The ability, motivation and opportunity to self-organize

(1)

The ability, motivation and opportunity

to self-organize

Master thesis, Msc. Human Resource Management University of Groningen, Faculty of Economics and Business

June 19th, 2013 Anneke Oenema Student number: 2221292 Multatulistraat 11 9721 NE Groningen Tel: +31 (0)623088577 E-mail: Anneke.Oenema1@rug.nl Supervisor/University: Prof. Dr. E. Molleman Supervisor/Noorderpoort: M. Riemersma-Diephuis

(2)

2

Towards self-organizing teams

The ability, motivation and opportunity

to self-organize

ABSTRACT

This study examined the relationship between the motivation, ability and opportunity to perform and performance. In this research performance refers to executing self-organizing tasks. Hypotheses are tested on a sample of 256 employees from different work teams at Noorderpoort with a response rate of 46.9%, using hierarchical multiple regression analysis. The results show that there are significant positive relationships between the ability, motivation and opportunity to perform organizing tasks and the performance of self-organizing tasks. Above this, at the task-level, opportunity to perform seems to be by far the most critical factor to perform self-organizing tasks. This study contributes to the body of work examining the effects of ability, motivation and opportunity to organize on self-organizing. Furthermore, the outcomes of this study give insight to the participating organization which empowered tasks are already done by the employees and identify possible barriers which hinder the performance of these tasks. In practice, the organization can use the outcomes of this study to implement self-organizing teams and achieve a higher level of performance.

(3)

3

INTRODUCTION

In literature, as well as in practice, a lot of attention has been given to empowered teams and related concepts, such as self-managing work teams and self-organization (e.g., Manz, 1992; Stewart & Manz, 1995; Hut & Molleman, 1998; Balkema & Molleman, 1999; Wellins, Byham & Wilson, 1991). Empowered teams are often seen as a solution for organizational problems and are introduced to increase the effectiveness of the organization, as well as improving the quality of working life for employees (Manz, 1992; Molleman, Delft & Slomp, 2001; Wellins et al., 1991). Critical for self-organization is the issue of local autonomy, which is described by Morgan (1986) as the principle of “minimal critical specification”. According to this principle “management should define as little as possible how a team should perform their tasks, but provide just enough guidelines to ensure that the team members are able to perform their tasks properly while still getting as much autonomy as they can handle to make their own contribution” (Balkema & Molleman, 1999).

Buurtzorg Nederland, a Dutch homecare organization, consists of self-organized teams and had the highest employee satisfaction amongst all large Dutch companies in 2011. In 2012 they received the ‘Best Employer Award’ out of 269 participating companies. As a response, a lot of Dutch organizations are thinking of the possible development and implementation of self-organizing teams. Academic researchers also have dealt with the matter of the development of self-organizing teams (Wellins et al., 1991; Van Amelsvoort & Scholtens, 1993; Hut & Molleman, 1998) and a lot of research has been done concerning the opportunities and barriers for the implementation of, and the effectiveness of self-managing teams (e.g. Balkema & Molleman, 1999; Molleman, 2000; Cohen & Ledford, 1994; Sims & Manz, 1995).

(4)

4 situational factors that enable action” (Rothschild, 1999; Siemsen et al., 2008: 427). The level of performance (i.e., self-organizing in the current research) can be referred to the verb “do”, which considers the extent to which employees already do specific tasks that are part of an empowered state (Molleman et al., 2001).

The relationships that will be examined in this study have both theoretical and practical relevance. Previous research focused on self-organizing teams and its effects on performance (Cohen & Ledford, 1994; Sims & Manz, 1995). What has been left out in previous research is that this relationship will not be obvious if team members do not have the ability, the motivation and the opportunity to perform the tasks that are delegated to the team by management. This study contributes to the body of work examining the effects of ability, motivation and opportunity to self-organize on self-organizing behavior (Vroom, 1964; Boxal & Purcell, 2011). The expectation is that the ability to perform as well as the motivation and opportunity to perform will have a positive influence on the level of performance. Furthermore, the opportunity to perform will moderate this relationship as such that when the opportunity to perform the tasks is high, the relationship between the motivation and ability to perform with performance will be strengthened.

This study will give the participating organization insight into which empowered tasks are already done by the employees and will identify possible barriers which hinder the performance of these tasks. In practice, the organization can use the outcomes of this study to implement self-organizing teams and achieve a higher level of performance. If the proposed interactions between the ability, the motivation and the opportunity to perform and the level of performance exist, the organization should not only consider the delegation of tasks to self-organizing teams, but also take into account the ability and the motivation employees have to perform these specific tasks.

While there have been much research on self-organizing teams, requirements such as the ability, motivation and opportunity to perform self-organizing tasks have received little or no attention in determining the precursors of performance in self-organizing teams. Hence, the broad objective of this study is to examine the relationships between the ability, motivation and opportunity to perform and the actual level of self-organizing. Therefore, the leading research question is as follows:

(5)

5 THEORY AND HYPOTHESES

Behavioral aspect of performance

Different authors have agreed that performance has behavioral and outcome perspectives (Campbell, McCloy, Oppler & Sager, 1993; Sonnentag & Frese, 2002). The behavioral aspect of performance refers to measurable behaviors that are relevant for the achievement of organizational goals. The behavioral perspective put forth that performance is in the doing, not in the result of what has been done (Campbell et al., 1993). The outcome aspect of performance refers to the objective consequences of individual behavior (Sonnentag & Frese, 2002). Tying performance to behavior has practical and conceptual advantages in terms of diagnosing and measuring performance (Motowidlo, 2003). There is a tendency to use task behaviors instead of outcomes, particularly when output standards are hard to define (Gerhart, Rynes & Fulmer, 2009) and this is probably the case for the teams involved in the current study. This study adopts the behavioral view of performance and will look at the extent to which employees actually execute self-organizing tasks.

Requirements for implementing self-organizing teams

(6)

6 Human Resource policies should be oriented toward influencing one of these domains (Jiang et al., 2012b). To influence the knowledge and skills of employees, recruitment, selection and training policies are required. The motivation to perform can be influenced by performance management policies such as performance management and career development. Beyond improving the employee’s knowledge, skills and motivation required to perform well, an organization needs to design work in such a way that it allows employees to exert their knowledge and skills and motivation via opportunities to achieve organizational objectives (Jiang, Lepak, Hu & Baer, 2012a). To offer these opportunities, job design policies and involvement policies are generally used. To realize involvement policies, practices like empowerment, employee involvement and self-organizing teams are common (Jiang et al, 2012b).

Motivation to perform

The first requirement for implementing self-organizing teams has to do with the attitudes of the team members, which are related to their psychological needs (Balkema & Molleman, 1999). Psychological needs may differ as such that some individuals may have growth needs, such as the desire for personal development and learning while others have social needs in the form of significant social relationships (Alderfer, 1972). Some team members are mainly intrinsically motivated and are looking for challenging jobs and will favor autonomous decision making and task variety while others are mainly extrinsically motivated, for example through financial rewards.

Individual performance is argued to be a function of both motivation and ability (Vroom, 1964). Motivation represents “the willingness of an individual to act” (Boxal & Purcell, 2011; Siemsen et al., 2008: 427). Often employees know how to perform correctly, but for one reason or another, choose not to do so because of lack of motivation. If team members are not willing to perform the tasks that are delegated to the team by management, performance will fall behind. Consequently, this study hypothesizes that;

Hypothesis 1. There is a positive relationship between the motivation to perform self-organizing tasks and the actual level of doing these tasks.

Ability to perform

(7)

7 of the employees involved. In general, ability represents “the knowledge and skills of an individual related to the action” (Rothschild, 1999; Siemsen et al., 2008: 427). In other words, this study will look at the ability of employees to conduct the delegated tasks. A deficiency with respect to these abilities may limit the development of self-organization (Balkema & Molleman, 1999). Consequently, this study hypothesizes that;

Hypothesis 2. There is a positive relationship between the ability to perform self-organizing tasks and the actual level of doing these tasks.

“Motivated employees without the necessary knowledge and skills will not achieve maximum contributions” (Jiang et al., 2012b: 77). Even if employees have the motivation to perform the tasks that are delegated to the team, if they do not have the skills and learning abilities that are needed to perform these tasks, performance will fall behind. Consequently, this study hypothesizes that;

Hypothesis 3. Ability to perform moderates the relationship between the motivation to perform and the level of self-organizing such that when an individual employee has the ability to perform self-organizing tasks, the relationship between the motivation to perform and the level of self-organizing will be strengthened.

Opportunity to perform

A lot of support is particularly profound for the role of ability and motivation in performance, while the role of opportunity is less often explicitly found. However, in many work situations individuals who are both capable and motivated to accomplish tasks successfully, may either be inhibited in or prevented from doing so due to situational constraints which are beyond their control (Peters & O’Conner, 1980; Tuuli & Rowlinson, 2009). “Opportunity to perform” can be defined as “the particular configuration of the field of forces surrounding a person and his or her task that enables or constrains that person’s task performance and that are beyond the person’s direct control” (Blumberg & Pringle 1982: 565). In other words, the opportunity to perform implies “the environmental or situational factors that enable action” (Siemsen et al., 2008: 427).

(8)

8 empowerment finally entails that on the operational level employees get more say (power) and management get less say. Eventually this will also ask for a shift in the role of management towards their employees. Instead of being a supervisor and have direct control they will become more a facilitator and coach to their employees (Stewart & Manz, 1995). Leaders may have resistance to the implementation of self-organizing teams because they realize that this could cost them their jobs or at least change their hierarchical position and their job (Golembiewski, 1995). In addition, leaders may believe that team members are incapable of managing themselves and therefore will not give them the opportunity for self-organization. Ford, Quinones, Sego and Sorra (1992) found evidence that lack of opportunity to perform tasks has a negative effect on performance. Consequently, this study hypothesizes that;

Hypothesis 4. There is a positive relationship between the opportunity to perform self-organizing tasks and the actual level of doing these tasks.

“Talented employees, who are motivated, but lack the opportunities to use their abilities, will likely display lower contributions than similar employees provided with such opportunities” (Jiang et al., 2012b: 77-78). This means that even if employees have the motivation to perform the self-organizing tasks, as long as management does not give them the opportunity to perform these tasks, this will negatively affects the relationship. Consequently, this study hypothesizes that;

Hypothesis 5a. Opportunity to perform moderates the relationship between the motivation to perform and the level of self-organizing such that when an individual employee has the opportunity to perform self-organizing tasks, the relationship between the motivation to perform and the level of self-organizing will be strengthened.

Besides, if employees are given proper opportunities to perform the self-organizing tasks, those who are capable of making a contribution to the organization will be able to fully exploit their knowledge and skills (Gerhart, 2007; Jiang et al., 2012b: 78). Consequently, this study hypothesizes that;

(9)

9 The hypotheses delineated above are depicted in Figure 1.

FIGURE 1 Conceptual model

In summary, this study distinguishes three classes of requirements for implementing self-organizing teams, which will be labeled as “motivation to perform”, “ability to perform” and “opportunity to perform” respectively. In the next section a more detailed description of the organization will be given as well as of the instruments that have been used to test the hypotheses.

METHODS

Sample and procedure

This study examined the ability, motivation and opportunity to perform self-organizing tasks and the actual performance of self-self-organizing tasks in a field setting. A survey research was conducted and participants were recruited at Noorderpoort, a large center for secondary, vocational and adult education and training (in Dutch: ROC) in the Northern part of the Netherlands. This organization trains 17,500 students, at 22 different schools with about 1,400 employees. The organization strives to strengthen its human capital; with pride, passion and professionalism towards the future. One of the conditions to make this happen is that employees have or will have the opportunity to regulate their own education and business. Inspired by the work structure of Buurtzorg Nederland, the management of Noorderpoort is thinking of the possibilities of the (further) development of self-organizing teams.

Motivation to perform

Opportunity to perform

Ability to perform Performance (e.g. level of

(10)

10 To make an inventory of tasks that may be empowered to the team by management, brainstorm sessions were held with different managers and employees working at Noorderpoort. In addition, the organizational concept of Buurtzorg has been examined. After this, possible self-organizing tasks were translated to the context of Noorderpoort and constructed into a questionnaire for the employees of the organization.

To test the hypotheses discussed in the theory section, twenty team managers at Noorderpoort were contacted by e-mail to arouse their interest for the research and to acquire permission and cooperation from the organization to approach employees. Eighteen managers indicated their interest in the research and agreed to participate. These managers, in turn, asked their subordinates whether they were willing to participate in the research. Questionnaires were prepared for all the 256 team members of the eighteen teams that had agreed to participate in the research (Appendix A). Teams were selected based on part of the results coming from an Employee Satisfaction Survey, which was held during the period January-February 2013 amongst all employees of Noorderpoort (Effectory, 2013). Amongst others, employees were asked in the Employee Satisfaction Survey questions about team performance and their team members. Specific items included; ‘I have sufficient influence on the work decisions in my team’, ‘I am satisfied with my immediate colleagues’, ‘My team is working on team development on a good way’. Item scores were averaged to form total scores for team performance and aggregated to the team level. The ten teams that had the highest score on team performance as well as the ten teams that had the lowest score on team performance were selected, to accomplish a high variance level of potential outcomes in this research.

(11)

11 bench-mark report which will indicate the differences in outcomes between their own team and the other teams at Noorderpoort.

The data were collected in March and April 2013. Questionnaires were distributed digitally using ROCspiegel, a tool that is specifically designed for quality assurance in education. With the help of this instrument, the questionnaire was quickly and easily distributed amongst the large group of respondents. Before the questionnaire was distributed, a trial version was distributed amongst some employees. On March 20th, 2013 the selected team members received an email with an introduction letter on why the survey was conducted and were requested to participate in the research by filling in the questionnaire before April 22nd, 2013. This email contained a link that led the respondent directly towards the questionnaire on ROCspiegel. Two weeks after the start of the questionnaire, an email was sent to all the respondents to thank participants for their response or to encourage them to respond if they have not already done so. On April 22nd, 2013 the link to the questionnaire was closed.

A total of 256 employees were invited to participate in the survey and questionnaires were distributed across eighteen teams, each team consists of 7 to 24 persons. The selected teams only consisted of teachers who had a fixed contract at Noorderpoort. 141 participants completed the survey resulted in an overall response rate of 55.1%. The final sample size used for the analysis was reduced to 120 (remaining response rate of 46.9%), due to 21 partially completed questionnaires. All respondents with missing values were deleted, as none of them filled-out the questions pertaining to the dependent variables.

Of the remaining participants, 49.2 % were male and 50.8% were female. Participating employees’ average age was 50.46 years (SD = 9.85), ranging from 26 to 64 years. The average time that respondents were member of their current organizational team (i.e., team tenure) was 8.55 years (SD = 7.69) and the average time that respondents were working at Noorderpoort was 13.91 years (SD = 10.88). To obtain demographic data of all respondents, the management personnel information system of the organization was used. The age of respondents and the time that respondents were working at Noorderpoort were based on the reference date of March 1st, 2013.

Measures

(12)

12 to Dutch, since the pool of potential respondents was Dutch speaking. A copy of the questionnaire including all items used is attached in Appendix A.

The questionnaire started with listing twelve possible self-organizing tasks. For each task the participants were asked to fill in the same four items. ‘I am motivated to perform this task’ (motivation to perform self-organizing tasks), ‘I have the knowledge and skills that are needed to perform this task’ (ability to perform self-organizing tasks), ‘My immediate supervisor gives me the opportunity to perform this task’ (opportunity to perform organizing tasks), ‘Within my team, this task is already performed’ (performance of organizing tasks). According to Mrs. M. Riemersma-Diephuis, some of the formulated self-organizing tasks cannot be performed at the individual level; therefore performance is measured at the team-level. The Cronbach’s alphas for these four scales were .81, .85,.87 and .77 respectively. Note that a reliability coefficient of .70 or higher is considered as being acceptable in most social science research situations (Nunnally, 1978), so it can be said that the internal consistency of these scales are sufficient.

Besides, the AMO variables were also measured by using existing scales that were already tested for validity and reliability (see below).

Motivation to perform was operationalized by asking team members to what extent

they have the motivation to do their job. This variable was measured amongst team members using four items developed by Hackman & Oldham (1976). These items were: “My opinion of

myself goes up when I do my job well”, “I feel a great sense of personal satisfaction when I do my job well”, “I feel bad and unhappy when I discover that I have performed poorly on my job”, “My own feelings generally are not affected much by how well I do on my job” (reverse

coded). Item scores were averaged to form total scores for the motivation to perform. The Cronbach’s alpha for these four items was .72, which is larger than .70, so it can be said that the internal consistency of this scale is sufficient.

Ability to perform was operationalized by asking team members to what extent they

have the ability to do their job. This variable was measured amongst team members using three items developed by Spreitzer (1995). These items were: “I have confidence in my ability

to do my work”, “I am self-assured in respect of my capabilities to carry out my work”, “I have learned the skills I need for my work”. Item scores were averaged to form total scores for

(13)

13

Opportunity to perform was operationalized by asking team members to what extent

they have the opportunity for actually conducting self-organizing tasks. This variable was measured amongst team members using six items developed by Spreitzer (1995). These items were: “I can decide for myself how I should tackle my work”, “I have ample opportunity and

freedom regarding the way in which I carry out my work”, “I have considerable autonomy in determining how I do my work”. Next to this, they were asked the following items (Spreitzer,

1995): “I have a great deal of control on what goes on within my team”, “My impact on what

goes on within my team is large”, “I have considerable influence on what goes on within my team”. Items scores were averaged, to form total scores for the opportunity to perform. The

Cronbach’s alpha for these six items regarding to the opportunity to perform was .86, which is larger than .70, so it can be said that the internal consistency of this scale is sufficient.

Control variables. A number of control and demographic variables were measured to

identify if these variables were related to the dependent variable. At the individual level, gender, age and team tenure were measured using single item questions. Team members’ age and team tenure (i.e. length of team membership) were used as control variables in all of the analysis. Past research has shown that team tenure influences the responses of individuals (e.g. Jehn, Northcraft, & Neale, 1999; Van Der Vegt, Emans, & Van De Vliert, 2000). Moreover, team tenure can have a positive influence on team performance because of the increasing expertise. On the other side, authors also mention that long team tenure could have a negative influence on team performance because team members may experience burnout (Morales & Marquina, 2009). Age and team tenure were measured in years. Gender was measured as a dichotomous variable and coded such that the number 1 represents male and 2 represents female (Becker, 2005). The respondents are members of a specific team and, therefore, respondents are nested in teams. An F-test showed that the teams differed with respect to the dependent variable (F [17,102] = 2.35, p < .01) and, therefore, the hypotheses were tested by means of multi-level analyses.

Data analysis

(14)

14 ability to perform, motivation to perform and opportunity to perform, and the interaction terms of the ability and motivation to perform and the opportunity to perform.

The purpose of this study was to explain how much variance in the level of self-organizing (performance) is explained by each independent variable (ability to perform, motivation to perform and opportunity to perform self-organizing tasks) and by the influence of moderating effects. To investigate if the proposed interactive relationships exist, six moderated hierarchical regression analyses were conducted. The hypotheses were tested using a hierarchical multiple regression analysis. In the first step, the control variables age, gender and team tenure were entered in the regression analysis. In the second step the independent variables motivation to perform, ability to perform and opportunity to perform were introduced to test hypothesis 1, 2 and 4. In the third step, the multiplicative term (independent variable x moderator) were entered to determine the interaction effect and to test hypothesis 3, 5a and 5b.

RESULTS Descriptive statistics and correlations

Table 1 presents all the means, standard deviations, and inter-correlations. Interestingly, when focusing on the model variables, significant positive correlations between these variables can be discovered. The motivation to perform self-organizing tasks is significantly and positive related to the motivation to perform ‘in general’ (r = .21, p < .05), the ability to perform self-organizing tasks (r = .85, p < .01), the ability to perform ‘in general’ (r = .31, p <.01), the opportunity to perform self-organizing tasks (r = .59, p < .01), the opportunity to perform ‘in general’ (r = .51, p < .01) and the performance of self-organizing tasks (r = .27, p < .01). In addition, Table 1 shows that the motivation to perform ‘in general’ has a significant and positive relationship with the opportunity to perform self-organizing tasks (r = .18, p < .05).

The ability to perform self-organizing tasks is significantly and positive related to the ability to perform ‘in general’ (r = .45, p < .01), the opportunity to perform self-organizing tasks (r = .56, p < .01), the opportunity to perform ‘in general’ (r = .50, p < .01) and the performance of self-organizing tasks (r = .35, p < .01). Furthermore, Table 1 shows that the ability to perform ‘in general’ has a significant and positive relationship with the opportunity to perform ‘in general’ (r = .29, p < .01).

(15)

self-15 organizing tasks (r = .61, p < .01). Furthermore, the opportunity to perform ‘in general’ has a significant and positive relationship with the performance of self-organizing tasks (r = .40, p < .01).

Regarding the control variables, age does have a significant positive relationship with team tenure (r = .34, p < .01). Above that, there is a significant negative relationship between gender and opportunity to perform ‘in general’ (r = -.21, p < .05), indicating that women perceive less opportunities to perform ‘in general’ than men.

TABLE 1

Descriptive Statistics (Mean and Standard deviations) and Research Variable Intercorrelations

Note. N = 120.

** Correlation is significant at the .01 level (2-tailed). * Correlation is significant at the .05 level (2-tailed).

Hypotheses testing

As the model assumes that the independent variables (ability, motivation and opportunity to perform) influence the dependent variable (performance of self-organizing tasks), linear regressions were conducted to test this model and the presented hypotheses. Analyses were repeated controlling for age, gender and team tenure, but the results were essentially identical. The results of the multiple regression analysis are shown in Table 2 to 7. The significance level (p-value) reflects the probability that this effect emerges by chance, which should be lower than 5% (n.s. stands for not significant).

Variables Mean S.D. 1 2 3 4 5 6 7 8 9 10

1. Age 50.46 9.85

2. Gender 1.51 0.50 .02

3. Team tenure 8.55 7.69 .34** -.07 4. Motivation to perform tasks 3.80 0.15 -.14 -.09 -.12 5. Motivation to perform 4.20 0.60 -.09 .17 -.09 .21* 6. Ability to perform tasks 3.78 0.62 -.12 -.11 -.05 .85** .13

7. Ability to perform 4.35 0.54 -.08 -.13 -.00 .31** .12 .45** 8. Opportunity to perform tasks 3.49 0.72 -.11 -.06 -.07 .59** .18* .56** .15

(16)

16 Hypothesis 1 stated that there is a positive relationship between the motivation to perform self-organizing tasks and the actual level of doing these tasks. Table 2 shows that the motivation to perform self-organizing tasks has a significant positive relationship with the performance of self-organizing tasks (B = .25, p < .01). This indicates that the actual level of doing self-organizing tasks is higher when employees have the motivation to perform these self-organizing tasks. This result supports the first hypothesis. In addition, the motivation to perform ‘in general’ shows an insignificant relationship with the performance of self-organizing tasks (B = -.02, n.s.). This means that this result does not support the first hypothesis.

TABLE 2

Results of Regression Analyses for Hypothesis 1

Model 1 Model 2

Steps and variables Estimate SE Estimate SE

Control variables

Age -.07 .10 -.05 .09

Gender .08 .10 .10 .09

Team tenure -.04 .10 -.02 .09

Main effect

Motivation to perform tasks .25** .08

Δ χ2 (2) 8.14

Model 1 Model 2

Control variables

Age -.07 .10 -.07 .10

Gender .08 .10 .08 .10

Team tenure -.04 .10 -.04 .10

Main effect

Motivation to perform ‘in general’ -.02 .09

Δ χ2 (2) 0.04

Note: unstandardized regression coefficients are presented, N = 120. ** Correlation is significant at the .01 level (2-tailed).

(17)

17 Next, hypothesis 2 stated that there is a positive relationship between the ability to perform self-organizing tasks and the actual level of doing these tasks. Table 3 shows that the ability to perform self-organizing tasks has a significant positive relationship with the performance of self-organizing tasks (B = .30, p < .01). This indicates that the actual level of doing organizing tasks is higher when employees have the ability to perform these self-organizing tasks. This result supports the second hypothesis. In addition, the ability to perform ‘in general’ has an insignificant relationship with the performance of self-organizing tasks (B = .07, n.s.). This means that this result does not support the second hypothesis.

TABLE 3

Model 1 Model 2

Control variables

Age -.07 .10 -.04 .09

Gender .08 .10 .11 .09

Team tenure -.04 .10 -.05 .09

Main effect

Ability to perform tasks .30** .08

Δ χ2 (2) 12.60

Model 1 Model 2

Control variables

Age -.07 .10 -.06 .10

Gender .08 .10 .08 .10

Team tenure -.04 .10 -.04 .10

Main effect

Ability to perform ‘in general’ .07 .09

Δ χ2 (2) 0.66

(18)

18 Hypothesis 3 stated that the ability to perform moderates the relationship between the motivation to perform and the level of self-organizing such that when an individual employee has the ability to perform self-organizing tasks, the relationship between the motivation to perform and the level of self-organizing will be strengthened. Table 4 shows that the two-way interaction between motivation to perform self-organizing tasks and the ability to perform self-organizing tasks on the performance of self-organizing tasks is not significant (B = .12, n.s.). This result does not support the third hypothesis. In addition, the two-way interaction between the motivation to perform ‘in general’ and the ability to perform ‘in general’ on the performance of self-organizing tasks is not significant (B = -.06, n.s.). This result also does not support the third hypothesis. Unfortunately, based on the results, it can be concluded that the ability to perform does not have a moderating influence on the relationship between the motivation to perform and the level of self-organizing.

TABLE 4

Model 1 Model 2 Model 3

Steps and variables Estimate SE Estimate SE Estimate SE

Control variables

Age -.07 .10 -.04 .09 -.04 .09

Gender .08 .10 .10 .09 .11 .09

Team tenure -.04 .10 -.05 .09 -.05 .09

Main effect

Motivation to perform tasks -.04 .09 -.03 .16

Ability to perform tasks .34* .16 .36* .16

Δ χ2 (2) 12.68

Interaction effect

Motivation to perform tasks x Ability to perform tasks .12 .08

Δ χ2 (1) 2.30

Control variables

Age -.07 .10 -.07 .10 -.07 .10

Gender .08 .10 .09 .10 .10 .10

Team tenure -.04 .10 -.04 .10 -.04 .10

Main effect

Motivation to perform ‘in general’ -.02 .09 -.02 .09

Ability to perform ‘in general’ .07 .09 .08 .09

Δ χ2 (2) 0.74

Motivation to perform ‘in general’ x Ability to perform ‘in general’ -.06 .09

Δ χ2 (1) 0.40

(19)

19 Hypothesis 4 stated that there is a positive relationship between the opportunity to perform self-organizing tasks and the actual level of doing these tasks. Table 5 shows that the opportunity to perform self-organizing tasks has a significant positive relationship with the performance of self-organizing tasks (B = .60, p < .01). This indicates that the actual level of doing self-organizing tasks is higher when employees have the opportunity to perform these self-organizing tasks. This result supports the fourth hypothesis. In addition, the opportunity to perform ‘in general’ has a significant relationship with the performance of self-organizing tasks (B = .33, p < .01). This result also supports the fourth hypothesis.

TABLE 5

Model 1 Model 2

Control variables

Age -.07 .10 -.02 .08

Gender .08 .10 .05 .08

Team tenure -.04 .10 -.04 .08

Main effect

Opportunity to perform tasks .60** .07

Δ χ2 (2) 50.70

Model 1 Model 2

Control variables

Age -.07 .10 -.04 .09

Gender .08 .10 .10 .09

Team tenure -.04 .10 -.06 .09

Main effect

Opportunity to perform ‘in general’ .37** .09

Δ χ2 (2) 15.70

(20)

20 Hypothesis 5a stated that the opportunity to perform moderates the relationship between the motivation to perform and the level of self-organizing such that when an individual employee has the opportunity to perform self-organizing tasks, the relationship between the motivation to perform and the level of self-organizing will be strengthened. Table 6 shows that the two-way interaction between the motivation to perform self-organizing tasks and the opportunity to perform self-organizing tasks on the performance of self-organizing tasks is not significant (B = .09, n.s.). This result does not support hypothesis 5a. In addition, the two-way interaction between the motivation to perform ‘in general’ and the opportunity to perform ‘in general’ on the performance of self-organizing tasks is not significant (B = -.01, n.s.). This result also does not support hypothesis 5a. Unfortunately, based on the results, it can be concluded that the opportunity to perform does not have a moderating influence on the relationship between the motivation to perform and the level of self-organizing.

TABLE 6

Results of Regression Analyses for Hypothesis 5a

Control variables

Age -.07 .10 -.03 .08 -.03 .08

Gender .08 .10 .03 .07 .03 .07

Team tenure -.04 .10 -.05 .08 -.06 .08

Main effect

Motivation to perform tasks -.14 .09 -.12 .09

Opportunity to perform tasks .69** .09 .67** .09

Δ χ2 (2) 53.06

Motivation to perform tasks x Opportunity to perform tasks .09 .07

Δ χ2 (1) 1.53

Control variables

Age -.07 .10 -.04 .09 -.04 .09

Gender .08 .10 .11 .09 .11 .09

Team tenure -.04 .10 -.06 .09 -.06 .09

Main effect

Motivation to perform ‘in general’ -.02 .08 -.02 .09 Opportunity to perform ‘in general’ .37** .09 .37** .09

Δ χ2 (2) 15.77

Motivation to perform ‘in general’ x Opportunity to perform ‘in general’ -.01 .07

Δ χ2 (1) 0.00

(21)

21 Lastly, hypothesis 5b stated that the opportunity to perform moderates the relationship between the ability to perform and the level of self-organizing such that when an individual employee has the opportunity to perform self-organizing tasks, the relationship between the ability to perform and the level of self-organizing will be strengthened. Table 7 shows that the two-way interaction between the ability to perform self-organizing tasks and the opportunity to perform self-organizing tasks on the performance of self-organizing tasks is not significant (B = .08, n.s.). This result does not support the last hypothesis. In addition, the two-way interaction between the ability to perform ‘in general’ and the opportunity to perform ‘in general’ on the performance of self-organizing tasks is not significant (B = .00, n.s.). This result also does not support the last hypothesis. Unfortunately, based on the results, it can be concluded that the opportunity to perform does not have a moderating influence on the relationship between the ability to perform and the level of self-organizing.

TABLE 7

Results of Regression Analyses for Hypothesis 5b

Control variables

Age -.07 .10 -.02 .08 -.01 .08

Gender .08 .10 .06 .08 .06 .08

Team tenure -.04 .10 -.04 .08 -.05 .08

Main effect

Ability to perform tasks .01 .08 .02 .09

Opportunity to perform tasks .60** .09 .57** .09

Δ χ2 (2) 50.70

Ability to perform tasks x Opportunity to perform tasks .08 .08

Δ χ2 (1) 1.01

Control variables

Age -.07 .10 -.04 .09 -.04 .09

Gender .08 .10 .11 .09 .11 .09

Team tenure -.04 .10 -.06 .09 -.06 .09

Main effect

Ability to perform ‘in general’ -.01 .09 -.01 .09

Opportunity to perform ‘in general’ .37** .09 .37** .10

Δ χ2 (2) 15.71

Ability to perform ‘in general’ x Opportunity to perform ‘in general’ .00 .09

Δ χ2 (1) 0.00

(22)

22 Secondary analyses

Additionally, to explore the relationships between the AMO variables and performance at a more detailed level, several multi-level analyses were conducted at the task-level, by considering the interactions between the AMO-variables and performance per self-organizing task. In this section only the significant interactions will be discussed.

Table 8 presents all the means, standard deviations, and inter-correlations. Interestingly, when focusing on the model variables, many significant positive correlations between these variables can be discovered at the task-level.

TABLE 8

Descriptive Statistics (Mean and Standard deviations) and Research Variable Intercorrelations

Variables Mean S.D. 1 2 3 4

1. Motivation to perform task 1 4.29 0.88

2. Ability to perform task 1 4.09 0.85 .72**

3. Opportunity to perform task 1 3.94 1.15 .42** .42**

4. Performance of task 1 3.49 1.19 .12 .15 .37**

4. Performance of task 2 3.35 1.09 .35** .38** .84** 1. Motivation to perform task 3 3.98 0.83

(23)

23

Variables Mean S.D. 1 2 3 4

3. Opportunity to perform task 7 3.81 1.11 .20* .38**

4. Performance of task 7 3.59 1.25 .06 .33** .68** 1. Motivation to perform task 8 4.07 1.04

4. Performance of task 8 4.18 0.89 .17 .24* .45**

4. Performance of task 10 3.99 1.20 .21* .26** .33** 1. Motivation to perform task 11 4.01 1.01

4. Performance of task 12 3.27 1.07 .02 .03 .45**

Note. N = 120.

The results of the multi-level analyses at the task-level are shown in Table 8 to 11. Hypothesis 1 stated that there is a positive relationship between the motivation to perform self-organizing tasks and the actual level of doing these tasks. Table 8 shows that for 7 of the 12 tasks there is a significant positive relationship between motivation and performance. So at the task level there is some support for this relationship.

(24)

24 this task for 10 of the 12 tasks. This indicated that for most tasks the actual level of doing a self-organizing task is higher when employees have the ability to perform this self-organizing task. These results mainly support the second hypothesis.

Next, hypothesis 3 stated that the ability to perform moderates the relationship between the motivation to perform and the level of self-organizing such that when an individual employee has the ability to perform self-organizing tasks, the relationship between the motivation to perform and the level of self-organizing will be strengthened. Table 9 shows that the two-way interaction between motivation to perform self-organizing tasks and the ability to perform self-organizing tasks on the performance of self-organizing tasks is significant for two tasks (task 7 and 12). For task 7 (B = -.25, p < .05) the direction of this effect is contrary to the expectations. According to this result, the ability to perform task 7 has a negative effect on the relationship between the motivation to perform and the level of self-organizing. This result is depicted graphically in a two-way interaction plot (Figure 3). In addition, Table 9 shows that that the two-way interaction between motivation to perform organizing tasks and the ability to perform organizing tasks on the performance of self-organizing tasks is significant for task 12 (B = -.16, p < .05) but again the direction is contrary to the expectation. The results are depicted graphically in a two-way interaction plot (Figure 4). Both graphs indicate that motivation has a negative relationship with performance when the ability to perform is high and a positive relationship when the ability to perform is low. This opposes hypothesis 3.

TABLE 9

Significant results of Regression Analyses for Hypothesis 3 at the task-level

Dependent variable: Performance

Steps and variables Estimate SE

Main effect

Motivation to perform task 7 -.23 .12

Ability to perform task 7 .23* .11

Motivation to perform task 7 x Ability to perform task 7 -.25* .11

Main effect

Ability to perform task 12 -.05 .12

Motivation to perform task 12 x Ability to perform task 12 -.16* .07 Note: unstandardized regression coefficients are presented, N = 120.

(25)

25 FIGURE 3

Two-way interaction plot hypothesis 3 – Self-organizing task 7

FIGURE 4

Two-way interaction plot hypothesis 3 – Self-organizing task 12

2,0 2,5 3,0 3,5 4,0 4,5 5,0 low high P e r fo r m a n c e o f ta sk 1 2

Motivation to perform task 12

(26)

26 Hypothesis 4 stated that there is a positive relationship between the opportunity to perform self-organizing tasks and the actual level of doing these tasks. Table 8 shows that the opportunity to perform self-organizing tasks has a significant positive relationship with the performance of all the self-organizing tasks. This indicates that the opportunity to perform self-organizing tasks has a strong positive relationship with performance. Opportunity to perform seems to be by far the most critical factor to perform self-organizing tasks.

(27)

27 TABLE 10

Significant results of Regression Analyses for Hypothesis 5a at the task-level

Main effect

Opportunity to perform task 6 .39* .15

Motivation to perform task 6 x Opportunity to perform task 6 .63* .24

Main effect

Opportunity to perform task 7 .68** .10

Motivation to perform task 7 x Opportunity to perform task 7 .35** .12

Main effect

Motivation to perform task 9 .08 .15

Motivation to perform task 9 x Opportunity to perform task 9 .40** .12

Main effect

Motivation to perform task 10 x Opportunity to perform task 10 .37* .15 Note: unstandardized regression coefficients are presented, N = 120.

FIGURE 5

(28)

28 FIGURE 6

Two-way interaction plot hypothesis 5a – Self-organizing task 7

FIGURE 7

(29)

29 FIGURE 8

Two-way interaction plot hypothesis 5a – Self-organizing task 10

(30)

30 TABLE 11

Significant results of Regression Analyses for Hypothesis 5b at the task-level

Main effect

Ability to perform task 2 .04 .17

Ability to perform task 2 x Opportunity to perform task 2 .45* .18

Main effect

Ability to perform task 5 -.03 .19

Opportunity to perform task 5 .21 .19

Main effect

Ability to perform task 11 x Opportunity to perform task 11 .27* .12 Note: unstandardized regression coefficients are presented, N = 120.

FIGURE 9

(31)

31 FIGURE 10

Two-way interaction plot hypothesis 5b – Self-organizing task 5

FIGURE 11

(32)

32 FIGURE 12

Two-way interaction plot hypothesis 5b – Self-organizing task 11

2,0 2,5 3,0 3,5 4,0 4,5 5,0 low high P er fo rm a nc e o f ta sk 1 1

Ability to perform task 11

opportunity :high opportunity :low

DISCUSSION

In this final section the finding results of the study are discussed. Besides that, both theoretical and practical implications, strengths and limitations, and directions for future research are discussed.

Findings and theoretical implications

The purpose of this study was to examine the relationships between the ability, motivation and opportunity to perform, and the actual level of self-organizing. The results of this study disclose that there is a positive direct relationship between the ability, the motivation and in particular the opportunity to perform and actual performance. Besides the study found some support for moderating effects at the task-level. Also for these moderations the opportunity to perform seems to be most critical.

(33)

33 found by several authors for the main effects of ability, motivation and opportunity on performance (e.g. Locke, Mento & Katcher, 1978; Lepak et al., 2006; Siemsen et al., 2008). By regarding performance as a function of three different dimensions (ability, motivation and opportunity to perform), individual traits and characteristics are emphasized that are important for performance as well as institutional responsibility and structures that are related to whether employees can perform optimally (Blumberg & Pringle, 1982).

The results of the secondary analysis of this study, at the task-level, indicate that opportunity to perform seems to be by far the most critical factor to perform self-organizing tasks. Although most of the research of the past has focused on motivational (such as motivation and satisfaction) or ability issues (such as knowledge and intelligence) as being the key mechanisms through which HR systems influence employee performance, researchers argue that more research is needed to consider the opportunity to perform, the third predictor of employee performance (e.g. Blumberg & Pringle, 1982; Lepak et al., 2006). Working conditions need to be examined such as “equipment, materials, leader behavior, actions of coworkers, mentoring, organizational policies, rules and procedures, norms and information” to enhance the opportunity to perform (Blumberg & Pringle, 1982).

Another hypothesis in this study was whether motivation and ability interact to produce performance (Locke et al., 1978). The results of this study do not support the proposition that the ability to perform self-organizing tasks moderates the relationship between the motivation to perform and the level of self-organizing (no support for hypothesis 3). The amount of variance that the motivation to perform accounted for in the actual level of performance of organizing tasks, was therefore independent of the ability to perform self-organizing tasks. This outcome is in line with and builds upon past research (Anderson & Butzin, 1974: 603; Locke, 1965), which argues that there is no interaction effect, “motivation and performance are positively related, independent of the value of ability”.

(34)

34 contradictory results. Some authors mention that performance was predicted by motivation only for high ability individuals (French, 1985; Fleischman, 1958; Lawler, 1966). On the other hand, Locke (1965) did not find evidence for an interaction effect of motivation and ability on performance at all. A possible explanation for these contradictory results is given by Locke et al. (1978: 270), is that “moderator effects are caused by differences in homogeneity with respect to causal factors”. The research of Locke et al. (1978) has shown that motivation was a better predictor of performance when groups were homogeneous with respect to ability, than within groups which were heterogeneous in their ability to perform.

In line with the former, the results of this study do not support the proposition that the opportunity to perform moderates the relationship between the motivation toperform and the level of self-organizing (no support for hypothesis 5a). The amount of variance that the motivation to perform accounted for, in the actual level of performance of self-organizing tasks, was therefore independent of the opportunity to perform self-organizing tasks. Lastly, the results of this study do not support the proposition that the opportunity to perform moderates the relationship between the ability to perform and the level of self-organizing (no support for hypothesis 5b). The amount of variance that the ability to perform accounted for, in the actual level of performance of self-organizing tasks, was therefore independent of the opportunity to perform self-organizing tasks. However, the results of the secondary analyses of this study, at the task-level, do support the proposition that the opportunity to perform moderates the relationship between the motivation to perform and the level of self-organizing (hypothesis 5a), with regard to four self-organizing tasks as well as the proposition that the opportunity to perform moderates the relationship between the ability to perform and the level of self-organizing (hypothesis 5b), with regard to four self-organizing tasks.

(35)

35 performance if opportunity is the constraining factor. Managerial interventions aimed at the AMO-variables that are not constraining in the organization, are less likely to be effective (Siemsen et al., 2008: 441) and therefore less likely to result in better performance within self-organizing teams.

Notably when considering the results of the secondary analysis, it is very important to give employees the opportunity to perform because although they have the ability or motivation to perform, as long as they do not get the opportunity to perform, performance will deteriorate. If employees are motivated to recruit and select new team members (task 6), and if they are motivated to mentor and coach new team members and trainees (task 7) it is especially important that they are given the opportunity to perform these self-organizing tasks. In this way, employees will remain motivated to perform tasks 6 and 7 which may positively influence their performance. The same applies for the situation in which employees are motivated to schedule team meetings and make an agenda (task 9) and if they are motivated to chair team meetings (task 10). When employees are given the opportunity to perform tasks 9 and 10, their motivation will remain and have a positive influence on the performance of the self-organizing tasks.

By looking at the two-way interaction between the ability to perform self-organizing tasks and the opportunity to perform organizing tasks on the performance of self-organizing tasks, there is no overlap in content between the different tasks (task 2, 5, 8 and 11) that show significant results. Notwithstanding it is still important to consider that if the opportunity to perform is low, ability to perform has a negative relationship with performance. For instance, even if an employee is able to inform team members about changes in work schedules (task 5), as long as he or she does not get the opportunity to perform this task, performance will fall behind.

Practical implications

(36)

36 perform these specific tasks. Therefore, it is valuable to focus on the minimum levels of ability, motivation and opportunity that are required to perform the self-organizing tasks. If these levels are not high enough yet, managerial interventions should be implemented to improve the AMO-variables to a certain level in order to make performance possible.

Opportunity to perform appeared to have the most positive significant relationship with performance. Especially with the implementation of self-organizing teams, it is important that managers give their employees the freedom to decide how to organize the work. This asks for a shift in the role of management towards their employees; from operational supervision to detached team leadership. Team managers will become coordinators as day-to-day supervisory responsibilities are increasingly devolved to empowered work teams (Hales, 2005). However, team managers remain accountable for overall team performance. This requires a conversation with the different team managers in order to investigate if these team managers are willing to give up their old role of supervisor in order to facilitate the performance of self-organizing teams. Team managers who resist should get informed and convinced of their career possibilities within an organization with organizing teams. The motivation of team managers towards the concept of self-organizing teams is determinative, they have to “lead others to lead themselves” (Stewart & Manz, 1995).

(37)

37 long as they do not have trust in their performance and integrity (Hakimi et al., 2010). Noorderpoort really needs to take this into account when considering the (further) implementation of self-organizing teams.

Furthermore, this study has given the participating organization insight into which empowered tasks are already done by the employees and identifies the possible constraining factors which hinder the performance of these tasks. A related benchmark-analysis will give the organization the opportunity to compare the results of the different teams that participated in this research. In addition, the level of self-organizing and constraining factors can be identified per team which may lead to certain managerial interventions.

Strengths and limitations

This research contains some strengths and limitations that should be mentioned. The first strength of this research is the fact that this research is based on multiple-team field data. 18 organizational work teams participated in this research with 120 team members from different organizational teams spread around different schools of Noorderpoort. The results of this research are therefore generalizable for other organizational teams in the organization.

Another strength of this research is that it is unique in the sense that previous research only focused on self-organizing teams and its effects on performance (Cohen & Ledford, 1994; Sims & Manz, 1995). What has been left out in previous research is that this relationship will not be obvious if team members do not have the ability, the motivation and the opportunity to perform the tasks that are delegated to the team by management. This study contributes to the body of work examining the effects of ability, motivation and opportunity to self-organize on self-organizing behavior (Vroom, 1964; Boxal & Purcell, 2011). The methods that were used in this research and the results that were found can serve as guidelines for the implementation of self-organizing teams.

This study has some limitations that should be mentioned. The first limitation is the small number of participants that actually filled in the questionnaire completely. Above the low overall response rate of 46.9%, there were twenty questionnaires that were not completely filled in and therefore excluded from this research. Besides, team size varied from 7 to 24 people and response rates for some teams were quite low. A consequence of this is that the group of respondents may not be representative for the population.

(38)

38 Another limitation is that this research was only performed at one organization, a large center for secondary, vocational and adult education and training in the Northern part of the Netherlands. Future research could improve generalizability if other organizations in different branches and different regions are included.

Finally, this research relied on self-reports to gather the data. A problem associated with self-reporting is common method variance (Spector, 2006). In order to avoid this problem, future research should, for instance, use multiple types of methods (e.g. longitudinal studies) and multiple types of respondents (e.g. team managers for assessing performance).

Directions for future research

Although the results of this study indicate that there are only significant moderating effects at the task-level between the different AMO-variables and performance, there are still some interesting areas for future research. Since research on the interaction effects of the AMO-variables on performance showed mixed results, it is still interesting to research them further. Although the research of Siemsen et al. (2008) discussed the constraining factor on knowledge-sharing behavior, future research might consider the effect of the constraining factor on performance in self-organizing teams.

Past research did not find much empirical evidence supporting the existence of the complementary roles of the AMO-variables on performance (Terborg, 1977). Theory of work performance indicates that ability, motivation and opportunity complement each other in influencing actual behavior (Cummings & Schwab, 1973). In contrast with the hypotheses stated in this research, Siemsen et al. (2008) argue that opportunity, ability, or motivation alone should not lead to performance. Blumberg and Pringle (1982) argue that the AMO-variables are related constructs. For example, an employee with no motivation to perform a self-organizing task, cannot successfully perform a task no matter how highly able he or she may be to do so (Cummings & Schwab, 1973). Future research might consider the existence of complementary roles of the AMO-variables on performance in self-organizing teams.

(39)

39 Lastly, since this research indicates that the opportunity to perform seems to be by far the most critical factor to perform self-organizing tasks, future research is needed to investigate which working conditions are critical and will enhance the opportunity to perform in self-organizing teams.

CONCLUSION

(40)

40 REFERENCES

Alderfer, C.P. 1972. Existence, Relatedness and Growth: Human Needs in Organizational

Settings. Newbury Park: Sage.

Aiken, L. S. & West, S. G. 1991. Multiple Regression: Testing and Interpreting

Interactions. London: Sage.

Anderson, N.H. & Butzin, C.A. 1974. Performance = Motivation x Ability: An Integration-Theoretical Analysis. Journal of Personality and Social Psychology, 30(5): 598-604. Argyris, C. 1998. Empowerment: the emperor’s new clothes. Harvard Business Review,

98–105.

Balkema, A., & Molleman, E. 1999. Barriers to the development of self-organizing teams.

Journal of Managerial Psychology, 14(2): 134-149.

Becker, T.E. 2005. Potential problems in the statistical control of variables in organizational research: A qualitative analysis with recommendations. Organizational Research

Methods, 8(3): 274-289.

Blumberg, M., & Pringle, C. 1982. The missing opportunity in organizational research: some implications for a theory of work performance. Academy of Management Review, 7(4): 560-9.

Boxall, P., & Purcell, J. 2011. Strategy and Human Resource Management, Third edition, Management, Work & Organizations, Palgrave MacMillan.

Campbell, J. P., McCloy, R. A., Oppler, S. H., & Sager, C. E. 1993. A theory of performance. In E. Schmitt, W. C. Borman, & Associates (Eds.), Personnel selection in

organizations, 35-70. San Francisco: Jossey-Bass.

Cohen, S.G., & Ledford, G.E. 1994. The Effectiveness of Self-Managing Teams: A Quasi-Experiment. Human Relations.

Coombs, C. H. 1964. A theory of data. New York: Wiley.

Cummings, L.L & Schwab, D.P. 1973. Performance in organizations: determinants &

appraisal. Scott, Foresman and Company, Glenview, IL.

Effectory, 2013. Customer Extranet, Medewerkersonderzoek Noorderpoort, meting 2 (januari 2013 – februari 2013). https://customer.effectory.com/project/9802/home, accessed on February 15th and March, 6th 2013.