Personnel controls for high potentials: Using self-‐regulation for selection?

(1)

Personnel controls for high potentials:

Using self-‐regulation for selection?

Mark de Jong (s1573853) First supervisor: dr. C. Heijes

Overwinningsplein 92 Second supervisor: Prof. dr. ir. P.M.G van Veen-‐Dirks

9728 CS, Groningen Course: Master’s Thesis BA O&MC

(2)

In front of you is my master thesis, the final product of my master Organizational and Management Control: Business Administration at the University of Groningen. The idea for this thesis originates from an often-‐heard saying that the mindset of elite level athletes could be beneficial in the business context as well. The reasoning is that selecting employees on the basis of this mindset could provide organizations with the high potentials that they are looking for.

This thesis is the result of my internship, During my internship, I worked on a project for Company X. Company X desired a questionnaire that visitors could fill in that predicts the development of talent and success compared to elite athletes.

I want to thank my parents for supporting me during my studies and being an example for me. Next, I would to thank dr. Lolle Schakel for the opportunity to work in this project and the guidance during the process of doing the research and writing my thesis. I also want to thank my supervisor from the university, dr. Coen Heijes, for the guidance during the writing of my thesis. Finally, a general thanks to my friends and family who have supported me during this process and my colleagues for the good times there.

(4)

ABSTRACT

This study’s aim is to investigate the effect of self-‐regulation on career success to determine whether organizations can benefit from using the concept self-‐regulation in their high potential identification controls. A self-‐regulation questionnaire is developed to measure self-‐regulatory skills on 6 dimensions: planning, monitoring, evaluation, reflection, effort and self efficacy. Career success was measured objectively (e.g. salary and promotions) and subjectively (e.g. career satisfaction). In the pretest phase (n = 104), the main aspect is investigating and evaluating the questionnaire’s quality. In the organizational phase (n = 166), the effect of self-‐regulation on career success is investigated. The pretest questionnaire’s results revealed good internal reliability and factor structure. Furthermore, the pretest showed that self-‐regulation was related to salary. The organizational phase results showed that self-‐regulation is related to promotions for Company X employees and is related to subjective career success. Besides, employees in higher performance categories tend to have higher self-‐regulatory skills than employees in lower performance categories. Furthermore, managers have higher self-‐regulatory skills than non-‐managers. Finally, management trainees showed to have significantly higher self-‐regulatory skills than regular employees. Management trainees’ self-‐regulatory skills are at the same level of elite athletes. Regarding self-‐regulation, management trainees have the same mind-‐set in business as elite athletes have in sports. In sum, it appears that self-‐regulation indicators can facilitate organizations in selecting high potentials because higher self-‐regulation is found to be related to aspects of career success and that managers and high potentials in organizations display higher levels of self-‐regulatory skills than regular employees.

(5)

INTRODUCTION

It appears that organizations are increasingly concerned about their talent management (TM). A new study by management consultancy Accenture has found more than two-‐thirds of executives are deeply worried about the threat of not being able to recruit and retain the best talents (Accenture, 2008). The survey of more than 850 top executives from the US, UK, Italy, France, Germany, Spain, Japan and China found that worries about TM were growing, with 67 per cent this year putting it second only behind competition as the key threat, up from six out of ten last year (Accenture, 2008). The necessity for TM appears to be accelerating, as organizations purport to find it increasingly more complex to source skilled labour (Frank and Taylor, 2004). Talent will be tomorrow's prime source of competitive advantage (Anderson, 2001; Chambers, Foulton, Handfield-‐Jones et al., 1998; Makela, Bjorkmann, Ernrhoot, 2010)

The financial health of an organization is predicated on the optimal selection and placement of employees (Hunter, Schmidt & Judiesch, 1990). From a leadership perspective, the ability to identify high potentials serves as a proxy measure for the overall health of the organization and the extent to which the organization will remain viable in the future (Silzer & Church, 2009). Misidentification can mean that individuals are placed in roles for which they are ill equipped which can lead to catastrophic events for organizations (McDonnell, 2011). Especially early selection of talents with the potential to become the future leader, the high potentials, can facilitate the future financial health and be a competitive advantage for the organization. The idea is that career success is not only of concern to the high potentials but that personal success eventually contributes to organizational success (Judge, Higgins, Thoresen, & Barrick, 1999). Individuals that are successful in their career are likely to be the same ones that are successful in their job and thereby help organizations to be successful in their endeavors as well.

(6)

more than their not selected counterparts and leads to a positive upward loop. The high potentials will get more opportunities and will improve more than their counterparts, making them more likely to be selected for higher positions with better opportunities.

However, early identification and selection of high potentials to become a successful manager or future leader is a very difficult task. Consequently, researchers try to identify factors that facilitate or predict career success. A review of Tharenou (1997) identified several categories of influences on career success. The most commonly investigated influences were human capital attributes (training, work experience, education) and demographic factors (age, sex, marital status). A meta analysis by Ng, Eby, Sorensen & Feldman (2005) provided insight in the various dispositional traits as predictors of career success, such as cognitive ability, Big Five personality factors and locus of control. These traits were highly related to career success with correlations ranging from .14 for Big Five personality factors to .47 for locus of control. Ng et al. (2005) also noted that there is only a limited range of variables assessed as predictors of career success. They state other predictors of career success should be identified (Ng et al. 2005). Other predictors could be measures for learning and adaptability. Silzer & Church (2009) mention that many companies do not have suitable assessment measures for learning and adaptability but that companies do have a great deal of interest in including these in their high potential identification process (Silzer & Church, 2009).

(7)

regulation can be an important determinant of future career success as it allows people to more effectively adapt to their roles, required competencies and skills. Therefore, self-‐ regulation could be an important trait of high potentials to be able to fulfill undefined and unknown roles.

Research shows that self-‐regulation has been found to be positively related to success in multiple domains, such as sports, music, task performance, sales and academia (Cellar et al., 2009, Kitsantas & Zimmerman 2002; Nota, Soresi, & Zimmerman, 2004; Porath & Bateman, 2006;). Toering et al., (2009) showed that international soccer players score higher on dimensions of self-‐regulation than national and grass-‐root level players. Kitsantas & Zimmerman (2002) showed that elite level volleyball players make more use of self-‐ regulatory aspects than non-‐experts and novice volleyball players. Porath & Bateman (2006) showed that self-‐regulatory behaviors were positively related to sales performance. Self-‐ regulation distinguishes between performances. Because high potentials are identified as high performers, self-‐regulation may be important for them as well.

Research question:

Can self-‐regulation indicators facilitate the personnel controls in organizations for selecting high potentials?

(8)

been published that comprehensively explore the underlying dimensions of high potential talent, yet this is exactly what organizations are looking for today (Silzer & Church, 2009). This study’s findings could provide insight in the underlying dimensions of high potentials and add to more efficient personnel controls for organizations.

THEORETICAL FRAMEWORK

Control perspective

Management control is broadly defined as all activities through which management tries to contribute to the continuity of the organization (Birnberg, 1998). Managing and controlling talent is becoming a key issue for organizations (Accenture, 2008). Besides this, TM emerges as a distinct strategic business activity because it calls for a greater focus on employees and positions that have the greatest differential impact on business strategy (McDonnell, 2011). Furthermore, TM is very important for succession planning, especially CEO succession (McDonnell, 2011). Therefore, TM is becoming an increasingly important activity through which management contributes to the success, viability and continuity of the organization. Talent management (TM) is becoming a more important part of management control.

(9)

The nature of the role of high potentials and the matching effective control system can be described by Perrow’s (1970) model of structure and technology. This model categorizes the routineness of tasks along two dimensions, task analyzability (high or low) and exceptions (high or low), which creates 4 situations. The degree of established techniques for handling tasks determines analyzability and the degree of variety and novelty in a task determines number of exceptions. The 4 different situations can be found in figure 1. Tasks in cell 1 are repetitive and familiar (few exceptions) and with known routines to handle them (high analyzability). In this situation, the organization can permit action or result controls (Perrow, 1970). Cell 3 is non-‐routine, tasks cannot be programmed and therefore behavior cannot be controlled by procedures or monitored by supervisors. Perrow (1970) suggests that personnel controls are most effective in this situation to ensure that employees themselves have a drive to act in the organizations goals.

Figure 1. Model of structure and technology (Perrow, 1970)

(10)

personnel controls important for high potentials. Research by Abernethy & Brownell (1997) confirms that personnel controls like selection are most effective in a cell 3 situation.

The characteristics of the cell 3 situation of Perrow (1970) can be combined with Ouchi’s (1979) control types (market, bureaucracy and clan). For a control type to be effective, the control type’s social and informational prerequisites have to be met. Market control requires a norm of reciprocity (social) and prices (informational). In a perfect frictionless market, prices can direct employee decision-‐making and behavior. This requires that transactions are fair and honest i.e. norm of reciprocity (social prerequisite). Bureaucratic control requires a norm of reciprocity and legitimate authority (social) and rules (informational). Rules control employee behavior in a bureaucracy and for these rules to be effective employees must accept these rules i.e. legitimate authority. Clan control requires, besides norm of reciprocity and legitimate authority, shared values and beliefs (social) and traditions (informational). Clan control relies on a deep level of common agreement between employees on what constitutes proper behavior. This requires shared values and beliefs and traditions.

(11)

Figure 2. Organizational control: People treatment (adopted from Kelman, 1958)

(12)

select starters that are willing to learn and adapt. Lombardo & Eichinger (2000) confirm this by indicating high potentials using a measure of learning from experience. This is in line with personnel selection controls that are already widely used by companies like intelligence tests.

Figure 3. Dimensions of the control problem (adapted from Birnberg, 1998)

(13)

first purpose of personnel controls is to help to ensure that each employee is able to do a good job. Employees should have all the capabilities and resources needed to do a good job. The second basic function in personnel controls is increasing the likelihood that each employee will engage in self control (Merchant & Van Der Stede, 2007). Self control is the naturally present force that pushes most employees to want to do a good job (Merchant & Van Der Stede, 2007). Learning and motivation variables from the model of Silzer & Church (2009) relate to the second basic function of employee selection. Learning results from self control and motivation is one of the phenomena underlying self control (Merchant & Van Der Stede, 2007). Learning and motivation variables are part of the growth dimension (Silzer & Church 2009). These components can facilitate or hinder a person’s growth and development. Both can be good indicators or whether a person will develop further and learn other skills (Silzer & Church, 2009). Learning and motivation are very important for growth, development and skill acquisition (Kanfer & Ackerman, 1989). From Silzer & Church’s (2009) model, the growth dimension can directly linked to the cell 4 situation of Birnberg (1998) as Sizler & Church (2009) state that typical examples of the growth dimension are adaptability and learning. Therefore, selection on adaptability and learning could be very important for selecting high potentials that should cope with the now unknown future roles and tasks.

Figure 4. Dimensions for the identification of the potential (adapted from Silzer & Church, 2010)

(14)

Self-‐regulation

From psychology we learn that the self-‐regulation process provides humans with an adaptive edge and facilitates their learning process (Zimmerman, 2000). Self-‐regulation is defined as the extent to which individuals are metacognitively, behaviorally and motivationally proactive participants in their learning process (Zimmerman, 2000; 2006). Metacognition is defined as the awareness of and knowledge about one’s own thinking (Zimmerman, 2002). According to Kanfer (1990), self-‐regulation refers to the proximate motivational processes by which persons influence the direction, amount, and form of committed effort during task engagement. Similar in both definitions are the self-‐regulatory processes within the human (metacognition, behavior, motivation) that facilitate certain outcomes (effort during task engagement or learning). This is also suggested by Vancouver (2000), who states that self-‐regulation refers not to the actions per se, but to the processes that mediate or support the actions. Vancouver (2000) defines self-‐regulation as processes involved in attaining and maintaining internal desired states, resulting in human functioning and action.

Self-‐regulation originates from theories about human functioning or actions, the social cognitive theory (Bandura, 1986) and the goal setting theory (Lock & Latham, 1984). In the social cognitive theory, people are neither driven by inner forces nor automatically shaped and controlled by external stimuli (Bandura, 1986). Human functioning is explained by a model of triadic reciprocal influences in which behavior, cognitive and environmental events all operate as interacting determinants of each other (Bandura, 1986). Self-‐regulation is a distinctive feature of the social cognitive theory (Bandura, 1986). Human functioning is motivated and regulated by internal standards and self evaluative reactions to their own actions. In the exercise of self-‐regulation, people set certain standards of behavior for themselves and respond to their own actions self evaluatively. Self-‐regulation mediates the effect of most external influences and provides the basis for purposeful action (Bandura, 1991).

(15)

and attaining these goals is a volitional process (Latham & Lock, 1991). Self-‐regulation occurs through goal setting because the setting of a goal is a discrepancy increasing process. The goal leads to a disequilibrium, requiring effort to reach the goal. As long as people do not attain their goal, this discrepancy will exist. Self-‐regulation processes direct the effort to reach the goal, reducing the discrepancy between the current state and the goal. Goal setting facilitates self-‐regulation in that the goal constitutes what the acceptable level of performance is. Efforts that fall short of the goal result in a negative performance evaluation, efforts that reach or exceed desired goals result in a positive performance evaluation. In both situations, the outcome is a learning experience. In the former situation, this may lead to actions like problem solving, increasing effort or identifying the source of dissatisfaction. In the latter situation, it can lead to actions like setting higher goals (Locke & Latham, 1994). Thus, the self-‐regulatory behavior sequence aligns the person’s current and future behaviors with some criterion that permits the person to evaluate progress toward a specific goal (Kanfer, 1990). The most widely accepted perspective in the industrial and organizational psychology is an integration of the social cognitive theory and the goal setting theory (Kanfer, 2005; Vancouver & Day, 2005). In the integrated perspective, person, social and environmental factors operate in concert to affect an individual’s goals and self-‐ regulatory activities.

(16)

Figure 5. Cyclical self-‐regulatory phases (adapted from Zimmerman, 2000)

According to Kanfer (1990), self-‐regulation also consists of three phases: self monitoring, self evaluation and self reactions. Self monitoring refers to the self-‐observation of thoughts, actions, behaviors, or events. Self evaluation refers to the comparison of current state with the goal, or standard. Self reactions are the implications that persons derive from the congruence of the goal with the standard. The concepts of Kanfer (1990) share similarities with the concepts of Zimmerman (2000). The concept performance /volitional control from Zimmerman (2000) corresponds to the concept of self monitoring of Kanfer, both emphasize the self observation of thoughts and actions. Zimmerman’s concept of self reflections correspond with Kanfer’s concepts of self evaluation and self reactions. Kanfer’s concepts of self evaluation and self reaction are sub concepts in Zimmerman’s concept of self reflection. They emphasize evaluating the experience and responding and reacting to that experience, resulting in learning. In general, Zimmerman’s theory (2000) can be considered more overarching because Kanfer’s concepts (1990) are viewed as sub concepts in the three phases in Zimmerman (2000, see figure 2).

(17)

and tends to produce a product of quality (Beyer, 1987). Important tasks in this step are selecting and sequencing a series of strategies and/or plans to achieve the goal and identifying potential obstacles to the successful attainment of the goal. The second step is self monitoring, checking if the anticipatory plan or strategy is effectively leading to the goal during execution of the plan. Zimmerman (2000) calls it the performance/volitional control phase and Kanfer (1990) names it self monitoring, observing if the (planned) thoughts and actions lead to the goal. Ertmer & Newby (1996) and Vancouver & Day (2005) incorporate (self) monitoring as the second phase as well. Self monitoring is a complex process of awareness of what one is doing, understanding where it fits in the sequence of planned steps and anticipation and planning of what ought to be done next (Ertmer & Newby, 1996). Besides, self monitoring is implementing the steps of the plan while monitoring the effectiveness of the strategy. This is all accomplished as one is engaged in the task itself. The third step is self evaluation, the second phase of Kanfer (1990) and part of the self reflection phase of Zimmerman (2000). After completing the plan from step 1, the process and the product achieved are assessed, to what extent does it complete the preset goal. In the self evaluation phase, the reasonableness and accuracy of the outcomes are assessed, the extent to which the goal was achieved is determined and the effectiveness of the overall plan and its supporting steps in achieving the goal are determined (Ertmer & Newby, 1996). Self evaluation is about determining how efficient and effective the overall plan was so that it can be modified for similar tasks for future use. The fourth step is self reflection, the third and final phase of Zimmerman (2000) & Kanfer (1990). Self reflection is an active process of making sense of past experiences for the purpose of orienting oneself for current and future thought and action, extracting meaning and learning from experiences (Ertmer & Newby, 1996). Reflection can also occur during the task, managing the progress of learning on-‐line and adjusting and changing as new information is assimilated (Ertmer & Newby, 1996). By employing self reflection to evaluate the results of a learning experience, awareness of effective plans and strategies can be increased and ways to use these plans and strategies can be understood (Ertmer & Newby, 1996).

(18)

that those with a belief in their capacity to perform (e.g. high self efficacy) will personally develop, or more likely accept and commit to, difficult performance goals. According to Vancouver & Day (2005), high self efficacy individuals will put more effort towards realizing their goals than low self efficacy individuals. In the face of inevitable setbacks that might evoke goal revision processes, high self efficacy individuals will more likely continue or increase effort, strategize, and persist to attain these more difficult goals than low self efficacy individuals. Or, when things are going well, high self efficacy individuals will more likely increase the difficulty of their goals and/or take on new challenges compared to individuals with low self efficacy. Finally, because of these motivational influences, high self efficacy individuals are more likely to perform well, increasing their self efficacy, and thus spiraling upward in a positive feedback loop (Latham & Locke, 1991). Maintaining this positive loop and high levels of self-‐regulation to accomplish the goals requires a lot of effort. Both effort and self efficacy are included as motivational variables of self-‐regulation. Toering (2010) combined these 6 concepts: planning, monitoring, evaluation, reflection, effort and efficacy in a framework for self-‐regulation. An overview of the 6 concepts of self-‐ regulation is given by Jonker, Elferink-‐Gemser, Visscher (2011) in figure 6.

(19)

Career success and high potentials

Career success can be defined as the real or perceived achievements individuals have accumulated as a result of their work experiences (Judge, Cable, Boudreau, & Bretz, 1995). In other words, career success is generally operationalized in two ways. The ﬁrst includes variables that measure objective or extrinsic career success (Gutteridge, 1973). These include indicators of career success that can be seen and therefore evaluated objectively by others, such as salary attainment and the number of promotions in one’s career (Judge et al., 1995). Others can be performance evaluations by managers or function level (Hoeksema, Vliert & Willams, 1997)

The second way that career success can be operationalized is by variables that measure subjective or intrinsic career success (Judge et al., 1995). Such variables capture individuals’ subjective thoughts about their career attainments, such as job and career satisfaction (Burke, 2001; Judge et al., 1999). The career satisfaction scale of Greenhaus, Parasuraman, Wormle (1990) is a often used self-‐referent measure of subjective career success. Self-‐referent measures are evaluated relative to personal standards. Self-‐referent success criteria generally reflect an individual’s career-‐related standards and aspirations (Heslin, 2005). In contrast, other-‐referent criteria involve comparisons with others, such as whether one is paid more or less than the industry average or a colleague who performs a similar role in the same or another organization (Heslin, 2005). There is preliminary evidence that other-‐referent subjective career success can add unique information about subjective career success (Heslin, 2003). Besides self-‐referent subjective career success, adding other-‐ referent measures could provide new insights when investigating subjective career success.

If organizations select high potentials, they expect that these individuals have the potential to be more successful in their career. Organizational definitions of high potentials clarify this. All refer to the ability to accumulate achievements to higher organizational positions or outstanding achievements (Silzer & Church, 2010).

• Role: 35% of the organizations define high potentials by their ability to effectively move into top / senior management roles

• Level: 25% by their ability to move and effectively perform two positions / levels above their current role

(20)

• Record: 10% by consistent track record of exceptional performance

Higher organizational positions are associated with career success. Higher organizational positions for example, lead to higher salary or have been preceded by promotions. Organizations therefore expect that high potentials achieve high levels of career success.

Hypothesis development

This study investigates the effect of self-‐regulation on career success to determine if self-‐ regulation can facilitate the personnel controls for selecting high potentials in organizations. High potentials are predicted to achieve high levels of career success. If self-‐regulation is related to career success, organizations may benefit from selecting high potentials on self-‐ regulation. To investigate the effect of self-‐regulation on career success, several hypotheses will be proposed. First, this will be investigated using a within group comparison in different populations and for different career success measures.

H1: Higher self-‐regulation is related to more salary.

H2a: Higher self-‐regulation is related to more promotions at the current employer. H2b: Higher self-‐regulation is related to more promotions across a career.

H3a: Higher self-‐regulation is related to higher self-‐referent subjective career success. H3b: Higher self-‐regulation s is related to higher other-‐referent subjective career success.

The above hypotheses operationalize career success as an individual measure between persons or within a group. Besides this, there are also groups that can be identified as being more successful than other groups. This reveals an opportunity to compare groups regarding self-‐regulation (between group comparison). The performance category of employees can indicate career success. Performance categories are often related to increases in pay or promotions. Therefore, if self-‐regulation is related to career success it can be hypothesized that employees in a higher performance category have higher self-‐regulatory skills than people in a lower performance category. Furthermore, organizations define high potentials by exceptional track record (Sizler & Church, 2009). If self-‐regulation is related to high performance categories, this also provides evidence for the importance of self-‐regulation for high potentials.

(21)

H4: Employees in high performance categories have higher self-‐regulation than employees in low performance categories.

A manager can be objectively regarded as more successful than a non-‐manager because management positions are often preceded by promotions, pay more salary, carry more responsibility and are higher in function scale. Therefore, if self-‐regulation is important for career success, it can be hypothesized that managers have higher self-‐regulatory skills than non-‐managers. Furthermore, if self-‐regulation distinguishes between managers and non-‐ managers it can also be important for high potentials because high potentials are expected to fulfill manager positions.

H5: Managers have higher self-‐regulation than non-‐managers.

Earlier research showed that higher self-‐regulatory skills are advantageous or necessary for elite athletes to reach the top. As mentioned in the introduction, Toering et al., (2009) showed that international soccer players score higher on dimensions of self-‐regulation than national and grass-‐root level players. Kitsantas & Zimmerman (2002) showed that elite level volleyball players make more use of self-‐regulatory aspects than non-‐experts and novice volleyball players. Research shows that there has seen a significant increase in the transfer of sport psychology practice to business settings, as evidenced by a huge surge in publications on the subject (Ievleva & Terry, 2008). Principles and predictors of athletic excellence could be transferred effectively to those engaged in business endeavors (Ievleva & Terry, 2008). When investigating the effects of self-‐regulation in a business setting, Porath & Bateman (2006) showed that self-‐regulatory behaviors predicted sales performance. Self-‐ regulatory skills seem to be skills that facilitate expert performance and success. Therefore, self-‐regulation may be important for high potentials who are expected to deliver expert performance.

(22)

The self-‐regulatory skills that facilitate elite athletes to be successful in their career and make the top can also be a factor for high potentials to be selected for the traineeship, make the top and be successful. The mindset to strive for success could be the same for management trainees and elite athletes. Regular employees can be thought of as having normal career success, management trainees being (potentially) high in career success and elite athletes being high as well, although in another domain. Therefore, self-‐regulatory skills in management trainees are hypothesized to be of the same level as those of elite athletes. Besides, it is hypothesized that the self-‐regulatory skills of management trainees and top elite athletes are higher than regular employees.

H6a: Management trainees and elite athletes do not differ in self-‐regulation. H6b: Management trainees have higher self-‐regulation than regular employees H6c: Elite athletes have higher self-‐regulation than regular employees.

METHODOLOGY

To investigate the hypotheses, self-‐regulation and career success are measured using a questionnaire. This study is subdivided in two phases to accurately investigate the hypotheses. The first phase is the pretest phase. The aim of the pretest phase is analyzing and improving the questionnaire and preliminary exploring the relation between self-‐ regulation and career success. Based on the pretest phase, the questionnaire will be adapted and prepared for the second phase. The second phase is the organizational phase. The aim of the organizational phase is to investigate the hypotheses in an organizational setting and clarify the research question: Can self-‐regulation indicators facilitate the personnel controls in organizations for selecting high potentials?

This section starts with the development of the questionnaire. First, the construction of the complete questionnaire and the origination of the different parts will be described. Second, the participants in this study will be described. Third, the variables will be described in detail and fourth the data analysis procedure is discussed.

(23)

Questionnaire development & procedure

The complete questionnaire consists of three parts. The first part is a questionnaire measuring self-‐regulation, the second part is a questionnaire measuring general aspects and the third and final part is a questionnaire measuring several aspects of career success. The complete questionnaire was administered in Dutch.

Self-‐regulation

The six dimensions of self-‐regulation: planning, self monitoring, self evaluation, self reflection, effort and self efficacy, form the self-‐regulation questionnaire. Planning, self monitoring, self evaluation and self reflection were based on Hong et al., 2001; Herl et al., 1999; Howard et al., 2000 and Peltier et al., 2006, respectively. The self efficacy domain was constructed from the Occupational Self efficacy Scale of Shyns & von Collani (2010). Effort was based on the Motivated Strategies for Learning Questionnaire of Pintrich (1990). New items were added to the existing items to be able to select the best items for the questionnaire in the pretest phase. Reverse items were added to avoid the acquiescence bias, the yes-‐saying bias. The newly developed items were constructed in cooperation with two organizational psychologists with a doctor’s degree in industrial & organizational psychology. All questions were modified to be answered on a 5 point Likert scale ranging from almost never (1) to almost always (5).

General aspects

Questions regarding general aspects included age, gender, level of education, years of working experience and years in current employment. These aspects were assessed because of their possible influence on aspects of career success. Age, gender and education were considered control variables in this study.

Career success

Objective career success is measured by several variables: salary, number of promotions, performance evaluation, if someone has a manager position in its organization (Hoeksema, et al., 1997).

(24)

career success questions are based on other-‐referent questions of Heslin (2003). All questions were answered on a 5 point Likert scale ranging from completely disagree (1) to completely agree (5).

Questionnaire development procedure

The self-‐regulation questionnaire was developed using the procedure of Drenth & Sijtsma (2006) for measuring psychological characteristics, which is based on the classical testing theory. The first step in the development of the questionnaire is identification of the psychological characteristic that will be measured. Furthermore, the theory relevant to the to be measured psychological characteristic should be investigated and analyzed. The second step is operationalizing the characteristic to be measured.

Step 1 and 2 are based on the theoretical description of the sub processes of self-‐ regulation (theoretical framework) and previous research on self-‐regulation. This study developed a first version of the self-‐regulation questionnaire (Appendix A). An online version of the complete questionnaire was developed using Qualtrics (Qualtrics Labs Inc., Provo, UT) to administer the questionnaire (Appendix H). An online questionnaire reduces the likelihood of non-‐response rates and increases the quality of the returned questionnaires (Denscombe, 2009).

In the third step according to Drenth & Sijtsma (2006), the questionnaire should be investigated and quantified by giving participants a test version of the questionnaire. The first version of the questionnaire is pretested (n = 104, pretest phase). The first version of the self-‐regulation questionnaire is adapted based on the feedback, response and psychometric analyses of the results. The self-‐regulation items in the first version are evaluated and the best items are selected for the second version of the questionnaire. The statistical analyses are described in the data analysis section: results pretest phase.

(25)

The second version of the questionnaire can be found in Appendix B. The adapted and improved second version of the questionnaire will be administered in an organizational setting, the organizational phase (n = 166). In the organizational phase, the self-‐regulation questionnaire is reduced to 8 items and again psychometrically analyzed. In both questionnaire administrations, career success aspects are measured to investigate if the outcomes of the questionnaire are related to career success. This is exploratory in the pretest phase and the main aspect of the organizational phase. The results of the organizational phase can be found in the section: results organizational phase.

Participants

Pretest phase: A total of 104 complete responses were collected in the pretest phase. In this sample, 42 (40.4%) participants are still studying and 62 (59.6%) participants are working. On average, the students are 23.09 (sd = 2.09) years old, 76.2% are male and most of the students attended school at university level (61.9%). The workers are on average 42,69 (sd = 12,17) years old, have 18,25 (sd = 11,68) years of working experience and are working 10,31 (sd = 10,81) years for their current boss. In the workers group, 53.2% is male and 51.6% completed their education at university level and 38.7% at higher vocational level.

Organizational phase: Company X participated in the survey of the organizational questionnaire. Company X is a large multinational company with over 5000 employees in the Netherlands and 28,700 employees globally. Company X is the second largest HR-‐service provider in the world with market leading positions in various countries. Company X’s annual revenue is 16.2 billion dollars in 2011 with a net income income of 179 million dollars. After receiving company consent, Company X’s employees were contacted for an explanation of the research and were requested for their assistance. The questionnaire was also administered at Company Y and Company Y’s parent company . Both organizations provide selection and assessment services for large companies in the Netherlands.

(26)

athlete sample, 21 participants are already working (see table 1). Company Y contributed 63 participants. Bibliographical descriptives can be found in table 1. All respondents participated voluntarily.

Table 1. Bibliographical descriptives for management trainees, Company X and Company Y employees and elite athletes. Man. Trainees Company X Company X Employees Company Y

employees Elite athletes

N Mean (SD) N Mean (SD) N Mean (SD) N Mean (SD) Age (y) 13 26,62 (1,39) 44 34,09 (7,41) 63 43,49 (9,99) 46 25,41 (5,18) Work experience (y) 13 1,81 (0,88) 44 10,10 (7,51) 63 19,72 (9,38) 21 3,38 (3,75) At current employer (y) 13 1,50 (0,58) 43 5,70 (6,01) 63 9,96 (7,40) 21 2,04 (3,12) Hours per week (h) 13 40,00 (0,00) 44 36,55 (3,82) 63 34,16 (6,58) 21 23,76 (12,13)

N % N % N % N % Gender Male 5 38,5 11 25 26 41,3 21 45,7 Female 8 61,5 33 75 37 58,7 25 44,3 Variables

Self-‐regulation is divided in 6 dimensions: planning, monitoring, evaluation, reflection,

effort and self efficacy. A mean score will be calculated for every self-‐regulatory dimension.

By adding up the 6 average dimension’s scores, a measure for total self-‐regulation is be calculated. The highest score per dimension is 5 (due to 5-‐point Likert scale) resulting in a maximum score for total self-‐regulation of 30 (dimensions * highest score per dimension, 6*5).

Objective career success variables:

Salary: Salary is measured as monthly salary before tax on a fulltime basis. Salary is divided

(27)

PromoCE: Promotions at the current employer is measured by the number of promotions

participants have received in their current occupation, divided by age. Not available for Company X management trainees. (H2a)

PromoTOT: Total promotions in a participant’s career is measured by the number of

promotions participants have received in their career, divided by age. Not available for Company X management trainees. (H2b)

PerfCat: Performance category: Based on the performance grade Company X employees can

determine their performance category: insufficient, reasonable, good, excellent. Company Y and employees were also able to rate themselves in one of these performance categories. For more information about the Company X performance measurement see appendix E for an interview with Nico van Loo, project manager performance management at Company X. Only available in the organizational survey. (H4)

Manager: Participants indicated (yes/no) whether they were manager in the organization

where they were working. (H5)

Subjective career success variables:

SSW: Subjective Self-‐referent career success for Workers was calculated by adding up the

scores of the 5 questions in the career satisfaction scale of Greenhaus et al. (1990). (H3a)

SOW: Subjective Other-‐referent career success for Workers was calculated by adding up the

scores of the 5 questions about other-‐referent subjective career success (based on Heslin, 2003). (H3b)

Control variables:

(28)

Data analysis

Pretest phase: The primary goal is investigating the quality of the questionnaire. A histogram analysis of the discriminative power of the items is performed. Too positive or negatively skewed items are removed as they reduce an item’s discriminative power. Next, internal reliability is examined using Cronbach’s Alpha (α). Items were investigated in comparison to their scale with a target of α = .80 for every scale (Evers, Lucassen, Meijer, Sijtsma, 2010; Field, 2005). Items reducing α below this level were removed. The structure of the questionnaire is investigated by an exploratory factor analysis (EFA). The factor analysis is fixed on 6 dimensions as the theory proposed 6 dimensions of self-‐regulation. An oblique rotation (delta = 0) was used because the dimensions on self-‐regulation are cyclically related and are therefore probably correlated to a certain level. An oblique rotation allows for correlation between factors that probably better fits the structure as proposed by theory. Factor loadings higher than 0.4 are considered major components of the factor (Field, 2005). The subjective self and other-‐referent career success measures are investigated as well. Both the pretest and the organizational version are analyzed in this manner.

Furthermore, the effect of self-‐regulation on career success is explored. In the pretest phase, the effect of self-‐regulation is explored on salary, promotions and subjective career success for workers (H1, 2b, 3). These analyses are within group analyses. Correlations will be used to explore the relation between self-‐regulation and career success. A Oneway ANOVA will be used to investigate if career success changes across self-‐regulation total quartiles.

Organizational phase: First, the questionnaire is analyzed using the same procedure that was used for the pretest phase. Second, the relation between self-‐regulation and career success is analyzed. The within group analyses investigate the relation between self-‐ regulation and promotions and subjective career success for workers (H2, 3).

Personnel controls for high potentials: Using self-­‐regulation for selection?