variables were highly internally consistent. To conclude, based mainly on the conceptual distinctions between the characteristics, but supported by the above performed factor and reliability analysis for the purpose of testing the reliability and validity of the measurement instrument, in total 23 characteristics are kept as separate variables in the analyses below.
As noted in the method section, multiple respondents were assessed from each team under the assumption that respondents’ ratings would reflect a shared reality within each team. Therefore, it would be expected that ratings from different respondents from the same team are similar to one another, and furthermore that they are more similar to one another than they are to respondent ratings from other teams. In order to examine this and to justify aggregation, the average interrater agreement coefficient (James et al., 1984) and the intraclass correlation coefficients (Bartko, 1976) are presented in Table 1.
Firstly, ICC(1) shows that all F-tests, one-way ANOVA analyses, are significant, which is a necessary condition for aggregating the variables to the team level. This suggests that all ratings within teams are significantly different between teams. Secondly, ICC(1) values show that most of the variance in ratings is due to team membership, with values around .15. Even though all were significant, some scores were modest in size <.05, specifically organizational appreciation and positive affect. Thirdly, the ICC(2) reliability of group means was for all variables around .60. However, these range from .18 till .81. Again, the same variables, organizational appreciation and positive affect, had modest values. Finally, without exceptions, all agreement scores (rᵇ mean in Table 1) are higher than .70, suggesting that raters within one team agreed with one another. However, the range of the variables suggests that some teams also have rather low interrater agreement scores (rWG scores).
26
TABLE 1
Means, standard deviation and reliability statistics
Mean SD n rᵃ ICC(1) ICC(2) rᵇ mean rᵇ sd rᵇ minimum
rᵇ maximum Control variables Age 43.63 10.81 759 Type of employment¹ 0.08 .27 753 Tenure 6.58 5.67 758 Gender² 0.35 .48 768 Gender diversity 0.34 .17 85 Education level³ 5.42 1.23 724 Educational diversity 0.46 .18 85 Gross salary 2656 1097 740 Team composition Team size 11.80 7.24 85 Functional diversity 0.49 .28 85 Team Fluidity 3.01 1.51 381 .86 .29*** .78 .82 .22 .00 1.00 Task design Team autonomy 4.52 1.33 381 .91 .05** .33 .83 .23 .00 1.00 Task interdependence 4.54 1.22 379 .76 .19*** .67 .90 .10 .50 1.00 Interteam interdependence 4.89 1.30 375 .89 .11*** .52 .85 .19 .02 1.00 Clear goals 4.60 1.38 377 .82 .15*** .61 .83 .15 .27 1.00 Challenging goals 4.91 1.11 375 .66 .16*** .62 .90 .15 .00 1.00 Context Team training 4.32 1.18 378 .60 .07** .40 .86 .16 .00 1.00 Organizational resources 4.22 1.17 375 .70 .11*** .51 .87 .13 .19 1.00 Organizational appreciation 2.48 1.30 371 .81 .03* .23 .76 .26 .00 1.00 Transformational leadership 4.96 .92 729 .95 .33*** .81 .99 .01 .97 1.00 Transactional leadership 3.85 .68 711 .61 .07** .40 .99 .01 .96 1.00 Transition processes Team reflexivity 4.31 1.31 378 .92 .09*** .46 .86 .26 .00 1.00 Temporal planning 4.21 1.18 375 .79 .09*** .46 .87 .21 .00 1.00 Action processes Team effort 5.43 1.05 373 .88 .07** .40 .92 .14 .00 1.00 Team monitoring 3.76 1.36 379 .94 .09*** .46 .87 .19 .00 1.00 Backup-behavior 5.52 1.17 378 .94 .13*** .57 .90 .16 .00 1.00 Team coordination (problems) 3.56 1.19 375 .84 .16*** .63 .93 .07 .55 1.00 Team communication 4.63 1.09 378 .81 .17*** .64 .95 .07 .54 1.00 Interteam coordination and cooperation 4.07 1.22 371 .86 .11*** .52 .86 .21 .00 1.00
Interpersonal processes Positive affect 3.79 .52 726 .84 .03* .18 .98 0.02 .91 1.00 Negative affect 2.13 .59 724 .85 .07*** .40 .98 0.03 .82 1.00 Collective efficacy 4.99 1.11 373 .89 .22*** .71 .98 .02 .91 1.00 Psychological safety 5.01 1.22 375 .84 .19*** .67 .95 .05 .77 1.00 Output Satisfaction 5.59 1.16 733 .88 .06** .37 .88 .14 .00 .99 Performance 4.70 .93 82 .92
* p< .10 ᵃInternal consistency reliability (cronbach's alpha) ¹Dummy variable 1 = temporary 2 = fixed contract ** p<.05 ᵇInterrater agreement (James et al., 1984) ²Dummy variable 1 = male
27 team monitoring and team reflexivity. These variables have many out-of-range rWG(J) scores, which suggests that raters did not agree on these items or interpreted the constructs differently. All in all, interrater reliability and agreement scores suggest high consistency and agreement of raters among most variables, only some variables showed low ICC scores and some had widespread ranges of rWG scores. These variables should be interpreted with cautious. Nevertheless, based on the above presented reliability statistics and interrater agreement values, individual variables were aggregated to the team level.
Descriptive statistics. Besides the reliability statistics and agreement scores, Table 1 also provides the means and standard deviations of all variables.
Intercorrelations. As an additional assessment of the independence of the characteristics, intercorrelations among the individual scales were examined. Table 2 presents the correlations between all individual variables. In particular, team processes appear to be highly intercorrelated. For instance, team communication and collective efficacy show a high positive intercorrelation. There are also some strong negative intercorrelations observed, for example clear goals correlates negatively with team coordination problems. Moreover, temporal planning correlates with team reflexivity, which are both transition processes. As a final example, all interpersonal processes, such as collective efficacy and psychological safety, are also highly intercorrelated.
28
TABLE 2
Correlations of individual variables
n 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Team composition 1)Team size 85 2) Functional diversity 85 .35** 3) Team Fluidity 381 .28* .15 Task design 4) Team autonomy 381 -.27* -.10 -.17**
5) Task interdep endence 379 .20 .26* .20** .01
6) Interteam interdep endence 375 -.19 -.03 -.04 .05 .06 7)Clear goals 377 -.25* -.24* -.26** .26** -.03 -.01 8)Challenging goals 375 .08 .19 -.09 .06 .18** .26** .17** Context 9) Team training 378 -.29** -.16 -.08 .18** -.02 .04 .25** .07 10) Organizational resources 375 -.27* -.18 -.13 * .15** -.02 -.11* .42** -.09 .25** 11) Organizational ap p reciation 371 .36** -.26* .-09 .05 .01 .06 .12 * .01 .28** .24** 12) Transformational leadership 729 -.29** -10 -.03 .20 ** .16** .07 .21** .07 .30** .24** .20** 13) Transactional leadership 711 -.21 .06 .05 .07 .00 -.06 .09 .10 .08 .14 ** .24** .33** Transition processes 14) Team reflexivity 378 -.08 -.04 -.23 .23 .48 -.21 .22 -.06 .59* .52 -.36 .39** .15** 15) Temp oral p lanning 375 -.09 -.13 -.40 -.44 -.44 .62* .56 .77** .27 .19 -.03 .37** .19** .64**
Action processes 16) Team effort 373 -.18 -.07 -.66* .21 -.25 .10 .68* .57 .73** .56* .04 .35** .14** .48** .,39** 17) Team monitoring 379 -.09 -.13 .41 .19 .44 -.25 -.36 -.40 -.47 -.13 -.52 .28** .17** .43** .36** .19** 18) Backup -behavior 378 -.13 -.20 -.49 .35 -.34 .23 .37 .57 .29 .30 .12 .29** .16** .45** .47** .60** .17** 19) Team coordination (p roblems) 375 .34** .23* .41 -.13 .29 -.09 -.75 ** -.40 -.50 -.38 .31 -.29** -.07 -.45** -.44** -.41** -.21** -.52** 20) Team communication 378 -.28** -.20 -.40 .12 -.25 .15 .65* .52 .40 .47 -.20 .39** .08 .63** .59** .49** .27** .60** -.63** 21) Interteam coordination
and coop eration 371 -.28** -.09 -.34 .57
* .32 -.39 .13 -.01 .56* .66* -.22 .30** .12* .34** .37** .26** .27** .28** -.45** .42** Interpersonal processes 22) Positive affect 726 -.21* -.19 -.19** .13* -.01 -.09 .23** .06 .15** .14** .17** .28** .05 .33** .26** .31** .20** .29** -.33** .42** .22** 23) Negative affect 724 .32** .17 .22** -.14** .14** .09 -.24** .07 -.24** -.30** -.20** -.17** -.01 -.21** -.19** -.14** -.12* -.24** .36** -.35** -.29** -.49** 24) Collective efficacy 373 -.25* -.08 -.51 .16 -.48 .33 .46 .57 .51 .31 -.12 .40** .11* .55** .50** .62** .25** .57** -.71** .69** .46** .42** -.39** 25) Psy chological safety 375 -.25* -.03 -.18 .18 -.15 .23 .36 .62* .17 .05 -.14 .34** .08 .55** .44** .54** .17** .63** -.57** .67** .36** .37** -.33** .67** * p <.05
29
Test of research framework
In order to answer the first part of the research question “Which input and process
variables affect team effectiveness?” multiple correlation analyses on the team level are performed. I correlated the control variables with the output variables (Table 3) in order to make sure that these do not influence the team effectiveness criteria. Furthermore, a correlation analysis was performed for the process with the output variables (Table 4), the input with the process variables (Table 5) and the input with the output variables (Table 6) for aiming to test the complete research model.
Control variables. Table 3 shows that the control variables do not have a significant relation with any of the team effectiveness variables, except for gross salary, which is positively related with financial performance.
TABLE 3
Correlations control variables and output variables
Process variables. Table 4 presents the correlations between process characteristics and team effectiveness. All process characteristics relate significantly with team satisfaction, except for interteam coordination and cooperation. In addition, all process characteristics also relate with team performance judged by team managers. Thus, in general all transition processes, action processes and interpersonal processes are important in predicting team satisfaction and team performance. Remarkably, there were only two process characteristics that related with financial performance, namely interteam coordination and cooperation and positive affect.
30
TABLE 4
Correlations of process and output variables
Input variables. Table 5 presents the correlations between the input and process characteristics. From Table 5 it is derived that the team composition characteristics were mostly related to action and interpersonal processes, except for functional diversity that only has some marginal relations with a few action processes. Likewise, task design characteristics show almost no relations with team processes, except for team autonomy and clear goals, which are related to almost all processes. So, team autonomy and clear goals might be factors of importance.
However, context characteristics appear to be most predictive for team processes. Some characteristics influence almost all three team processes, especially team training is of great importance. Moreover, it is worth noting that there are also some team processes that are only related to context characteristics, these are team reflexivity and team monitoring. Other
31
TABLE 5
Correlations of input and process variables
Transition processes Action processes Interpersonal processes
32 interesting context variables are transactional leadership and organizational appreciation. Transactional leadership shows no relation with interpersonal processes, and organizational appreciation is merely related with negative affect (see Table 5). So, more appreciation from the organization means less negative affect within teams. As a final point, team reflexivity and positive affect show the least correlation with input characteristics.
Table 6 presents correlations between the input characteristics and team effectiveness. Table 6 shows that some input characteristics are also directly related with team effectiveness. First of all, team fluidity relates negatively with both team satisfaction and team performance. It was already shown that team fluidity also relates to several processes, which in turn relate with team effectiveness. Therefore, it is possible that some processes mediate this effect. Secondly, team autonomy relates positively to all three team effectiveness criteria. Since team autonomy was also to a large extent related to team processes, again mediation of
TABLE 6
Correlations of input and output variables
33 processes may play a role. Finally, clear goals suggests a quite strong direct relationship with both team effectiveness characteristics, probably without mediation.
Interesting are the relations between input characteristics and financial performance. Although financial performance was only marginally related to some team processes, financial performance is related with input characteristics. It appears that especially context characteristics relate with financial performance. Noteworthy, organizational resources relate positively to all three team effectiveness criteria. So, the resources provided by the organization are of particular interest.
34
TABLE 7
Most important process – output relations
TABLE 8
Most important input – process relations
Team satisfaction Team performance Transition processes Team reflexivity + + Temporal planning ++ ++ Action processes Team effort ++ ++ Backup-behavior ++ ++
Team coordination (problems) -
--Team communication ++ ++
Interteam coordination and cooperation + Interpersonal processes Positive affect ++ ++ Negative affect -- --Collective efficacy ++ ++ Psychological safety + ++ + p<0.01, r<.40 - p<.01, r<-.40 ++ p<.01, r>.40 --p<.01, r>-.40
Transition processes Action processes Interpersonal processes
35
DISCUSSION
This research examined an input, process, output model of team effectiveness. This model proposed a relationship between input characteristics and team effectiveness through team processes. Team effectiveness was measured among 62 teams using three different sources, team satisfaction judged by team members, team performance judged by team managers and financial performance obtained from the management information system. It was expected that the input variables would be more related to the process variables and that the process variables would be more related with the output variables. Results show that in general this model is supported. The next section discusses the value of the I-P-O model and the contributions of this study to research and practice.
Theoretical contributions to literature about team processes
This study contributes by highlighting the importance of the I-P-O model. First of all, results suggest that all transition, action and interpersonal processes related strongly with team satisfaction and team performance judged by managers. Both transition processes, namely team reflexivity and temporal planning, related positively with team effectiveness. This finding shows that it is important for teams to regularly plan and evaluate between performance episodes. This is conforming earlier found literature (Carter & West, 1998; Hu & Liden, 2011; Janicik & Bartel, 2003).
Furthermore, the results of action processes clearly suggest that these processes are strongly related to team effectiveness. This is in accordance with previous research (Hoegl et al., 2004; Steiner, 1972; Wilke & Meertens, 1994). One remarkable finding in this perspective is that interteam coordination and cooperation did positively relate with financial performance. This means that teams must carefully coordinate their activities with other teams. It has to be remarked that a characteristic of this healthcare organization is that teams are overall highly interdependent. Therefore, a lot of cooperation and coordination between the teams is needed in order to align activities. If this is done effectively, it appears to relate to higher financial performance.
36 and Bell (2001) respectively Edmondson (1999). Moreover, positive affect shows a positive relation with financial performance, as well as team satisfaction that was also positively related with financial performance. This means that financial performance is related with two individually measured variables. Nevertheless, the causality of these relations is not clear. Are teams more satisfied when they perform well financially or do teams financially perform better if they are more satisfied? The first option is more likely, but a longitudinal study must examine the direction of this relation. Knowing which factors predict financial performance will also be very important for the organization.
Concerning the interpersonal processes, previous research (Kozlowski & Ilgen, 2006; Marks et al., 2001; Mathieu et al., 2008) distinguished emergent states from team processes. According to Marks et al. (2001) emergent states are cognitive, motivational and affective states of teams, which are typically dynamic. Examples of emergent states are team efficacy, trust, team cohesion, affect and climate (Kozlowski & Ilgen, 2006; Mathieu et al., 2008). The focus of this study was mainly on processes, since the distinction between team processes and emergent states was not explicitly made. However, they are included in the I-P-O model because all interpersonal processes used in the I-P-O model of this study can be regarded as emergent states. So, this study is very complete in investigating team effectiveness since it also covers the emergent states of teams.
Theoretical contributions to literature about input characteristics
37 difficult to assist and help each other, because they lack knowledge or competence. Ancona and Caldwell (1992) already demonstrated that functional diversity did not yield higher performance since it inhibits processes and team effectiveness through reduced information sharing. Accordingly, Bunderson and Sutcliffe (2002) found that dominant function diversity had a negative effect on information sharing and team performance. Future research focusing on functional diversity might help to gain more insight into the relationship between functional diversity, team processes and team effectiveness.
Furthermore, results about task design characteristics suggest that especially team autonomy and clear goals were important for almost every team process. Besides the indirect effect of both factors through team processes, team autonomy and clear goals were directly related to team performance, also found by Cohen and Bailey (1997) and Wageman et al. (2005). In particular, this study contributes by highlighting the importance of clear goals for team effectiveness. Results show that having clear goals is highly related to most other characteristics in this study, including both team effectiveness characteristics. Results fail to support relations with challenging goals, contrary to the earlier findings of Hackman et al. (2000). This effect might be mediated by motivation, as already suggested by Hackman et al. (2000), who argues that teams will be more motivated to succeed in their task if they have challenging goals. Consistent with this finding, interdependence within and between teams did scarcely relate with team processes or team effectiveness, in contrast to what was found by Shea and Guzzo (1987). It may be that goal interdependence is important here, since Van der Vegt et al. (2001) found that task interdependence was only related with team satisfaction and team performance when goal interdependence was high. What is found is that teams that were highly interdependent within teams experienced more coordination problems, as expected by Galbraith (1978).
38 relates positively with financial performance. This finding is conform Wageman et al. (2005), who argue that organizations must provide positive consequences for excellent team performance. For this organization it is argumentative that teams obtain more appreciation if they score high on financial performance. However, this causal relationship is not tested. Future research might also investigate the direction of this relation. Results about managerial support show that both transformational leadership and transactional leadership related with a number of process variables. In contrast to findings of Bass and Avolio (1993), both were not directly related to team effectiveness. As expected and suggested by the findings of this study, this effect is probably being mediated through team processes, for instance team communication or coordination. It goes without saying that transactional leadership was not related with interpersonal processes due to the ignorance of personal feelings by transactional leaders.
To conclude, the organizational context is of high importance for this organization. If teams obtain more training, resources and support from the organization, their satisfaction, performance and even their financial performance will be higher. Though, the direction of the relationship between organizational resources and financial performance is not straightforward, since it sounds reasonably that teams with more resources perform financially better.
Theoretical contributions to literature about the I-P-O model
39 addition, this study contributed by validating the measurement instrument since all constructs were highly reliable and could be aggregated to the team level. In summary, the I-P-O model remains, and the measurement instrument is a reliable tool for researching team effectiveness in field settings.
PRACTICAL IMPLICATIONS
Besides theoretical contributions, this study has practical implications for organizations. Firstly, information provided about the relations between input and process characteristics with team effectiveness induce the practical implication that we could design more effective teams based on these results. Though, some characteristics found in this study are easier to change than other characteristics. For instance, input characteristics are more difficult to change by team managers. Team composition and task design are mostly changed in more rigorous ways. However, clear goals are quite easily established by the team itself or by the team manager, which is a very important characteristic in relation with team processes and team effectiveness. Context characteristics, such as organizational support and training can only be influenced by the organization. In particular for this organization, context characteristics were of high importance. So, it is important for this organization to realize that they could affect their team effectiveness by for instance providing the teams with enough resources and information to carry out their tasks.
Team managers should try to intervene at the process level, since this study showed that all process characteristics related with team effectiveness characteristics. Above that processes are easier to change by team managers than input characteristics, but though input characteristics can be changed. For instance, transformational leadership can be used to influence team processes, since transformational leadership related to many processes. Thus, managers might train themselves in having a clear vision, and motivate, inspire and develop their employees.
40 So, identifying and validating the input and process characteristics in this study was a first step in learning how to design effective teams. Furthermore, the study contributed by providing a practical measurement tool for team effectiveness in field settings (Appendix A). This tool can be used in many organizations to provide insight in the functioning of their teams and the characteristics that are most predictive for these organizations in designing more effective teams.
LIMITATIONS AND FUTURE RESEARCH
It is important to note some limitations of this study, which relate directly to future research issues. Some of them were already discussed. First of all, the objective measure used in this study was doubtful. Financial performance and team performance judged by team managers did not correlate at all, in contrast of what one should expect. Team managers assessed their teams on achieving goals, achieving deadlines, operating speed, quality of the work, productivity, and effectiveness. No correlation between objective and subjective performance implies that team managers did not take the financial situation of teams into account when assessing the teams on performance. Another reason could be that the used measure of financial performance was not a sufficient indicator of actual performance of teams. For some teams it was not clear what their revenues and costs exactly were and whether this financial indicator was a good parameter for these teams. Doubtful teams were mostly overhead teams, such as the management team, secretaries and the HR team. For these teams the data was a bit indistinct since it was not always clear where their costs were actually budgeted. Moreover, these teams had mainly costs and no revenues. It should also be taken into account that financial data was collected from January till April, whereas the individual data was collected in April and May. Therefore, the results about financial performance should be interpreted cautiously. The study did also not include other objective measures of team performance, so results were only based on subjective team performance judged by team managers. These scores could for example be more positively than they in fact were. It is not sure whether this was a reliable team effectiveness characteristic, because it is not measured against other objective performance criteria. For future research it would be advisable that studies include reliable objective measures of team performance, like absence, turnover rates, customer satisfaction or productivity measures.
