Perceived Investment in Employee Development and Turnover Intentions:
Antecedents and Behavioural Outcome of PIED
Master Thesis
Submitted On: August 1st, 2014
Submitted By: Allison-Erika Gould, 10446869 MSc Business Studies, Strategy Track
University of Amsterdam, Faculty of Economics and Business
Supervisor: J.J. Ebbers
TABLE OF CONTENTS
1. INTRODUCTION ... 4
2. THEORETICAL REVIEW AND HYPOTHESIS ... 8
2.1 PERCEIVED INVESTMENT IN EMPLOYEE DEVELOPMENT (PIED) ... 8
2.2 ANTECEDENTS OF PIED ... 9
2.3TURNOVER INTENTIONS ... 12
2.4 PIED AND JOB SATISFACTION ... 14
2.5 PIED AND PCO ... 15
2.6 RESEARCH MODEL... 17
3. METHODOLOGY ... 18
3.1RESEARCH SETTING AND DATA COLLECTION ... 18
3.2OPERATIONALIZATION OF VARIABLES ... 19 3.3ANALYTICAL APPROACH ... 22 4. RESULTS ... 23 4.1DESCRIPTIVE STATISTICS ... 23 4.2 CORRELATIONS ... 23 4.3 REGRESSION ANALYSIS ... 26 5. DISCUSSION ... 31 5.1THEORETICAL IMPLICATIONS ... 31 5.2BOUNDARY CONDITIONS ... 33 5.3MANAGERIAL IMPLICATIONS ... 36
5.4LIMITATIONS AND SUGGESTIONS FOR FUTURE RESEARCH ... 37
6. CONCLUSION ... 39
REFERENCES ... 40
Abstract
Training and development opportunities are recognized as a critical human resource strategy that can lead to a competitive advantage for organizations. Yet, to attain competitive advantage, organizations must be able to gain on their investment by ensuring a readily available workforce by decreasing turnover intentions. This study proposes that perceived investment in employee development (PIED) plays a significant role in increasing employee retention and furthermore investigates this construct by examining the antecedents and behaviour outcomes of PIED. Additionally, this study purposes that job satisfaction is an important mediator in the relationship between PIED and turnover intentions and that PCO moderates this relationship. In a sample of 122 employees in the Dutch advertising industry, this study found that participation in formal development activities, leader-member exchange, and challenging job assignments were each positively related to PIED. This study also found that job satisfaction was a strong mediator in the relationship between PIED and turnover intentions. However, this study did not find PCO to moderate the relationship between PIED and turnover intentions. Furthermore, implications of this study are drawn from management and organization literature as well as for management practice. Lastly, this study contributes to literature on social exchange theory and career motivation theory by indicating how PIED is developed and when it influences turnover intentions.
Perceived Investment in Employee Development and Turnover Intentions: Antecedents and Behavioural Outcome of PIED
1. Introduction
Organizations trench through a talent war as they continue to compete in an environment in which their ability to attract and retain talented employees is far more difficult than ever before (Michaels, Handfield-Jones & Axelrod, 2001). Not only are organizations threatened by globalization, technological advancements, and demographic changes but also over the past decade the employment relationship has shifted from lifelong employment security to employees increasingly managing their own careers over multiple organizations (Direnco & Greenhaus, 2011).
One way that organizations compete for talent is by providing opportunities for skill development through on-the-job training (Doeringer & Piore, 1971, Mincer 1962) and employer-provided formal training (Bidwell & Briscoe, 2010). In today’s labour market, workers value the opportunity to develop work-related skills as it allows them to remain employable as they manage their future careers (D’Amato & Herzfeldt, 2008). Furthermore, employee development is not only vital in maintaining and developing the capabilities of the individual employee but for the organization as a whole (Lee & Bruvold, 2010) as employee development affects organizational outcomes by shaping employee behaviours and attitudes (Whitener, 2011).
However, despite the advantages employee development affords both the individual and the firm, organizations have become reluctant in providing training opportunities for they do not always trust that they will be compensated for their time, money and efforts due to the changing nature of the employee-employer relationship (O’Mahony & Bechky, 2006). This preoccupation is more than evident in the advertising industry in which talent acquisition and retention is a major problem (Benett, 2011). In a report tilted Transforming Talent Management1, Benett (2011) found that almost a third of the 3,000 agency staffers that completed the survey planned on leaving their current employer within the following 12 months. Accordingly, results of Bennett’s research show that, 90% of agency staff said that they have to figure things out on their own due to the lack of training and 50% of talent in the advertising industry felt undertrained and with no definable career path. Therefore, it can be
1
http://www.aaaa.org/careers/Documents/transforming_talent_study.pdf
deducted that employees are leaving advertising agencies for they find little opportunity for advancement and seek employers who will provide the training they need to develop their careers.
Academics seek to understand whether providing training and development opportunities will actually increase employee retention and if so what factors then cause employees to stay. A central premise in understanding human resource practises is perceived investment in employee development (PIED). When an employee forms perceptions that the organization cares about their development and employability (Cavanaugh & Noe, 1999, Lee & Bruvold, 2003) the employee in turn is expected to respond in a reciprocal manner by working hard and staying within the organization. A number of studies have suggested that consistent with social exchange theory (Blau, 1964), PIED is positively related to retention. However, only a limited number of studies have examined the relationship between developmental support and turnover intentions and moreover these studies have reported mixed findings. For example, although Allen et al., (2003) and Wayne et al., (19997) have found employee participation in development activities to lead to greater perceived organizational support which in turn leads to lower turnover intentions, Lee & Bruvold (2003) did not find a direct relationship between these constructs. Furthermore, Ito & Brotheridge, (2005) found that employees’ perception of supervisor support for development directly increased turnover intentions when coupled with employee adaptability. Therefore, as suggested by Kraimer et al., (2011), studying a possible mediator and/or moderator effect between PIED and turnover intentions would contribute to a better understanding of PIED. This study examines both a mediator and a moderator in the relationship between PIED and turnover intentions, the first being, job satisfaction and the latter being perceived career opportunities (PCO).
Target Group
The target group of this study are individuals working in advertising agencies in the Netherlands. Despite the advertising industry being a major driving force in the Western economy2 agencies are unable to retain talented employees (Benett, 2011). The lack of training in the industry has come to be understood as a major factor of employee turnover. For example, the average Starbuck’s barista in the United States receives more training than
2
In the American advertising industry champions between 60-80% of clients’ marketing expenditure with every $1 of ad spending accounting for $20 in economic output (Benett, 2011).
an advertising agency employee per year. Furthermore, 30% of the individuals surveyed stated that they planned on leaving their current employer within the next 12 months (Benett, 2011). Since the objective of this study is to further understand the relationship between organizational support for development and turnover intentions this focus group provided an interesting backdrop for more robust findings as turnover is quite high, and agency employees seek development opportunities.
Research Questions
This study addresses five research questions. First, it asks: What factors influences employees’ perceptions of an organizations investment in development? As defined by Lee and Bruvold (2003), PIED is “an employee’s perception of how supportive an organization is of skill development, employability, and career management” (p. 983). In similar research by Kraimer and colleagues (2011), the authors considered two types of antecedents to a similar construct: employees’ participation in formal development activities provided by the organization and informal developmental activities with senior managers (i.e. leader-member exchange (LMX), and career mentoring). This study will examine, participation in formal developmental activities, and LMX, which is defined as the quality of the social exchange relationship between a leader and follower (Liden & Maslyn, 1998) as antecedents to PIED. Furthermore, this study suggests an additional predictor of PIED as being that of challenging job assignments.
Next, this study further examines how PIED relates to employee turnover intentions. Kraimer et al., (2011), addressed the gap in the literature in the perceptions of organizational policies as a predictor of employee behaviour rather than actual policies themselves. This study is limited in that it will examine turnover intentions (rather than actual turnover) but will contribute to Kraimer et al.’s (2011), initial research by examining the advertising industry where turnover is expected to be high and therefore more robust findings are predicted to further validate the authors findings. Therefore this study asks: Does PIED negatively influence turnover intentions?
Subsequently, as mentioned above, unreliable results between organizational support for development and its effect on turnover has lead Kraimer et al., (2011), and other academics (Lee & Bruvold, 2003, Ito & Brotheridge, 2005) to believe that a possible mediating and/or moderating contextual variable exists. Job satisfaction is introduced as a mediator of this study and this study asks: Does job satisfaction mediate the relationship
between PIED and turnover intentions? Job satisfaction is considered a primary driver of turnover intentions (Griffith, Hom & Gaertner, 200) and thus believed to be a mediator in this relationship.
Lastly, based on career motivation theory (London, 1983) and more recently, Arthur and Rousseau’s (1996), notion of the “boundaryless” career, perceived career opportunities (PCO) is introduced as a possible moderator between PIED and turnover intentions and is defined as, “an employees’ belief that jobs or positions that match their career goals and interests exist within the organization” (Kraimer et al., 2011). Therefore the question is asked: Does PCO moderate the relationship between PIED and turnover intentions? Although the authors find PCO to significantly predict turnover further investigation into PCO as both a main effect and moderator is suggested to be valuable. As mentioned above, the authors suggest that in an organization characterized by higher variance in performance, and high turnover will account for more vigorous findings. Below the research questions are restated:
1. Does participation in formal development activities have a positive influence on PIED?
2. Will developmental work activities (i.e., leader-member exchange and challenging job assignments) also lead to a positive relation with PIED?
3. Does PIED have a negative relationship with turnover intentions?
4. Does job satisfaction mediate the relationship between PIED and turnover intentions? 5. Does PCO moderate the relationships between PIED and turnover intentions?
2. Theoretical Review and Hypothesis
2.1 Perceived Investment in Employee Development (PIED)
The opportunity for employees to participate in training and development is essential to remain employable in today’s marketplace. Training and development practises constitute as one of the most important ways to assist personnel in gaining new knowledge and skills required to comply with competitive standards (Tsai, & Tai, 2003). Moreover, selecting the appropriate human resource practices is critical to an organization’s success (Ferris et al., 1999) as employee development is acknowledged as one of the most significant human resource practices (Lee & Bruvold, 2010). Furthermore, past research has indicated that employee development represents a “high commitment” human resource strategy that has the ability to affect organizational outcomes by shaping employee behaviour and attitudes (Whitener, 2001; Arthur, 1994). By “investing” in employee development the organization is equipping employees with new skills and knowledge for personal growth (Aguinis & Kraiger, 2009), which also enables individuals to be ready for new job requirements (Rothwell & Kazanas, 1989). Investing in employee development refers to organized learning experiences provided by the employer to enhance performance (Nadler & Nadler, 1989). Examples of employee development opportunities include: job rotation, training classes and workshops, tuition reimbursement programs, career development, planning and mentoring (Kraimer et al., 2011).
Studies have found that training and development opportunities to be linked with greater organizational commitment among employees, higher job satisfaction (Maurer & Lippstreu, 2008) and increased employee motivation (Ichniowski et al., 1997; Macduffie, 1995). Furthermore, competitive advantage is believed to be derived not only through the development of employees into a more competent workforce through continuous learning but also through worthy organizational outcomes such as cost savings in reducing absenteeism and turnover rates (Kaye & Jordan Evens, 2000; Lam & White, 1998) and as well attracting people who find developmental experiences attractive (Barbeite & Maurer, 2002). Perceived investment in employee development (PIED) is defined as the employees’ assessment of their organizations’ commitment to help employees learn to identify and obtain new skills and competencies that will allow them to move to new positions either within or outside of the organization (Lee & Bruvold, 2010). PIED creates conditions where employees believe that the organization values their contribution and cares about their employability (Lee & Bruvold, 2010). It is due to this investment that it is expected that the employee will devote greater
effort towards the organization (Wayne et al., 1997). This effort leads to employees that are more motivated to stay with and work hard for organizations that show that they care about employee development (Aguinis & Kraiger, 2009).
2.2 Antecedents of PIED
As stated by Kraimer et al., (2011), “little is known about the factors that influence employees’ perception of development support” (p.485). As suggested by Aguinis and Kraiger (2009), “future research could investigate the extent to which training opportunities are seen as a message that the organization cares [for its employees], which could be a powerful and important message in today’s corporate world plagued by downsizing and employee layoffs” (p. 467). Thus it is of value to understand the antecedents and thus perceptions of development support and the extent to which these perceptions are related to employee turnover intentions.
Kraimer et al., (2011) developed two types of antecedents of organizational support for development: formal and informal learning. These categories are based on the assumption that personal experiences are the stimuli that form an employees’ perception of the organization (Ehrhart & Holcombe, 2000). According to Maurer et al., (2003), two entities represent two distinct but related sources of support for development; they are the organization itself and the supervisor. As an agent of the organization, the supervisor has the ability to demonstrate his or her enthusiasm (or lack thereof) for developing subordinates. On the other hand, the organization has the ability to provide opportunities for development and learning in a variety of ways. The antecedents of PIED are presented below.
2.2.1 Participation in Formal Development Activities
Formal development activities fall under two categories: the first being specific and the second being general. In the former, employees’ receives training explicit to the organization and therefore employee mobility is predicted to increase only within the firm (Becker, 1962). This type of training of course limits the employee in their “employability” and actually rarely occurs in practise since it is difficult to come up with quality examples of training that are only useful for one employer (Loewenstein & Speltzer, 1997). The latter type of training is “general” in which skills of the employee will increase the productivity of labour across firms equally. Of particular interest to this study is general training since employees acquire job-related skills, and work competencies (Aguinis & Kraiger, 2009; Kozlowski et al., 2001)
that can easily be transferred to another employer therefore increasing their market value and thus mobility outside of the firm (Green, Felstead, Mayhew & Pack, 2000).
Human Capital theory assumes that employees will be expected to pay for general trainings as it is theorized that the mutual commitment will weaken turnover intentions since general training is of equal value to organizations other than the training firm, personnel turnover is expected to increase (Green, et al., 1997). For example, Lynch (1991) found the employees who participated in some form of training through company funding outside of the workplace were more likely to leave their jobs. In contrast however, research by Benson, Finegold and Mohrman (2004), examined the role of tuition reimbursement programs in predicting turnover. They believed that the level of education obtained by the employer through training would be a significant predictor of turnover. However, Benson et al., (2004) found that company sponsored development programs lead to participant’s positive
perceptions of organizations support, which lead to a decrease in turnover. These findings are in line with Lee and Bruvold (2003) who found that firms’ investing in employee training contribute to the creation of positive perceptions in individuals of the organization’s
willingness to support their development. Basis for these results can be found in social exchange theory that advances that by investing in the training and development of employees, turnover intention decreases (Benson et al., 2004; Sieben, 2007).
Lastly, research has demonstrated that organizational support for development motivates employees to participate in training (e.g., Maurer et al., 2003). Based on the work of Kraimer et al., (2011), the authors theorized that an employees’ level of participation in developmental activities in the organization influences their perceptions of organizational support for development. Therefore, it is expected that employees who personally participate in development activities in the organization will find these activities more prevalent than those colleagues that do not. Furthermore, since specific activities correlate to the level of resources invested by the organization in support of the employee’s development (Tsui, Pearce, Porter, & Tripoli, 1997), employees who have participated in these activities will be more likely to form perceptions that their employer is supportive of their overall development (Allen, Shore, & Griffeth, 2003; Wayne et al., 1997). Therefore in line with the above reasoning and that of social exchange theory, the following hypothesis is developed:
H1: Participation in formal development activities has a direct and positive relation to PIED.
2.2.2 Developmental Work Exchange
As previously mentioned, two distinct types of developmental activities exist. The first being formal developmental programs and the second being informal, relational-based developmental activities (Hall, 1996; Sturges, Guest, Conway, & Davey, 2002). This section introduces the predictors in the second category, leader-member exchange, and challenging job assignments.
Leader-Member Exchange
Rooted in both role theory (Kahn, Wolfe, Quinn, Snoek & Rosenthal, 1964), and social exchange theory (Blau, 1964), leader-member exchange (LMX) refers to the quality of the social exchange relationship between supervisor and subordinate (Liden & Maslyn, 1998). The LMX construct describes the relationship between supervisor and subordinate against the backdrop of a formal organization (Graen & Cashman, 1975).
As the LMX model suggests, there is a differential relationship between supervisor and subordinates and thus the quality of relationships can range from high to low. A high LMX relationship is characterized by emotional support and trust (Dienesch & Liden, 1986) and more specifically when LMX is high, the social-exchange is characterized by a more empowering relationship which emphasizes the exchange of resources and support that extends beyond the employment contract (Wayne et al., 1997). In low LMX contexts, support in the subordinate-supervisor relationship is limited to completing the tasks and activities satisfying the contract (Wayne et al., 1997). Eisenberger et al., (2002) fond that supervisor support is strongly related with changes in employees’ perceptions of organizational support and therefore it is expected that an employee will evaluate the organizations level of support for development on the basis to the extent to which their immediate leaders provide them with development opportunities (Kraimer et al., 2011). Therefore the following hypothesis will be presented:
Challenging Job Assignments
To be challenged at work means that performance expectations are set that are reasonably high or as Berlew and Hall (1966) stated “a stretch”. When experiencing a challenge at work, individuals are faced with a non-routine task, which forces them outside of their comfort zone (McCall, Lombardo, & Morrison, 1988) and thus existing tactics and routines are inadequate driving individuals to develop new strategies and skills (Lombardo & Morrison, 1988). Furthermore, challenging job assignments have been documented as the most important type of learning within organizations (Yeo & Marquardt, 2010) as on-the-job training leads to “permanent changes in knowledge, attitudes, or skills” (Berings, Poell & Simons, 2008, p. 418).
Rooted in Signaling theory, job assignments are an underlying process through which employees may interpret as indicative of organizational support as they serve as signals to an employee’s potential and possibly an indication of future organizational support (Wayne et al., 1997). Wayne et al., (1997) studied the possible antecedents of POS and LMX. They hypothesized that a number of developmental experiences and promotions will be positively related to POS. The types of developmental experiences they focused on were early job assignments, promotions and other organizational experiences. Their results showed that employees who had participated in more formal and informal training and developmental experience than others reported high levels of POS. Since challenging job assignments is a way to signal to employees that their career goals are important but also that they are valued in the company the following hypothesis can be derived:
H3: Challenging job assignments have a direct and positive relation to PIED. 2.3 Turnover Intentions
For several reasons, retention is a critical to organizational strategy especially when it comes to those employees working in knowledge intensive organizations. To start, companies depend on the stock of knowledge within the boundaries of the firm therefore; retaining employees with the highest intellectual capital is vital in a tight labour market where finding talent is difficult (D’Amato & Herzfeldt, 2008). Next, employee turnover is linked to a number of tangible and intangible costs. Costs such as recruitment, training, quality issues and costs of temporary staff are among the measurable expenditures organizations’ suffer when loosing employees, the intangible costs are associated to consequences for
organizational culture, employee morale, social capital and knowledge in the organization (Morrell et al., 2004, Newman et al., 2011).
Turnover intentions have been recognized as the final cognitive variable having an immediate causal effect on turnover (Bedeian et al., 1991) and actual turnover is expected to increase as the intention increases (Mobley et al., 1978). Numerous studies have sought to identify factors that cause employees to quit (e.g., Griffeth, Hom & Gaertner, 2000). These factors include, job satisfaction, organizational commitment, job search, comparison of alternatives, withdrawal cognitions, and quit intentions. Furthermore in the work of Hausknecht and colleagues (2009), the authors identified factors that compelled high versus low performing employees to stay within an organization. Results indicated that high performers were more likely than low performers to report staying because of advancement opportunities, constituent attachments, job satisfaction, organizational justice, and organizational prestige.
As McConnell (1999) stressed, effective training and skill development programs have a measurable impact in reducing turnover. According to research by Lee and Bruvold, (2010), there are several reasons that PIED reduces employees’ intention to leave. First, employees engage in social comparison processes (Adams, 1965) and are highly likely to compare their situation to their peers in other organizations who may invest more or less into employee development programmes. If their peers receive less training, the employee becomes more attached to their current organization. Second, employees’ who receive training and development feel more indebted to the firm, and more or less feel the need to payback the organization (Wayne et al., 1997). This concept is also prominent in social exchange theory when organisations invest in their employees; employees tend to reciprocate in positive ways (Settoon et al., 1996; Cropanzano & Mitchell, 2005). In continuing with research in social exchange theory, the relationship that forms between employees and employer is built on obligations and trust, and one in which employees are willing to exchange work performance for additional, less tangible values, such as feelings of being valued and supported (Eisenberger et al., 2001). When employers provide training and development opportunities, employees perceive that their organization cares about their well-being and values their contribution to the firm and turnover intentions is expected to decrease. Therefore the following hypothesis can be conferred:
2.4 PIED and Job Satisfaction
Job satisfaction represents another advantage that may be related to PIED. Job satisfaction is the affective response an individual may have to specific aspects of the job (Lee & Bruvold, 2010). PIED may lead to increased job satisfaction for a number of reasons. First in line with Georgellis and Lange (2007), organizations are able to enhance the psychological contract between employees and employer by providing developmental support. In this way the employer is believed to become more emotionally attached to the employer and thus increase employee satisfaction. Second, employees will perceive that the organization in which they are employed is concerned with their growth and development and therefore individuals will feel a greater sense of control over their careers. Employees are expected to make themselves more valuable to their employers while at the same time making themselves more employable in the external labour market (Feldman, 1996) It has been shown that dissatisfaction in lack of training opportunities has a stronger effect on job satisfaction than dissatisfaction with workload and pay (Shields & Ward, 2001) and therefore it is expected that greater employability leads to higher job satisfaction. Lastly, it is expected that individuals may perceive an organization offering employees development opportunities as concerned with their long-term growth (Lee & Bruvold, 2010). A number of studies have examined the positive relationship between employee development and job satisfaction (Edgar & Geare, 2005; Georgellis & Lange, 2007). For example, Guest (2002) found that organizations providing training and skill development lead to greater job satisfaction due to benefits that training affords, such as the necessary skills, knowledge, attitudes to function autonomously and responsibly. Based on the notion that PIED represents the employees’ perception that the organization invests in their development through providing training and development activities, it is expected that PIED to be positively related to job satisfaction. Therefore the following hypothesis is proposed:
H5: PIED will have a direct and positive relation with job satisfaction.
Furthermore, it is believed that people who perceive that their firm is investing in their employability through training are more likely to experience job satisfaction and thus are less likely to leave their job (Ferris & Urban, 1984). According to social exchange theory, positive perceptions such as higher levels of job satisfaction, and a stronger willingness to work hard for the organization (Arthur 1994; Eisenberger, Armeli, Rexwinkel, Linch & Rhodes, 2001), cause employees to act in a reciprocal manner towards their employer (Wayne, Shore & Liden, 1997; Lee & Bruvold, 2003). Amongst these attitudes job
satisfaction has received considerable attention in relation to labour market behaviour, in particular intention to quit (Clark, Georgellis, & Sanfey, 1998). In a study by Shields and Ward (2001) employees dissatisfied with their job was a major predictor of their intention to quit. Furthermore, when coupled with dissatisfaction in promotions and training opportunities intention to quit was even stronger. Therefore it is believed that employees who are satisfied with their job and in which a relational exchange exist between the employee and organization will be less likely to leave their job. Hence:
H6: Job satisfaction will mediate the relationship between PIED and turnover intentions. 2.5 PIED and PCO
Perceived career opportunities (PCO) is a construct developed by Kraimer et al., (2011) to account for inconsistent findings between organizational outcomes: turnover intentions and job performance and an employee’s perceived organizational support (POS). As defined by the researchers, PCO is “an employees’ perceptions of the degree to which work assignments and job opportunities that match their career interests and goals are available within their current organizations (p.486)”. This definition is believed to account for the varying perceptions of individuals and the opportunities present to them in their current organization. According to Arthur and Rousseau (1996) and their contribution to career motivation theory, the post-industrial age is defined by a new organizational era of “opportunity, insecurity, flexibility and uncertainty” (p. 3). That being said, Kraimer et al., (2011) believe that PCO is an important contextual concept and recognizes “the new employment relationship” (Arthur & Rousseau, 1996; Mirvis & Hall, 1994) that of increased boundaryless and ever self-directed. Both employee and employer expect that careers will unfold across multiple organizations and therefore it is primarily up to the employees to manage their career success (Pearce & Randel, 2004; Roehling, Cavanaugh, Moynihan & Boswell, 2000).
According to London (1983), one of the earliest theorist of career motivation, an individuals’ work behaviours and decisions are to varying degree motivated by their desire to achieve their own career goals. An employee may therefore voluntarily leave an organization if external opportunities match their career goals. According to matching theory, workers may move across organizations as their careers develop (Bidwell & Briscoe, 2010). Either, the rewards of the job, such as money, career progression and learning provide a good match for the individual or the worker’s resources such as skills, knowledge or relationships may be a good fit to the specific demands of that job, improving job performance (Jovanovic, 1979).
The assumption of matching theory is that an individual will end up in a job for which they are best matched. Additional research on this topic that has contributed to the understanding of mobility between organizations has been on the type of factors triggering work exits from firms. Outward career mobility has often been seen as an internal problem of the firm instead of an opportunity that the nature of the new organization affords (Arthur & Rousseau, 1996). Therefore the level of developmental support may be irrelevant and/or account for negative and/or lackluster relationship between organizational support and turnover intentions. Kraimer et al., (2011), found that PCO was a worthy predictor of an employees’ external job search activity and a call to action to further investigate the PCO construct as a moderator between organizational support for development and turnover intentions has been suggested. Therefore in line with Kraimer et al., (2011), the following hypothesis will be investigated:
H7: PCO moderates the relationship between PIED and turnover intentions, such that the relationship is negative when career opportunity is high but positive when career opportunity is low.
2.6 Research Model Regression Model Mediation/Moderation Model Participation in Formal Development Activities Leader-Member Exchange Challenging Job Assignments Perceived Investment in Employee Development H1 (+) H2 (+) H3 (+) Perceived Investment in Employee Development Job Satisfaction Turnover Intentions Perceived Career Opportunities H4 (-) H5 (+) H6 (-) H7 (-)
3. Methodology
3.1 Research Setting and Data Collection
The empirical setting of this study was the Dutch advertising industry. More specifically the research sample consisted of members of the RA*W network and additional employees working in advertising firms located throughout the Netherlands. This study relied on quantitative research methods. Data was gathered using two collection methods: 1. Paper-administered surveys, and 2. Computer-Paper-administered surveys. Lastly, the survey was administered in English. Although it could have been possible to also create a Dutch version of the survey it did not seem necessary given the international nature of the advertising industry in the Netherlands. For example, many advertising agencies based in Amsterdam work with clients worldwide and therefore the respondents would most likely speak English on a daily basis. Furthermore, it would be suspected that most respondents would have completed some degree of higher education and therefore would have at least a moderate working knowledge of English. Insights into the two collection methods are presented below. Paper-Administered Survey
The first method was a paper-administered survey which was distributed February 4th, 2013 at a monthly RA*W network meeting. In this context a paper-administered survey was ideal as it ensured that the survey would be completed immediately instead of trusting that the group would complete the survey after the meeting. As well, it was believed that individuals at the meeting might feel more compelled in completing the survey due to increased group conformity. Of the 56 paper-administered surveys that were administered through the RA*W network, the response rate was 100%. However out of the 56 surveys administered, only 44 (78.6%) were suitable to use in this study. The 12 surveys that were not considered in this study were due to being incomplete or the respondent not fitting the sample (e.g., did not work in the advertising industry, owned their own agency etc.).
Computer-Administered Survey
The remaining data was collected electronically through a computer-administered survey. The collection date was between May 15th, 2013, and July 5th, 2013. Internet-mediated surveys have many advantages; they are completely anonymous, have faster response rates, are less costly and as well statistically speaking, have less missing data (Londsdale et al.,
2006). Furthermore, an advantage of an internet-mediated questionnaire for the researcher is that internet mediated questionnaires do not have to be filled into SPSS manually, which is particularly time consuming and also diminishes the risk of human error when transferring data from paper version to an electronic data, in this case SPSS. The online survey tool used to collect the online data was Qualtrics. In total 201 respondents started the survey however only 78 (38.8%) actually completed it. According to reports generated by Qualtrics 34.83% or 70 respondents quit after completing 0% of the survey. Of the remaining 131 respondents, the greatest majority (16.03% or 21 participants) quit after the preliminary questioning (7th question) and the second greatest majority (13.1% or 10 participants) quit after the 8th question. This may indicate that respondents were not interested in the content of the survey. As these questions were interesting to the research but not mandatory, these questions could have been positioned at the end of the survey instead, which may have increased the survey response rate. On average, the duration to complete a survey was 10 minutes. However, response times ranged between 10 minutes and 4 hours. Furthermore, most people started the survey around 14h. The duration and time that the survey was active may indicate that respondents were likely nearing the end of their workday, and looking for entertainment. However, it would be expected that some respondents were distracted and hence why surveys were opened for longer than the average 10 minutes. This would suggest that sending a reminder or setting up some type of safeguard would have been useful to prevent respondents from not completing the questionnaire.
3.2 Operationalization of Variables
This survey used scales measuring PIED, turnover intentions, job satisfaction, PCO, participation in formal development activities, LMX, and challenging job assignments. The items from these seven variables are presented below and a description of each scale is provided. A copy of the survey can be found in Appendix A. In order to test the reliability of each scale, the Cronbach’s alpha was applied. As indicated by current literature, a scale with a high level of internal reliability has a Cronbach’s alpha > 0.7.
Dependent Variable
Turnover intentions is the dependent variable of this study and was measured using four-items from Shore and Martin (1989) Intent to Stay scale. These four four-items are measured on a five-point scale with item 1 and 3 being reverse scored. Shore and Martin (1989), define
turnover intensions as a predictor of turnover independent of the intent to change professions or type of work performed. The coefficient alpha of the four-item scale was α = 0.893.
Independent Variables
Participation in formal development activities was measured using a scale, which was specifically developed for this survey. The scale reflected the types of formal development activities present in the Dutch advertising industry. In guidance with Kraimer et al., (2011), who developed a similar scale, preliminary research was conducted to narrow down potential activities of interest. It was determined that the following formal development activities may be present in the advertising industry in Amsterdam: 1. Training classes/workshops, 2. Job rotations, 3. Workshops/computer-based training classes to develop technical skills, and 4. Workshops/computer-based training classes to develop managerial skills. Furthermore, since this study was not created for only one advertising organization it was possible that other forms of formal development activities would be present, other than those listed therefore participants could select the option of 5. Other. Participants were asked to indicate on a scale from 1 (not at all) to 5 (a very large extent) to which “you have participated in the following career development activities while employed by your organization. Compare yourself to other colleagues in your company”. Employees who previously participated in developmental activities are more likely to have extra knowledge about these activities and see them as more prevalent in the organization than someone who has not previously participated (Kraimer et al., 2011). The coefficient alpha of the five-item scale was α = 0.716. When examining the output more closely it was determined that if “job rotation” was deleted from the scale, that the internal validity would increase to α = 0.813. Given the insights that this study concerns about the advertising industry, the deletion of job rotation from the scale did not come as a surprise since those working in the industry tend to participate in one specialization (i.e., accounts, creative etc.) and therefore employees rotating between these functions does not seem very likely. It could be possible that employees would take on different responsibilities within their role but this would have had to be indicated differently in this study. Removing any other item from the scale would decrease the validity of this measure (α<0.7) and therefore all remaining items were included as a measure of participation in formal development activities. Thus, by deleting the item “job rotation”, the value of the Cronbach’s alpha coefficient increased from α = 0.716 to α = 0.813 for the reaming four-items.
Leader-member exchange was measured using Scandura and Graen’s (1984) seven-item Leader-Member Exchange (LMX) scale. This LMX construct is the most consistently used measure of LMX. An example item is ‘my immediate supervisor recognizes my potential’. Responses were made on a seven-point Likert-scale from 1 (strongly disagree) to 7 (strongly agree). The score of the seven-items were averaged and had a coefficient alpha reliably of α=0.922.
Challenging job assignments was assessed by means of Wayne, Shore and Liden (1997) three-item Developmental Experiences scale. An example item is, “in the position that I hold at my current organization, I have often been assigned projects that have enabled me to develop and strengthen new skills”. Responses were made on a seven-point Likert-scale 1 (strongly disagree) to 7 (strongly agree). The score of the three- items were averaged α=0.878.
Perceived investment in employee development was measured using Lee and Bruvold (2010) seven-item Perceived Investment in Employee Development scale (α=0.85). The first two items from this scale were originally adapted from Tsui, Pearce, Porter, & Tripoli (1997). However, to ensure an appropriate number of items for the scale, Lee and Bruvold (2003) developed a seven-item scale focussing on statements regarding the support employees received in the domain of their personal development. An example item is “my organization provides support when employees decide to obtain ongoing training”. Responses were made on a 7-point Likert-scale 1 (strongly disagree) to 7 (strongly agree). The reliably of this scale was α=.865.
Mediator Variable
Job Satisfaction was assessed using Cammann, Fichman, Jenkins, and Klesh’s (1983) 3-item scale Global Job Satisfaction scale (α=.88). Sample items included “All in all, I am satisfied with my job”, and “In general, I don’t like my job” (reverse scored). The coefficient alpha of the three-item scale was α = 0.916.
Moderator Variable
Perceived Career Opportunities was assed using Kraimer et al.’s, (2011) PCO construct. Kraimer et al., (2011) define PCO as “an employees’ perceptions of the degree to which work assignments and job opportunities that match their career interests and goals are available within the organization relative to their own subjective career goals and interest;
these goals may or may not involve promotion and upward mobility along a vertical career track” (p. 486). Kraimer et al., (2011), developed the PCO construct to justify varying results in the relationship between organizational support for development and employee performance and turnover. In this study we will continue to test Kraimer et al.’s, (2011), hypothesis by once again using PCO as a moderator in the relationship between PIED and turnover intentions. An example item is “there are job opportunities available within my current organization that are of interest to me”. The 3-item scale is a Likert scale (1=strongly disagree to 7=strongly agree). The coefficient alpha of the three-item scale was α = 0.912. Control Variables
Several demographic control variables were considered: age, gender, and education, which have been shown to relate to turnover (Griffeth et al., 2000). Age was measured with open-ended questions in terms of years. Gender (0=male, 1=male), and educational level were each measured with categorical response options (1=below HBO, 2=HBO bachelor-level, 3= University bachelor-level, 4= Master’s level, and 5= Doctorate level or above). Lastly, the control variable paper vs electronic, (o=electronic, 1=paper) was measured to ensure that statistically different responses did not occur between the samples. Additionally, in testing a moderation effect between PIED and turnover intentions via PCO, job satisfaction was controlled for as it is expected that job dissatisfaction was considered a primary driver of turnover intentions (Griffeth, Hom, Gaertner, 2000).
3.3 Analytical Approach
The analysis was performed using SPSS 20. Means, correlations, and standard deviations were calculated for the study variables (see Table 3). Missing value analysis indicated that gender (3.2%), education level (3.2%) and lastly age (4.8%) had missing values. These values were replaced using the series-mean method. In testing the hypothesis the variables were step-wise entered in a prearranged order. First, the impact of the control variables on the dependent variable was measured in Model 1. Next, the predictor variables of interested were included in Model 2. Subsequently, if applicable, the mediation or moderation variable was added in Model 3. Lastly, in the case of a moderation effect an interaction term between the moderation variable and the independent variable was created.
4. Results
4.1 Descriptive Statistics
Table 1 presents a descriptive summary for the sample. In total 257 respondents voluntarily participated in the study. In total 122 respondents (47.5%) completed the questionnaire and therefore were considered for this study. Within the sample the respondents were 50% male (N=61) and 50% female (N=61) (M= 1.5, SD= .5)(1=male, 2=female). The age ranged between 22 and 50 years of age (M= 29.21, SD= 5.21). In regards to the education level of the respondents, the following frequencies occurred: below HBO Bachelor’s level 5.7% (7), HBO Bachelor’s level 41% (50), university bachelor’s level 8.2% (10), Master’s level 39.3% (48), and Doctorate level or above 4.1% (5).
Table 1: Descriptive of Categorical Variables
Variable Frequency Percentage Cumulative Percentage Gender
Male 61 50 50
Female 61 50 100
Educational Level
Below HBO Bachelor’s Level 7 5.7 5.7
HBO Bachelor’s Level 50 41.0 46.7
University Bachelor’s Level 12 9.8 56.6
Master’s Level 48 4.1 95.9
Doctorate or Above 5 4.1 100
4.2 Correlations
Table 2 presents the means, standard deviations, and correlations of this study. A dummy variable was created for the categorical variable education (0=under HBO level; 1= all other educational levels) and paper vs. electronic (0=paper;1=electronic). The coefficients, which are significant at the 1% and 5% levels, are marked with stars and bolding.
To test the presence of multicollinearity due to a significant correlation between the predictor variables, a collinearity diagnostics in linear regression analysis were performed. The analysis indicated that the highest tolerance-level between all predictor variables and the
dependent variable was 2.175 and the lowest tolerance level was 0.516 therefore results indicate that multicollinearity was not an issue (Tabachnick & Fidell, 2007).
Table 2: Means, Standard Deviations, Correlations and Reliabilities
Note. N=122; *p<.05. ** p<.01.
Variable M SD ∞ 1 2 3 4 5 6 7 8 9 10
1. Turnover Intentions 2.96 1.10 .89 -
2. Perceived Investment in Employee Development 3.89 1.30 .87 -.53** -
3. Job Satisfaction 5.13 1.54 .92 -.75** .58** -
4. Perceived Career Opportunities 3.71 1.76 .91 -.58** .54** .52**
5. Participation in Formal Development 2.02 .89 .81 -.37** .50** .39** -
6. Leader-Member Exchange 4.80 1.30 .92 -.53** .61** .75** .42* -
7. Challenging Job Assignments 4.80 1.47 .88 -.37** .59** .53** .36** .40** -
8. Gender (1=male, 0=female) 1.50 .50 - .03 -.03 -.115 .09 -.14 -.01 -
9. Age 12.21 5.21 - .01 .09 .01 -.20* -.09 -.01 .03 -
10. Education (0=Under HBO, 1=Other) .94 .23 - -.23** -.21* -.15 .07 -.08 -.14 -.12 -.04 -
In examining the correlation table it was determined that all predictor variables have a moderate to high correlation to turnover intentions. In particular, PIED (r=-.528; p<.01), PCO (r=-.58; p<.01) and LMX (r=-.526; p<.01). This suggests that LMX, PCO and PIED are important predictors of turnover intentions. Furthermore, participation in formal development activities, leader-member exchange and challenging job assignments were positively correlated to PIED (resp. r=.50, p= p<.01; r=.61, p<.01; r=.59, p<.01). This suggests that these variables are individual predictors of PIED. Additionally job satisfaction (r=-.075; p<.01) had a strong negative influence on turnover intentions which may suggest that individuals who are satisfied with their jobs were less likely to leave the organization. Moreover, job satisfaction had a moderate to high positive correlation with LMX (r=.746, p<.01) and a moderate positive relationship with PIED (r=.58, p<.01) and PCO (r=.516, p<.01) suggesting that individuals who perceive their organizational as investing in employee development, possibly through leader-member exchange, and have the opportunity for advancement are more satisfied at their job, and thus less likely to leave the organization.
4.3 Regression Analysis
As mentioned in Section 3, all hypotheses in the theoretical framework were tested using multiple hierarchical regressions. Results of the analysis are presented below.
Antecedents of PIED
Table 3 presents the estimated results with respect to the hypothesized relations between participation in formal development activities, LMX, challenging job assignments and PIED. In Column 1 (Model 1, Table 3) the control variables, age, gender, education, and paper vs. electronic were entered. It was believed that age, and education level might influence the perception employees have about the development opportunities their agency provided. As a result, the four controls produced R2=.056, F (4, 117)=1.723, p>.05. Gender appeared to be significant (β=-.210, p<.05) . In column 2 (Model 2, Table 3), the three predictor variables produced R2=.488, F (7, 114)=19.435, p<.01. As can be seen in Table 3, participation in formal development activities, leader-member exchange and challenging job assignments had a significant positive regression weights, indicating that employees with higher scores on these scales were expected to have higher PIED, after controlling for the other variables in this model. Therefore, H1, H2, and H3 were supported.
Table 3: Regression Analysis for Direct Relation between Antecedent Variables and PIED
Note. N=122; *p<.05 **p<.01
Turnover, PIED and the Mediating Role of Job Satisfaction
Table 4 presents the estimated outcomes with regards to the hypothesized relations between PIED and turnover intentions and the mediating effect of job satisfaction in this relation. The following relations are tested: that between PIED and turnover intentions (Column 2, H4), between PIED and job satisfaction (Column 3, H5), and the mediating effect of job
satisfaction in the relation between PIED and turnover intentions (Column 4, H6). The effects of the control variables (Column 1, Table 4) on the model produced R2=-.080, F (4,
117)=2.535, p<.05. Furthermore, the first requirement that needs to be fulfilled when testing for mediation was met when testing hypothesis 4: PIED is significantly (negatively) related to turnover intentions (β =-.495**, p<.01). Furthermore, when PIED was introduced to the model, the predictor produced R2=.495, F (5, 116)=10.463, p<.01. As can be seen in Table 4, PIED had a significant negative regression weight, indicating that employees with higher scores on this scale were expected to have lower turnover intentions, after controlling for the other variables in this model thus hypothesis 4 is supported. The second requirement that
Predictor Variable (1) Control Variables
PIED (2) Test of H1, H2, H3 PIED β Std. Error β Std. Error Control Variables Age .041 .025 .010 .018 Gender -.210* .534 -.104 .382 Education -.036 .232 .035 .166 Type -.102 .254 .045 .185 Regression Coefficients Formal Development .320** .075 Leader-member Exchange .404** .081
Challenging Job Assignments .204** .077
Model Summary
Model R2 .056 .488**
Model F 1.723 19.435**
needs to be met is that PIED has a significant positive relationship on job satisfaction (H5). Table 4 indicates that this condition was met (β =-.568**, p<.01) and therefore hypothesis 5 is supported indicating that individuals who perceived their employer as providing training and development opportunities are more satisfied in their jobs. Next, the 3rd, and 4th
requirement of mediation is that there must be a significant relationship between the mediator and the dependent variable in the presence of the independent variable, and that there must be an insignificant relationship between the independent variable and dependable variable in the presence of the mediator. Table 4 confirms that when job satisfaction is entered into the model that the predictor has a significant (negative) relationship with turnover intentions (β =-.667**, p<.01) indicating that condition 3 is met. Furthermore, Table 4 shows that PIED becomes insignificant in the presence of job satisfaction (β =-.116, p>.05) signifying that job satisfaction mediates the relationship between PIED and job satisfaction. Hypothesis 6 is supported.
Table 4: Regression Analysis for Mediation Effect of PIED on Turnover Intentions via Job Satisfaction
Note. N=122; *p<.05 **p<.01
Predictor Variable (1) Control Variables (2) Test of H4 Turnover Intentions
(3) Test of H5 Job Satisfaction
(4) Test of H6 Turnover Intentions
β Std. Error β Std. Error β Std. Error β Std. Error Control Variables Age .073 .021 .026 .169 -.038 .030 -.042 .131 Gender .044* .195 .093 .018 -.101 .277 .068 .014 Education .274 .448 .170* .397 -.051 .638 .136* .306 Type .122 .213 .071 .186 -.065 .303 .028 .143 Regression Coefficients PIED -.495** .067 .568** .091 -.116 .063 Job Satisfaction - - - - -.666** .053 Model Summary Model R2 .080* .231** .305** .285** Model F 2.535* 10.463** 12.87** 28.285** Model Adjusted R2 .048 .281 .329 .575
Moderating role of PCO
Table 5 presents the findings of H7, which examined the direct effect of PIED on turnover intentions via perceived career opportunities (PCO). In the first column the control variables (Model 1) were entered. In the second column, the centered scores for PIED and PCO were entered (Model 2). Lastly, in the third column, the interaction term of PIED X PCO was inputted (Model 3). The centered scores of the predictor variables were used to create the interaction term in order to improve the interpretation of the interaction effect. The control variables accounted for a significant amount of variance in the model R2=.587, F (5, 116), p=.000. Job satisfaction had a strong significant negative relation to turnover intentions (β= -.732, p<.01) indicating that job satisfaction is a strong predictor of turnover intentions. Furthermore, education is also a significant predictor of turnover intentions (β= -.150, p<.05) which indicates that individuals are more likely to leave the organization with higher education. These results are not surprising, as individuals with greater education will have more opportunities in the workforce. In Model 2 the centered scores of PIED and PCO produced R2=.055, F (7, 114)=29.203, p=.000. As can be seen in Table 5, PCO had a significant negative regression weight, indicating that employees who perceived that there are career opportunities in their organization are less likely to leave, after controlling for the other variables in the model. However, PIED did not contribute to the model, and job satisfaction was still strongly negatively correlated (β =-.576**, p=.000) to turnover intentions. In Model 3, the interaction term was introduced however, as seen in Table 5, the introduction of the interaction term was insignificant indicating that PCO did not moderate the relationship between PIED and turnover intentions. Furthermore, the overall model was still significant F (8, 113)=26.001, p=.000. However, hypothesis 9 was not supported.
Table 5: Regression Analysis for Direct Effect of PIED on Turnover Intentions via PCO
Note. N=122; *p<.05 **p<.01
Predictor Variable (1) (2) (3) Test H7
Turnover Intentions β SE β SE β SE Controls Variables Age .063 .014 .021 .014 -.011 .014 Gender -.045 .132 .126 -.005 .126 -.025 Education .150* .305 .177** .293 .174** .293 Type .032 .144 .007 .136 .024 .138 Job Satisfaction -.732** .044 -.576** .053 -.596** .053 Regression Coefficients PIED -.001 .065 .004 .065 PCO -.288** .047 -.266** -.048 PIED X PCO -.082 .028 Model Summary Δr2 for step .587** .055** .006 Model F 33.027 13.425** 11.856** Model adj. r2 .570 .620 .623
5. Discussion
The purpose of this study was to develop a deeper understanding of the antecedents and behavioural outcomes of the employees’ perceptions of organizational investment for development. The construct perceived investment in employee development (PIED) was introduced as a measure of employees’ perceptions of organizational support and proposed that participation in formal development activities and informal developmental work activities contributed to the formation of PIED. Furthermore, this paper was interested in whether PIED reduced turnover. Specifically, successful retention efforts were assessed by turnover intentions, which are believed to be the final cognitive process that an employee goes through before voluntarily leaving the organization. On the basis of social exchange theory, this paper predicts that PIED would decrease an employees’ intention to turnover, as employees are more likely to act in a reciprocal manner when their own needs for skill development are being met. Moreover, this paper shows that job satisfaction acts a moderator between PIED and turnover intentions. Furthermore, this study proposes that, perceived career opportunities (PCO), moderates the relationship between PIED and turnover intentions. On the basis of London’s (1983) career motivation theory, and more recent notions of the “boundaryless” career (Arthur & Rousseau, 1996), it is predicted that employees’ will act positively towards organizational support when they perceive that there are career opportunities they find desirable present in the organization. Conversely, it is expected that when career opportunities are absent, PIED will have a neutral or even negative effects on employee retention. Lastly, this paper contributes to research investigating the link between turnover and the lack of employee development in the advertising industry. In recent years, greater attention has been given to this phenomenon in which despite being a major driving force in the Western economy, advertising agencies are unable to retain talented employees. This study makes an important contribution to Benett’s study, Transforming Talent Management (2011), by investigating the Dutch advertising industry and contributing to these findings.
5.1 Theoretical Implications
Participation in Formal Development Activities, Developmental Work Activities and PIED In spite of increased attention towards organizational support for development in improving retention efforts, research has not specifically examined how employees form perceptions that the organization cares about their growth and development (Aguinis & Kraiger, 2009).
The first contribution of this study was to show a strong positive relationship between participation in formal development activities, developmental work activities and PIED. In line with findings of Kraimer et al., (2011) and the views of Aguinis & Kraiger, (2009) and other scholars (e.g., Tsui et al., 1997), results indicate that participation in formal development activities increases positive perceptions of an organizations investment in their employees (H1). Contributing to findings of Kraimer et al., (2011), a key point raised by the
results of this study is that actual participation in formal development activities leads employees to perceive organizational support as greater than those who do not participate in these activities. This suggests that employees who participate in formal trainings have increased feelings of self-worth and that organizational training and development programs send an important indication that the organization values its employees. Furthermore, this study found LMX, and challenging job assignments to be positively related to PIED (H2, H3).
This suggests that as an agent of the organization, the supervisor has the ability to demonstrate his or her enthusiasm (or lack thereof) for developing subordinates through acting both as a mentor for the employee and as well by awarding diverse projects to employees. These findings make an important contribution to the field of human resource strategy by offering insights into what activities form an employees’ perception about their employer’s investment into development activities. It is an important insight since few studies previous to this one have actually studied these antecedents. Furthermore by making this distinction and examining the statistical strength of each activity it may lead to an important distinction between these activities. It could be that certain activities have greater statistical strength in decreasing turnover and thus it could be advised that organizations seek to focus on those activities that satisfy both the needs of their employees and their own organizational goals.
PIED, Turnover Intentions and Job Satisfaction
The second contribution of this study was to show that job satisfaction was an important mediator in the relationship between PIED and turnover intentions. Despite academics best efforts to define a clear relationship between organizational support for development and turnover, their results have been mixed (Kraimer et al., 2011). However, opportunities for development are rendered as vital for organizations and employees to compete in a competitive market and increased work demands and furthermore opportunities for development are shown to be an important factor in motivating and retaining key employees.
The results of this study found that PIED was negatively related to turnover intentions (H4) indicating that PIED plays an important role in reducing intentions to turnover. In addition, the relationship between PIED and turnover intentions was significant when other variables were controlled for, establishing a direct relationship between these variables. According to social exchange theory, employees who receive organizational support for development perceive that the organization cares for them and are thus more likely to remain in the organization (e.g., Kuvaas & Dysvik, 2010). However, the direct relation between PIED and turnover intentions challenges research that has concluded mediating factors to validate this relationship (e.g. Allen et al., 2003; Lee & Bruvold, 2003) and as well research which has found a positive relationship between PIED and turnover intentions (Ito & Brotheridge, 2005). Therefore it was of significance to test a possible mediation effect via job satisfaction. Results of this study suggest that PIED increases job satisfaction in advertising agency employees’ in the Netherlands and also mediates the relationship between PIED and turnover intentions.
PIED and PCO
The next contribution of this study sought to demonstrate that perceived career opportunities (PCO) are an important factor that influences the relationship between PIED and turnover intentions. Kraimer et al., (2011), found that PCO acted as a moderator between organizational support for development and turnover intentions such that the relationship was negative when career opportunities were high and positive when career opportunity was low. However despite strong evidence in theory to support these predictions (London, 1983; London & Mone, 2006; Benson et al., 2004), this current study could not replicate the findings of Kraimer et al., (2011), and thus H9 was not supported.
5.2 Boundary Conditions
The advertising industry is a fascinating setting for examining the relationship between organizational support for development and turnover intentions as turnover is remarkably high and opportunities for training and development are considerably low. Moreover, the adverting industry is a major driving force of the Western economy3 and therefore it is of importance to investigate whether investment in employee development encourages or reduces voluntary turnover not only for the advertising industry but as well for the health of the Western economy.
3
Every $1 of advertising spending results in $20 of economic output (Benett, 2011).
An advertising agency is a creative organization in which to a greater extent, the creative ideas of the organization or the “magic” are seen as being responsible for creating the output. To a lesser extent, the “logic” or business side of an agency is often ignored. However, the inefficiencies of this side of this business are evident in the industries inability to attract and retain talented employees. Nowadays, training and development opportunities are the most important requirements employees have as they want to be able to learn new skills that can transfer to new job requirements either within or outside of the firm. However, advertising agencies are not prepared to offer such incentives. For example, those starting out in the industry are often unable to get their foot in the door as agencies demand that these individuals have years of working and/or agency experience. Agencies are not willing and/or are ill-equipped to provide training opportunities for these individuals starting their careers. Furthermore, personal development of workers currently employed within an agency is not a priority either as there are very few possibilities for training and development and career opportunities are vague.
An ulterior motive of this study was to first show that training and development opportunities are important incentives for individuals in the advertising industry but second that these opportunities lead to more productivity/efficiency/profit for the agency in the form of decreased turnover. An assumption of this study was that there are not many opportunities for advertising employees to participate in formal development activities and that informal training mechanisms typically “take” their place. This is in part due to the creative nature of the agency in which talent is believed to be the natural ability of the individual but as well the simple nature of the business places high pressured work demands on employees which leaves little time or space for systemic formal training and staff development (Grabher, 2002) therefore the type of formal development mechanisms are quite vague. Furthermore, creative-based skills in particular as developed through “learning-by-watching” (Grabher, 2002). This implies that a certain amount of experimentation and failure must be tolerated (Mumford et al., 2002), and that mentorship is an important part of the creative process. In order to test these assumptions, this study introduced leader-member exchange and challenging job assignments as indicators of the informal training process of creative-based skills and found that these two constructs were important indicators of employee development. This is an key insight as it shows that employees value the opportunity to learn through their superiors and that one way this can happen is through providing opportunities to learn through strategically placing employees on certain projects they might find challenging. Furthermore, as predicted,
