to retain the older worker
Master Thesis Author: Claire Valk Student number: 5672082 Supervisor UVA: Dr. S.T. Mol Co-reader UVA: Dr. A.E. Keegan
Supervisor PricewaterhouseCoopers: Ms. J. Molina MBA University of Amsterdam, Faculty of Economics and Business, Business Studies
Preview
This research project for my master thesis in Business Studies has been conducted from October 2007 to January 2008 during an internship with the Human Resources Management department of PricewaterhouseCoopers (PwC) in Amsterdam. It was a great opportunity to finalize my study in such a reputable international audit and assurance company, which nowadays is also one of the largest global consulting firms in Human Resources Management.
The subject of my master thesis has been established in dialogue with PwC. Research into a better employability of an ageing workforce is a relevant and actual subject for PwC in the current tight labour market. Chapter 1 is an introduction to the research, which includes a literature overview, the research questions and the conceptual model. In chapter 2 the method of the research will be described. In the subsequent chapters the results will be described and discussed. The final chapter consists of conclusions and recommendations.
Content Preview...2 Content ...3 1. Introduction ...5 1.1. Demographic trends...5 1.2. Overview...6
1.3. Human Resources Planning ...8
1.4. Research questions...11
1.5. Conceptual model and hypotheses...12
1.5.1 Job satisfaction, organizational commitment and turnover/intent to stay ...12
2. Method...14
2.1 Participants and procedure...14
2.2 Instruments ...14
2.2.1 Interview...14
2.2.2 Questionnaire ...15
2.2.2.1. Demographic variables ...15
2.2.2.2. Independent variables ...15
2.2.2.2.1. Reasons for retirement ...15
2.2.2.1.2. The HR strategies...16
2.2.2.1.3. Job satisfaction and Organizational commitment ...17
2.2.2.3. Dependent variables retirement intentions...18
2.3. Data reduction and analysis ...18
2.3.1. The discrepancy between employer and employee on the perceived implementation of HR practices ...18
2.3.2. Factor analysis ...18
3. Results...21
3.1. The employer ...21
3.1.1. Point of view of employer on ageing ...21
3.1.2. HR tools...22
3.2. The employee...23
3.2.1. What is the main reason for older employees to continue employment at their company? ...23
3.2.2. Are employees more willing to continue employment at their company when human resource strategies of their choice are being used?...25
3.2.2.1. Correlations...25
3.2.2.2. Normality and linearity ...30
3.2.2.3. Regression analysis...32
3.2.3 Is there one specific HR strategy that is perceived to be the most effective in retaining the older worker? ...34
3.2.4 Do employers and older employees differ in their view about the offered HR tools? ...35
3.3. Analysis per hypothesis ...38
3.3.1. Hypothesis 1 ...38
3.3.1.1. The discrepancy between employer and employee on the perceived implementation of HR practices will have a positive relationship with desired retirement age. ...38
3.3.1.2. The discrepancy between employer and employee on the perceived implementation of HR practices will have a negative relationship on career opportunities, autonomy, social support, work-life balance and job satisfaction...39
3.3.1.3. Career opportunities, autonomy, social support, work-life balance and job satisfaction will have a positive relationship on desired retirement age. ...40
3.3.2. Hypothesis 2 ...42
3.3.2.1. The discrepancy between employer and employee on the perceived implementation of HR practices will have a positive influence on Job stress, Routine, high Workload and Pressure from the employer. ...42
3.3.2.2. Job stress, Routine, Workload and Pressure from the employer will have a negative influence on the desired retirement age ...43
4. Discussion ...44
4.1. Limitations and Future Research ...48
References ...50
Appendices...55
Abstract
The baby boom generation, now between 52 and 62 years of age, is currently retiring from active labour participation. On the one hand they take with them unique levels of experience. On the other hand the number of younger employees, joining the workforce is too small (because of a lower fertility rate) to replace those who are leaving. (McDonald & Kippen, 2001)
Dealing with an “ageing workforce” is an important challenge for companies in an increasingly competitive environment. In that respect employers should have a sense of urgency to anticipate and act on these demographic developments. It is, however, surprising that employers do not anticipate this so called “grey wave” and do not seem to develop strategies within their human resource planning to retain these older workers (Remery, Henkens, Schippers & Ekamper, 2003).
The current research will examine the discrepancy between the need for Human Resources strategies for the ageing workforce and the lack of anticipation of the management of companies to fulfill this need. Why do organizations not (or barely) anticipate the ageing workforce, how are they planning to solve the adverse effects of early retirement, which HRM tools will they apply and what are the most effective HRM tools for retaining the older worker are questions which will be answered in this research.
Based on the results of this study the conclusion can be drawn that organizations may effectively retain their employees for 2.15 years if they focus on the HR practices of job design and compensation. Within job design particularly reducing workload pressures and job demands as well as creating new roles and challenging and meaningful tasks are the HR instruments which are most effective in retaining the older employees. Within compensation practices the instrument of improving pension plans was the most effective in retaining the older employees.
1. Introduction
1.1. Demographic trends
Many children were born in the period after the second World War (roughly between 1945-1955), which is why this generation is referred to as the baby boom generation. Because of this “baby boom” the population between 50 and 65 years of age is increasing the coming years above average in all industrialized nations, but mostly in the countries with a large baby boom generation (Armstrong-Stassen & Templer, 2004). The population of aged persons (65+) in the Netherlands has increased from a mere 6 % in 1900 to 14.3% in 2006. Figure 1 shows the population pyramids of the age structure in the Netherland in 2007, 2010 and 2040. In 2010, the number of aged persons is expected to increase to 15.3 % of the population and this is expected to escalate to 25% in 2030 (CBS, 2007). The grey pressure (the number of people aged 65 + in comparison with people aged 20–65 years), has almost doubled from 12.1% in 1900 to 23.2 % in 2006 (CBS, 2007).
Figure 1: Population pyramids of the age structure in the Netherlands
In spite of the fact that the Dutch active workforce has increased with 2.5% in the first quarter of the year 2007 compared to the same quarter of last year, the number of vacancies has doubled during the period 2004-2006. The number of vacancies which are difficult to fulfill has even tripled in that same period. (CBS, 2007)
Not just the population of aged persons is increasing in the Netherlands also the workforce is aging. The net workforce of individuals aged 50-65 has also increased from 38.5 % in 1995 to 52.6 in 2006. This is an increase of almost 37% in 12 years. (CBS, 2008)
1.2. Literature overview
The baby boom generation has started to retire from the workforce and this will continue during the next decade. When this generation retires they will take with them unique levels of knowledge and experience. At the same time the supply of younger employees who will join the workforce during the next decade is decreasing because of low fertility rates (McDonald & Kippen, 2001; Access Economics, 2001 in Jorgensen, 2005). The combination of the retirement of the baby boom generation and the lower supply of younger workers will result in a structural tightness in the labour market. A sharp competition for skilled and talented employees will be the result of these trends. Organizations have to move fast to capture and nurture intellectual capital which is often difficult to replace. Organizations must therefore develop innovative methods to retain existing older employees and to make use of their experience and skills in a better way. Dealing well with this will be a challenge for all companies. (Jorgensen, 2005)
Almost all employees in the industrialized nations retire before the age of 65 (Van Dalen & Henkens, 2002). In an article in “De Volkskrantbanen”, Van Alphen (2006) claims that in the Netherlands only 20% of the employees want to work until their retirement.
There are three major factors which have an impact on the early retirement of older workers. These include:
1) In the nineties of the twentieth century the disincentives to continue working beyond the age of 55 have been reinforced by a variety of de facto early-retirement programmes, while the incentives to leave the labour market early have risen robustly since the 1960’s (Van Dalen & Henkens, 2002). The latter include better employer benefits, higher interests rates on savings and better social security benefits (CHOO,
1999). Several studies have shown that a large majority of the older workers would like to leave the labour force at the earliest possible time (Van Dalen & Henkens, 2002). The tool of early retirement has been widely used by employers since the mid-1970’s to reduce labour costs and to replace relatively expensive older workers. In the mid-1990 formal early retirement programmes in almost all Organization for Economic Co-operation and Development (OECD) countries made it financially unattractive for employees to continue working after the age of 55 years. (Van Dalen & Henkens, 2002)
2) Older workers have to deal with age discrimination, defined as a less positive treatment of a person because of his chronological age (CCH, 1994 in; Choo, 1999). Workers are defined as older workers when big changes, such as functional (physical) capacities, take place in relevant work related functions during their working life. In general the ages from 45-50 have been used for the term older worker. (Ilmarinen, 2001). Older employees are frequently seen as less flexible and thus less capable to deal with fast changes taking place in organisations (Rhebergen & Wognum, 1997). Overall there seems to be however no significant difference between the job performance of mature and younger workers (Fortera & Prieto, 1994; Warr, 1994; in OECD, 2007) Variations in productivity can exist in different groups. For instance poor health is related to age and can reduce productivity and cause early retirement (Bound, Schenbaum, Stinebrickner, & Waidmann, 1997). This effect is becoming less important due to the increase in longevity. Another aspect that affects productivity is that older workers are generally thought to have difficulty in learning new skills. (Kubeck, Delp, Haslett & Mc Daniel, 1996; OECD, 2007)
The degree of physical decline associated with older workers is hard to measure because of the great variability among older workers. The physical changes in capacity are dependent on work and living lifestyles of the ageing employees. (Ilmarinen, 2001)
The myth of high sickness absence of older workers has been investigated by “TNO Arbeid”. TNO found that the sickness absence of older workers is not higher than the mean sickness absence of other employees in organizations. Older workers do have however a longer-lasting sickness absence. (Nauta, de Bruin & Cremer, 2004).
3) It is more difficult for older workers to find new jobs than for younger workers. When older workers do find a job, very often they have to accept a pay cut (Choo,
1999). Research has shown that training is an essential tool to increase the productivity potential of older workers (Armstrong-Stassen & Templer, 2005; Rhebergen & Wognum, 1997). On the whole older workers have the potential to remain productive up to and beyond presently standard retirement ages, provided they will be given adequate training. (OECD, 2007; Armstrong-Stassen & Templer, 2005; Choo, 1999)
As changes to pension policies would stimulate many workers to retire later or more gradually, retirement programs and social security systems in many OECD countries have been reduced or removed. (OECD, 2007) Recently the Dutch government has contributed in several ways to stimulate later retirement. The national government does not financially support early retirement and they plead in favour of employees to keep their retirement pay when they chose for a demotion. Besides these arrangements, the national government is holding back alternative withdrawal and at the same time a policy is provided to keep the employee more employable. The Dutch government wants to remove the barriers to work after 65. (Kabinetsstandpunt stimuleren langer werken, 2007)
With employees growing older and retiring early there is an apparent need to retain older employees in the workforce. Lefkovich (1992) argued that one of the main business and industrial topics in the coming decades will be to recruit and retain older workers. Employers could face a severe workforce scarcity in the future when they fall short to recognize and take proactive steps to cope with the topic of the ageing workforce. (Lefkovich, 1992)
The ageing workforce necessitates major innovations in employment practices, as well as a fundamental change in human resource planning strategies (Walker, 1998). Organizations have to make special efforts to adapt their human resource practices to the needs and preferences of older workers (Agarwal, 1998).
1.3. Human Resources Planning
Cascio (1998) defines Human Resource Planning (HRP) as; “An effort to anticipate future business and environmental demands on an organization and to meet the HR requirements dictated by these conditions” (page 154). This wide-ranging vision of human resource planning suggests some interrelated and specific actions that together cover a human resource planning system. This human resource planning system
includes according to Cascio (1998): Talent inventory – to review current resources (potential, abilities, and skills) and analyze existing use of employees; Human
resource forecast – to foresee future HR requirements (skills mix, numbers and
internal supply versus external labor supply); Action plans – to expand the pool of capable individuals by selection, training, recruitment, transfer, placement, compensation, promotion and development; Control and evaluation – to make closed-loop feedback available to the rest of the system and to supervise the level of success of human resource objectives and goals. (Cascio, 1998; Walker, 1980)
According to Cascio (1998) human resources planning is a two step process. The first step, the analysis of needs, is to anticipate on and react to the wishes of workers, to find out priorities and to assign resources where they are most effective. The second step, the planning of the programme can be carried out when the analysis of needs is completed. Activities and programmes relate to existing and anticipated topics and must be weighted in terms of costs, expected benefits and needs. Within this process there are three areas are of main concern to management (Cascio, 1998):
1) Needs forecasting: enhanced planning and control over staffing and organizational necessities.
2) Performance management: making the performance of teams, individuals, and the entire company better.
3) Career management: actions to determine, select, develop, and manage the careers of individuals in a company.
Specific HRM programmes have to be linked to the broad range of external and organizational factors. When these are linked to each other, recruitment, development, compensation, performance management, and other activities become integral parts of a dynamic process. The ultimate objective of human resource planning is the most efficient use of talent in the company. (Cascio, 1998)
Walker (1998) claims that the ageing workforce will necessitates major innovations in employment practices, as well as a fundamental change in human resource planning strategies (Walker, 1998). According to Agarwal (1998) the implementation of individualized human resource planning provides in a functional tool for successfully utilizing the older employee and the possibility to manage the remaining career of the older worker with the organization.
An examination of the ageing workforce literature shows that within the process of human resource planning the following six human resource strategies are
especially essential in the retention of older employees (Armstrong-Sassen, 2006). The first strategy is flexible working options, including flexible work programmes and days, job sharing, reduced hours, and unpaid leaves for elder care and education. The second strategy includes training and development practices targeting older workers for training to improve skills, as well as educating executives on the successful utilization of older workers. The third strategy is job design practices that offer older workers with important and challenging tasks and assignments and create new functions for older workers, such as mentoring. The fourth strategy includes
recognition and respect practices that identify the contributions of older workers and
make sure that older workers are valued with respect. The fifth strategy is
performance evaluation practices that make sure that performance appraisals are free
of age unfairness and provide older workers with valuable feedback on their performance. The last strategy compensation practices includes better benefits and increased financial compensation, such as more holiday time and extra time off. (Armstrong-Sassen, 2006; Armstrong-Stassen & Templer, 2005; Rhebergen & Wognum, 1997).
According to the literature these six human resource strategies should be effective to retain older employees. In Canada there are examples of companies who already effectively use these strategies to attract and retain the mature worker (Armstrong-Stassen, 2006). However in the Netherlands companies are not or only barely using these successful human resource practices in retaining mature employees. Remery, Henkens, Schippers and Ekamper (2003) showed in their research among more than a thousand companies that the older workers are not perceived to be an issue by employers. There is a risk that not enough will be invested in keeping the trainability and mobility of workers as they become older (OECD, 2007). Older workers have a relatively unfavorable status within organizations and being deficient in coherent and systematic attention to the situation of the older employees demonstrates that keeping older employees within the workforce is not a priority in current human resource planning. Employers do not seem to have a sense of urgency to foresee and act on the development of an ageing workforce and the tight labour force. (Remery et al., 2003)
1.4. Research questions
Ageing of the workforce (“the grey wave”) is an important development. One would expect that employers would develop strategies in their human resource planning to retain older employees, in particular in a tight labor market and with a scarcity of skilled young starting employees. It is surprising that, according to the literature (Remery et al., 2003) it seems that employers do not seem to anticipate this “grey wave”. In addition the literature treats this subject mainly from the viewpoint of the employer while, the view of the older employee seems to be neglected. The research for this study includes both employers and employees. This makes it unique and creates the opportunity to achieve different and meaningful results.
The current research is divided in two parts. In the first part the following questions will be answered:
1. To what extent do companies take the ageing workforce into account in their human resource strategies?
1.1. Which of the six human resource tools are used by the companies to retain the older workers?
1.2. Do the employers think that their human resource tools to retain the older worker are effective?
1.3. If companies do not take human resource strategies into account, how do they manage the ageing workforce in a tight labor market?
In the second part of the research the following questions will be answered: 2. Do employees find their employers effective in retaining them?
2.1. What is the main reason for older employees to continue employment at the organization?
2.2. Are employees more willing to stay longer employed in their company when the 6 human resource strategies are being used?
2.3. Is there one specific human resource strategy that is perceived to be the most effective in retaining the older worker?
1.5. Conceptual model and hypotheses
1.5.1 Job satisfaction, organizational commitment and turnover/intent to stay
Job satisfaction and organizational commitment have been widely studied over the last four decades of organizational research (Currivan, 1999). In this study job
satisfaction is seen as the level of positive emotions an employee has towards his job
and the role he’s playing (Kalleberg, 1977). The intrinsic elements social support, job stress, job involvement and career opportunities can be used to generate an overall measure of job satisfaction (Rayton, 2006).
Organizational commitment is defined as the degree to which an employee feels loyal
towards the organization (Rayton (2006). It is seen as an affective form of commitment, comparable to Allen and Meyer’s dimension of affective commitment (1990).
Turnover or intent to stay is seen as an outcome of job satisfaction and organizational
commitment. In research voluntary separations or quits are used to measure turnover, however this form of turnover is hard to collect. Researchers often use intent to stay instead of turnover. Studies have shown that intent to stay and turnover are negatively related (Iverson, 1992 in Currivan, 1999).
Empirical studies report a direct link between higher commitment and (to a lesser degree) job satisfaction to a higher intent to stay. (Lincoln & Kalleberg, 1996; Hom & Griffeth, 1995; Mueller, Boyer, Price & Iverson in Currivan, 1999). Given that link, satisfied and committed employees will have negative intentions about early retirement and are willing/want to stay longer in their organization.
Hypothesis 1: The discrepancy between the employer and employee on the perceived implementation of the HR Practices (flexible working, training and development, job design, recognition and respect, performance evaluation and compensation) will have a negative influence on organizational commitment, job satisfaction, career opportunities, autonomy, social support and work-life balance. Job satisfaction, organizational commitment, career opportunities, autonomy, social support, work-life balance will have a negative influence on the desired retirement age.
Hypothesis 2: The discrepancy between the employer and employee on the perceived implementation of the HR Practices (flexible working, training and development, job design, recognition and respect, performance evaluation and compensation) will result in Job stress, Routine, high Workload and Pressure from the employer. Job stress, routine, workload and pressure from employer will have a positive influence on the desired retirement age.
Figure 2: Conceptual model.
Organizational commitment Job satisfaction Career opportunities Autonomy Social support -Support from managers
-Support from co-workers
Work-life balance
Job involvement (+) Desired
retirement age - Role conflict - Role ambiguity -Workload - Routine Control variables - Retirement due to circumstances - Pursue own interests
-+ The discrepancy between the employer and employee on the perceived implementation of the HR Practices -Flexible working - Training and development - Job design - Recognition and respect -Performance evaluation
-Compensation - Job stress- Pressure from employer
2. Method
2.1 Participants and procedure
Date to answer to the research questions of chapter 1.3 have been obtained both from the interviews with HR directors (or their representatives) of 12 different organizations and from questionnaires filled in by representative samples of 850 employees of 3 organizations (a transport company, public government body and an electricity company). Of these 850 employees 320 filled in the questionnaire leading to a response rate of 37.6%. The mean age of the respondents is 54.46 years (SD =
3.14). 82.8% of the sample was male. Of the respondents 9.7% were employed in
management positions, 28.3% in middle management positions and 61.9% had operational positions. The questionnaire has been sent to relevant employees from three organizations in a random selection which had been made of all the employees of the organizations who were of 50 years or older. The questionnaire was either linked to an e-mail message or (when the employee had no e-mail address) sent as hardcopy.
2.2 Instruments
The current research consisted out of two phases for which the following two instruments were developed:
2.2.1 Interview
The first phase of the research consisted of the interviews held with 12 HR directors or the representatives of the participating organizations. The interviews consisted of both open and closed questions. The open questions were concentrated on the viewpoint of the HR directors or the representatives on the ageing workforce. The closed questions were related to the six human resource strategies which were mentioned in the chapter “overview” as important for the retention of older workers (flexible working options, training and development practices, job design practices, recognition and respect practices, performance evaluation practices and compensation practices (Armstrong-Stassen, 2006). The HR directors or representatives were asked if and to which extent they use the mentioned HR instruments and if they also use other HR tools to better manage the ageing workforce. The latter was an open question.
2.2.2 Questionnaire
The second step of the current research was a Dutch language questionnaire. The questionnaire contained scales of the independent variables aimed at assessing demographic variables, reasons of retirement (job stress, own interest, pressure from employer and circumstances), HR strategies (flexible working options, training and development practices, job design practices, recognition and respect practices, performance evaluation practices and compensation practices) and job satisfaction, pay satisfaction, organizational commitment, support, routine, workload, autonomy and role ambiguity. In the following paragraph the scales have been described. The reliability from the scales in this study are given in table 3. The reliabilities in the following paragraphs are from previous research. The reliabilities of the scales in the present study are all above α = .70, except for career opportunities (α = .62).
2.2.2.1. Demographic variables
The first part of the questionnaire consisted of four general questions (age, gender, year of service and job position). On job position there were three options; management, middle management or operational.
2.2.2.2. Independent variables 2.2.2.2.1. Reasons for retirement
The second part of the questionnaire consisted of 15 questions about possible reasons for retirement, obtained from “The Retirement Satisfaction Inventory” (Floyd, Haynes, Doll, Winemiller, Lemsky, Burgy, Werle & Heilman, 1992) and chosen for their relatively good test-re test reliability. The ratings to these items were provided on a six point scale ranging from 0 (not suitable for me) to 5 (very important).
Job stress scale (3 items, α = .80, test-retest correlation = .75). Examples of items on
this scale are “Experiencing too much stress at work” and “Difficulty with physical demands of job”.
Pressure from employer scale (5 items, α = .79, test-retest correlation = .77).
Examples of items on this scale are “Pressured by my employer to leave” and “Laid off, fired, or hours reduced”.
Pursue Own Interests (4 items, α = .66, test-retest correlation = .74). Examples of
my family”. The reliability of this variable in current research falls slightly below Nunnally and Bernstein’s (1994) α = .70 (α = .65). When removing item four from the variable the alpha becomes .71.
Retirement due to Circumstances (4 items, α = .61, test-retest correlation = .65).
Examples of items on this scale are “My spouse is in pour health” and “My own health”. The reliability of this variables is in current research lower than .70 (α = .65). When removing item four from this variable the alphas becomes .70.
2.2.2.1.2. The HR strategies
The third part of the questionnaire consisted of questions about the six human resource strategies (Armstrong-Stassen, 2006). The ratings on these items were provided on a five point scale ranging from 1 (totally not) to 5 (totally).
Flexible working options (4 items) including adaptable work days and programmes,
reduced hours, job sharing and unpaid leaves for education and elder or parental care. An example is “Flexible work schedules”.
Training and development practices (5 items) include targeting older employees for
training to upgrade skills, as well as educating executives on the successful utilization of older workers. An example is “Training to update scales”.
Job design practices (5 items) offer older workers with challenging and important
tasks and assignments and create new functions for older workers, such as mentoring. An example is “Challenging and meaningful tasks”.
Recognition and respect practices (5 items) identify the contributions of mature
employees and make sure that older workers are treated with respect. An example is “Recognize role mature employees can play”.
Performance evaluation practices (2 items) that make sure that performance
appraisals are free of age unfairness and provide older workers with useful feedback on their performance. An example is “Provide useful feedback”.
Compensation practices (4 items) including increased financial compensation and
better benefits, such as more holiday time and extra time off. An example is “incentives to continue employment”.
2.2.2.1.3. Job satisfaction and Organizational commitment
The last part of the questionnaire consists of items which were obtained from Rayton (2006).
Job satisfaction (9 items, α = .70) is defined as “the level of positive emotions an
employee has towards his job and the role he’s playing” (Currivan, 2006). An example of an item is “How satisfied are you with the sense of achievement you get from your work?”. Job satisfaction ratings were provided on a five point scale from 1 (very dissatisfied) to 5 (very satisfied)
Organizational commitment (3 items, α = .75) is defined as “the degree to which an
employee feels loyalty towards the organization” (Currivan, 2006). An example of an item is “I feel proud to tell people who I work for”. Organizational commitment ratings were provided on a five point scale from 1 (totally disagree) to 5 (totally
agree).
Autonomy (5 items, α = .75) is defined by Currivan (2006) as “the degree to which an
employee exercises discretion over the performance of tasks”. An example of autonomy is “Generally, how much influence do you have over how you do your job?”. Autonomy ratings are provided on a five point scale ranging from 1 (very few) to 5 (a lot) or 1) never to 5) very often.
Support from managers (7 items, α = .86) is defined as “the degree of consideration
individuals receive from others in their social work” (Currivan, 2006). An example of support from managers is “How good would you say mangers here are at keeping everyone up to date about proposed changes?”.
Career opportunities, routine, workload and role ambiguity ratings are provided on a three point scale (no(1), yes(2) and don’t know(3)). Career opportunities (1 item). An example of career opportunities is “Do you feel there are opportunities for you here for a promotion”.
Routine (1 item) is defined as “the degree to which job tasks are repetitive” (Currivan,
2006). Workload (1 item) is measured as the amount of overtime hours. Workload is recoded in no (1), don’t know (1) and yes (2). Role ambiguity (1 item) is measured as the amount of the awareness in achieving the business objectives of the organization. Role ambiguity, career opportunities and routine are recoded in 1 (no) 2 (don’t know) and 3 (yes).
Support from colleagues (1 item) is defined as “the degree of consideration expressed
item of support from colleagues is “How good do you feel the level of cooperation is with employees in your work group/work team?”.
Work-life balance (1 item) is defined as the assist of the employer in achieving a
work-life balance. Support from managers, support from colleagues and work-life balance are measured on a five point scale ranging from 1 (very bad) to 5 (very good).
2.2.2.3. Dependent variables retirement intentions
The desired retirement age is defined as the desired retirement age of the employee. The desired retirement age with the employment of HR instruments is defined as the
desired retirement age of the employee when the HR instruments of their choice should be implemented. The variable extension in years of service due to
implementation of HR practices is calculated by subtracting the preferred retirement
age with the implementation of HR instruments from preferred retirement age.
2.3 Data reduction and analysis
2.3.1. The discrepancy between employer and employee on the perceived implementation of HR practices
The variable the discrepancy between employer and employee on the perceived implementation of HR practices is the mean difference between the ratings of the HR practices of the employer and the employee. The ratings of the employees on the HR instruments are subtracted from the ratings on the HR instruments of the employers. The measurement of the employer is seen as the real employment of the HR instrument in the organization. The scores of the employees are perceived scores. The subtracted scores are added together to one score, which is the discrepancy between employer and employee on the perceived implementation of HR practices.
2.3.2. Factor analysis
To examine the underlying structure, the items for job satisfaction, support from managers, support from colleagues, autonomy, organizational commitment and work life balance were factor analyzed. With the aid of the parallel analysis of Hayton, Allen and Scarpello (1994) four factors were retained. Parallel analysis was used because it is one of the most reliable and accurate ways to determine the amount of factors. A parallel analysis is conducted in four steps. The first step concerns generating random
data and the second step involves extracting eigenvalues from the random data correlation matrix. These two first steps were repeated 50 times which resulted in 50 sets of eigenvalues. The third step involves averaging the eigenvalues and finally the fourth step concerns comparing the real data with the parallel random data. (Hayton, Allen & Scarpello (1994) The four factors, which were determined by the parallel analysis, were labeled are autonomy, job satisfaction, support from managers and organizational commitment. Table 1 shows the factor loadings and the composition of the subscales. Support from managers 5, Autonomy 3 and 6, Job satisfaction 1, 2, 3 and 7 were extracted from the subscales. Work-life balance was assigned to the third subscale, because this item loaded on factor 3.
Table 1: Principle components analysis; rotated component matrix Component Support from managers Autonomy Job satisfaction Organizational commitment
How good do you feel the level of cooperation is with
managers outside your work group or work team? .814 How good would you say managers here are at keeping
everyone up to date about proposed changes? .787 How good would you say managers here are at dealing
with problems at the workplace .777
Overall, how good do you feel <company name> is at
sharing and exchanging knowledge and experience? .713 How good would you say managers here are at providing
everyone with a chance to comment on propose changes?
.712 .332
How good do you feel the level of cooperation is with line managers responsible for your work group or work team?
.673
How good would you say managers here are at treating
employees fairly? .656
How often are you asked by managers for your views on
changes to work practices? .747
Generally, how much influence do you have over how
you do your job? .731
How often are you asked by managers for your views on
future plans for the workplace? .730
How often are you asked by managers for your views on
pay issues? .681
How often are you asked by managers for your views on
staffing issues .403 .673
How often are you asked by managers for your views on
health and safety at work? .427 .621
Overall, how satisfied do you feel with the reward and
recognition you receive for your performance? .560 .50 Overall how satisfied are you with your current career
opportunities? .678
How satisfied do you feel with your pay compare with
the pay of other people that work here? .480 .668
How satisfied are you with the sense of achievement you
get from your work? .332 .661
How satisfied do you feel with your pay? .642
How satisfied do you feel with the benefits you receive
other than pay? .548
How satisfied are you with the amount of influence you
have over your job? .422 .520
Overall how satisfied are you with the influence you
have in company decisions that affect your job or work? .502 My company assists me in achieving a balance between
home life and work? .342
Support from colleagues
I feel loyal to <company name> .844
I will do something extra for my company .819
I share the values of <company name> .776
I feel proud to tell people who I work for. .762
How satisfied are you with your current job? .470 .470
3. Results
3.1. The employer
The research questions have been answered based on the interviews which were held with the HR directors or their representatives.
3.1.1. Point of view of employer on ageing
The HR directors in the questioned organizations have different views about the ageing workforce and handle the ageing workforce differently. However all interviewees agree that the ageing workforce is currently a hot and important topic. A representative of a Dutch employers’ federation quoted: “The necessity to do something about the ageing workforce can be seen as a soiled pipe. When you leave it now, it will smell within ten years. “In all but one of the participating organizations, it was expected that the mean employee age would rise within the next five years”.
Views expressed were:
Some of the questioned organizations do not see the ageing workforce as a threat. They struggle with a surplus and want to reduce the number of older employees. They are of the opinion that they will not have a workforce problem in the future. Organizations where older employees are already leaving earlier than desired, feel a greater urgency to do something about it.
None of the questioned organizations have a specific policy for the older employees as they believe that a specific policy for the older employee is discriminative. Organizations rather speak of a stage of life policy or an age concerning policy.
The issue of the baby boom, although often dealt with in the course of business still does not have enough urgency at the top of organizations. Some organizations have taken steps in their policies and collective agreements but these steps are not implemented or visible in the whole organization. A representative of the HR department quoted: “There’s still not enough urgency, tomorrow’s result is seen as more important”. According to a representative of the HR department of a Dutch bank, banks are “playing a game” with the unions. The representative called this game window-dressing. Much is written about it, but the top of the organization is hesitant.
The central HR departments of the questioned organizations regard the topic of employability of the ageing workforce as very important to the development and implantation of strategies for dealing with the ageing workforce. Unfortunately many HR departments (such as HR departments of factories) are decentralized in most organizations which focus on everyday problems. Because of this long-term planning on age related issues often disappears to the background.
3.1.2. HR tools
All of the HR directors and representatives agree that the keyword in thinking about the ageing workforce is employability. Organizations have to do their utmost to keep all their employees employable. In this respect, vitality is perceived to be very important. Organizations invest in vitality and look at the lifestyle and well being of employees. According to the questioned organizations they successfully use and have to use workshops, internships, posting, internal and external mobility, part-time work, performance interviews and medical checkup as tools to keep their employees employable and vital.
The employee has to carry the responsibility about their career and not the employer. The employer has to facilitate the wishes of the employee within stated policies. At this moment it is still the employer who takes care of the career of the employee, but this has to change according to the questioned organizations.
In collective labour agreements in the Netherlands clauses are included specifically for older workers, such as no obliged overtime and night shifts, free days and part-time work. According to three of the HR directors these clauses are not necessary. Elder workers are not necessarily less employable and are often in such a good condition that they can fulfill nightshifts. According to the questioned organizations both the mindset and culture of organizations have to be changed. Elder workers are not pitiful and staying in the workforce is normal and not by definition bad for the employer.
3.2. The employee
The research questions will be answered based on the collected data. The research questions will be considered individually.
3.2.1. What is the main reason for older employees to continue employment at their company?
To answer this question the quartiles of the demographic variable age were calculated. Differences between the first and fourth quartile should demonstrate that older workers have different needs than younger workers. The mean measurements of the 6 HR instruments in the first quartile (<52) and the fourth quartile (>56) are compared in an independent t-test. The Levene's Test for Equality of Variances shows that equal variances can be assumed. The figure below shows that on average it seems likely that employees in the first quartile score higher on all the HR instruments when compared to the fourth quartile, these differences are however not significant for the 6 HR instruments (see Table 2).
0 0,5 1 1,5 2 2,5 3 3,5 Impor tance of HR instru ments (Mean score) 1 2 3 4 5 6 HR instrument Series1 Series2
Figure 3: Mean ratings on the importance of the HR instruments
Note. Series 1 = first quartile, series 2 = fourth quartile
Note. HR instrument 1 = flexible working options, HR instrument
2= Training and development practices, HR instrument 3=Job design practices, HR instrument 4= Recognition and respect practices, HR instrument
5= Performance evaluation practices and HR instrument 6 = Compensation practices.
Table 2: Independent t-test of the mean differences on the importance of the HR instruments between employees in the first quartile and the employees in the fourth quartile
Mean first quartile Mean fourth quartile Mean
difference t Sig (2-tailed) 95% Confidenceinterval of the Difference Lower Upper Flexible working options 2.62 2.48 0.14 1.01 .31 -0.14 0.42 Training and development practices 2.47 2.34 0.13 0.83 .41 -0.18 0.44 Job design practices 2.72 2.58 0.13 0.91 .36 -0.15 0.42 Recognition and respect practices 3.25 3.13 0.13 0.66 .51 -0.25 0.50 Performance evaluation practices 2.89 2.84 0.05 0.28 .78 -0.30 0.41 Compensation practices 3.42 3.21 0.21 1.16 .25 -0.15 0.57
Note. Equal variances assumed Note. N=320
3.2.2. Are employees more willing to continue employment at their company when human resource strategies of their choice are being used?
3.2.2.1. Correlations
Table 3 shows the correlations between the dependent variable extension in years of service due to implementation of HR practices and the independent variables importance of flexible working options (r = .07, n.s.), training and development (r =
.19, p < .01), job design (r = .29, p < .01), recognition and respect (r = .21, p < .01),
performance evaluation (r = .17, p < .01) and compensation (r = .27, p < .01). The independent variables are also significantly intercorrelated. Highly correlated independent variables can indicate multicollinearity. According to Pallant (2001) multicollinearity exists when the correlations between the independent variables are .90 or above. To check if there is really a matter of multicollinearity the collinearity statistics tolerance and VIF are also computed. Table 4 shows the collinearity statistics. The Variance Inflation Factor (VIF) measures the impact of collinearity among the variables in a regression model. When the tolerance of a variable is near one this indicates independence, when the tolerance value is close to zero, the variable is multicollinear. According to (Jeeshim and KUCC, 2002) there is not a formal criterion for determining the critical tolerance value or VIF. A tolerance value below .1 or a VIF that is greater than 10 roughly points towards significant multicollinearity. The independent variables job design practices and recognition and respect practices have a VIF value of 5.90 and 3.90 and a tolerance value of .17 and .21 which is not greater than 10 and less than .10. It is still possible that these variables are multicollinear because of the high correlations between the independent variables. When running the regression analysis it is important to account for this possible multicollinearity.
Table 3: Means Standard deviations, internal consistency and scale intercorrelations Mean SD 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 1. Age 54.46 3.14 - -.03 .16** .01 -.19** -.11* -.15** -.20** .07 .04 .05 .15** .12* .03 -.04 2. Gender 1.17 0.38 - -.37** .06 .03 .16** -.03 .02 .02 .04 .03 .04 .01 .00 .08 3. Years of service 24.46 10.78 - -.07 .02 .04 .12* .05 .00 -.20** -.21** -.14* -.14* -.19* -.09 4. Job Position 2.52 0.67 - .04 .03 .05 .16** .10* .05 .02 -.01 .04 .13* -.06 5. Job stress 2.98 1.02 (.74) .48** .33** .27** .05 .02 -.05 -.03 .04 .00 .06
6. Pressure from employer 3.28 0.93 (.80) .33** .40** -.03 -.10* -.10* -.10* -.07 -.04 -.01
7. Own interest 3.71 0.85 (.71) .44** .05 -.09 -.06 -.04 -.05 .01 -.08
8. Circumstances 4.08 0.74 (.70) .04 -.04 -.06 -.06 -.02 -.01 -.05
9. Flexible working options 2.52 0.91 (.76) .43** .47** .33** .38** .34** .10*
10. Training and development practices
2.64 0.93 (.86) .66** .54** .63** .47** .14**
11. Job design practices 2.18 1.00 (.91) .64** .70** .68** .10*
12. Recognition and respect practices 3.06 1.07 (.95) .77** .60** .08 13. Performance evaluation practices 2.59 1.10 (.88) .62** .02 14. Compensation practices 1.87 1.01 (.93) .11*
15. Importance flexible working
options 2.54 1.03 (.76)
Note. Reliabilities are reported in the diagonal. * Correlation is significant at the .05 level (1-tailed). ** Correlation is significant at the .01 level (1-tailed).
Table 3: Means Standard deviations, internal consistency and scale intercorrelations (part 2) 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 1. Age -.03 -.03 -.01 .01 -.04 -.04 -.01 -.03 -.15* .12* .12* .11* .14* .03 .01 .30** .19** -.11* 2. Gender -.04 -.09 -.04 -.01 -.08 .03 .09 .01 .06 .06 .00 .08 .13* .04 -.10* .05 .04 -.01 3. Years of service -.17** -.10* -.12* -.13* -.08 -.04 .04 .01 -.21** .05 -.06 -.07 -.04 -.06 .32** -.07 -.12* -.05 4. Job Position .04 -.04 .03 .06 .08 -.12* -.19** -.22** -.12* -.04 -.48** -.16** -.13* -.19** -.03 -.01 -.10* -.13* 5. Job stress¹ .02 .07 .01 -.03 .03 -.03 .01 -.01 -.05 -.14* -.13* -.05 -,13* -.05 .02 -.17** -.10* .03 6. Pressure from employer¹ -.03 .07 .01 -.01 .08 .00 .01 .03 -.17** -.05 -.10* -.05 -,09 -.09 .08 -.20** -.09 .14* 7. Own interest¹ -.10 -.02 -.04 -.08 .00 -.14* -.02 .02 -.10* -.04 -.23 -.08 -.14* -.10* .05 -.31 -.24** .06 8. Circumstances¹ -.01 .02 .01 -.01 .12* -.07 -.08 -.02 -.06 -.05 -.17** -.05 -.02 -.06 -.03 -.19** -.12* .09 9. Flexible working options .14* .04 .04 .08 .01 .04 -.03 -.13* -.01 .04 .04 .20 .04 .08 -.43** -.04 -.10* -.13* 10. Training and development
practices
.18** .06 .00 .14* .04 .26** .10* -.05 .17** -.02 .24** .39** .22** .19** -.64** .01 -.03 -.04
11. Job design practices .19** .06 .09 .19** .04 .30** .08 -.10* .16** .06 .26** .39** .25** .20** -.74** .11* .04 -.11* 12. Recognition and respect
practices .08 -.02 .02 .09 -.03 .36** .19** -.11* .24** .17** .37** .53** .33** .31** -.69** .12* .00 -.14* 13. Performance evaluation practices .10* -.01 -.02 .10* -.02 .36** .14* -.15** .11* .07 .31** .49** .26** .25** -.72** .11* -.06 -.22** 14. Compensation practices .20** .15** .19** .26** .19** .11* .06 -.08 .06 .00 .07 .20** .11* .16** -.69** .11* .00 -.16** 15. Importance flexible working options .56** .57** .50** 0.46** .45** .04 -.04 .09 .10* -.09 .07 .06 -.01 .10* -.08 .15** .18** .07 Note. Reliabilities are reported in the diagonal.
* Correlation is significant at the .05 level (1-tailed). ** Correlation is significant at the .01 level (1-tailed). ¹ Retirement due to
Table 3: Means Standard deviations, internal consistency and scale intercorrelations (part 3)
Mean SD 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33.
16. Importance training and
development practices 2.44 1.13 (.94) .79** .71** .72** .60** -.05 .00 .09 .12* -.04 .05 -.05 -.15** .13** -.16** .14* .27** .19** 17. Importance job design practices 2.69 1.05 (.94) .78** .75** .70** -.10* .05 .13* .09 -.07 -.01 -.08 -.22** .06 -.05 .10* .30** .29** 18. Importance recognition and
respect practices 3.21 1.33 (.86) .88** .76** -.11* -.07 .16** .02 -.03 -.05 -.14* -.23** .06 -.06 .14* .28** .21** 19. Importance performance evaluation practices 2.87 1.28 (.97) .68** -.03 -.02 .09 .06 -.03 .00 -.06 -.16** .10** -.15** .16** .27** .17** 20. Importance compensation practices 3.34 1.27 (.92) -.10* -.02 .12* -.01 -.13* -.10* -.14** -.26** .05 -.04 .10* .28** .27** 21. Routine 1.90 0.88 - .16** -.03 .25** .11* .43** .51** .45** .29** -.26** .17** .06 -.10* 22. Role ambiguity 2.82 0.50 - .05 .16** .09 .22** .25** .15** .28** -.11* .07 .04 -.03 23. Workload 1.25 0.49 - .04 .07 .12* -.08 -.05 .14* .09 .02 .12* .18** 24. Career opportunities 1.25 0.36 (.63) .03 .32** .20** .26** .21** -.16** .07 .06 -.04
25. Support from colleagues 4.00 0.65 - .09 .21** .29** .17** -.01 -.03 -.01 .06
26. Autonomy 2.39 0.86 (.83) .52** .44** .37** -.33** .15** .11* 0.03
27. Support from managers 2.88 0.72 (.90) .51** .35** -.41** .15** .07 -.11*
28. Job satisfaction 3.46 0.60 (.71) .32** -.16** .11* .03 -.08
29. Organizational commitment 3.95 0.66 (.86) -.24** .17** .22** .05
30. The discrepancy between employer and employee on the perceived implementation of HR practices
8.55 5.72 - .,08 .01 .13
31. Desired retirement age 59.65 3.16 - .73** -.27**
32. Desired retirement age with the employment of HR practices
61.81 3.46 - .47**
33. Extension in years of service 2.15 2.47
-Note. Reliabilities are reported in the diagonal * Correlation is significant at the .05 level (1-tailed). ** Correlation is significant at the .01 level (1-tailed).
Table 4: Collinearity Statistics of the demographic variables and the independent variables (flexible working options, training and development practices, job design practices, recognition and respect practices, performance evaluation practices and compensation practices)
Independent Variables Collinearity Statistics Tolerance VIF (Constant) .96 1.04 Age .81 1.23 Gender .80 1.25 Years of service .95 1.06 Job Position .62 1.63 Flexible working options .31 3.23 Training and development practices .25 4.09
Job design practices .17 5.90
Recognition and respect practices .21 4.82 Performance evaluation practices .38 2.61 Compensation practices .96 1.04
3.2.2.2. Normality and linearity
To answer this research question first an examination of the relationship between the predicted values of the dependent variable extension in years of service and the residuals in the regression model with the four demographic variables and the six HR instruments as independent variables was made. It is necessary to check if the residuals are normally distributed with a variance which is constant for the different predictable values on the dependent variable extension in years of service. Besides the examination of residuals deviations from linearity were also examined. Figure 4 shows the predicted values on retirement intentions (X-axis) and the standardized residuals (Y-axis), which remain in the regression model. In this model, the demographic variables as well as the HR instruments are included. Figure 4A shows the frequency partition of the residuals, which can be seen as that part of the scores on retirement intention that can not be predicted on the basis of the 10 independent variables (age, gender, years of service, job position, flexible working, training and development, job design, recognition and respect, performance evaluation and compensation). Figure 4B shows a P-P Plot. When deviations of normality exist the diagonally pulled curve deviates from the straight line. The assumption of normality is controlled by inspecting figure 4B. It seems there is no reason for concern. Figure 4C shows the relationships between the independent variables and the dependent variable. The relationships between the independent variables and the dependent variable is roughly linear.
Regression Standardized Residual 7.5 5.0 2.5 0.0 -2.5 F re q u e n c y 60 50 40 30 20 10 0 Mean =8.97E-16 Std. Dev. =0.98 N =249
Observed Cum Prob
1.0 0.8 0.6 0.4 0.2 0.0 E xp ec te d C u m P ro b 1.0 0.8 0.6 0.4 0.2 0.0
Normal P-P Plot of Regression Standardized Residual
A B
Regression Standardized Predicted Value
2 0 -2 -4 Re gr es si on St an da rd ize d Re si du al 7.5 5.0 2.5 0.0 -2.5 C
Figure 4: relationship between the predicted values of the dependent variable extension in years of service and the residuals in the regression model with the four demographic variables and the six HR instruments as independent variables
3.2.2.3. Regression analysis
Table 5 shows the main outcomes of the regression analysis. The table consist shows both the β-weights as the relative weights. The regression model with only the demographic variables accounts for 2.0% of the variance in the dependent variable extension in years of service due to implementation of the HR practice. Of the demographic variables predominantly job position (β = -2.23, p < 0.05) contributes to the most predicted variance in the extension in years of service due to implementation of the HR practices. Adding the 6 HR practices leads to a significant 12.0 % extra explained variance in the extension in years of service.
Especially flexible working options (β = 2.21, p < 0.05), job design practices (β =
3.05, p < 0.01) and compensation practices (β = 2.47, p < 0.05) seem to contribute to
the predicted variance in the dependent variable extension in years of service due to implementation of the HR practices.
Table 5: Results of Regression analysis, Relative weight analysis and Hierarchical regression analysis of the predictors on extension in years of service
Full Model Regression Hierarchical
regression Independent variables Standardized regression coefficients (β) Relative weights (ε) Percentages of predictable variance (%) R² Δ R² 1.Demographic variables 13.5 0.02 0.02 Gender 0.76 .00 0.3 Years of service -0.41 .00 0.8 Job position -2.23* .02 12.4 2. Predictors 86.5 0.14** 0.12** Flexible working options -2.21* .01 5.1 Training and development practices -0.26 .01 8.5 Job design practices 3.05** .04 29.8 Recognition and respect practices -0.12 .01 10.0 Performance evaluation practices -1.14 .01 6.5 Compensation practices 2.47* .04 26.4
Note. * Correlation is significant at the .05 level (1-tailed). ** Correlation is significant at the .01 level (1-tailed).
3.2.3 Is there one specific HR strategy that is perceived to be the most effective in retaining the older worker?
The first column of table 5 shows the weights from the regression model. The β-weights demonstrate that job design practices seem to be the most effective human resource strategy in retaining the older worker. The second column of table 3 shows the epsilon values or relative weights. The third and the fourth columns show the relative percentages of predictable variance in retirement intentions.
In the presence of multicollinearity relative weights or epsilon weights (Johnson, 2000) can provide for estimates of variable importance. Relative weights regard the predictive power of a variable isolated from other variables and combined with other variables. The desirable quality for relative importance index is obtained by breaking up the overall model R square such that the variables’ set a value of importance represent independent components of R square and all components sum to R square. The relative weights can be utilized to rank order variables. (Lebreton, Hargis, Griepentrog, Oswals & Ployhart, 2007) The relative weights analysis is conducted because relative weights are more sensible than β-weights because they express the proportional contribution of a predictor to the prediction of a dependent variable (Johnson, 2000). The relative weights analysis shows that job design practices (29.8%) and compensation practices (26.4%) predict the highest amount of variances in years of extension years in service due to implementation of HR practices.
3.2.4 Do employers and older employees differ in their view about the offered HR tools?
Figure 4 shows the perceived degree of the employed HR instruments in the organizations of the employees (purple) and the employer (blue). It is obvious that in two of the three organizations the employer scores higher on the HR instruments. In the first organization the employer only scores higher on flexible working options.
0,00 1,00 2,00 3,00 4,00 5,00 6,00 1 2 3 4 5 6 HR instruments E m p lo ym en t H R in st ru m en ts Ser i es1 Ser i es2 0,00 1,00 2,00 3,00 4,00 5,00 6,00 1 2 3 4 5 6 HR instruments E m p lo ym en t H R in st ru m en ts Ser i es1 Ser i es2 A B 0,00 0,50 1,00 1,50 2,00 2,50 3,00 3,50 4,00 4,50 5,00 1 2 3 4 5 6 HR instruments E m p lo ym en tH R in st ru m en t Series1 Series2 C
Figure 4: The mean scores of the employer and employee on the different HR instruments Note. Series 1=employer, series 2=employee
Note. HR instrument 1 = flexible working options, HR instrument 2= Training and development practices, HR instrument 3=Job design practices, HR instrument 4= Recognition and respect practices, HR instrument 5= Performance evaluation practices and HR instrument 6 = Compensation practices. Note. A= public government body, B= transport company and C=electricity company
Table 7 shows the multiple comparisons from the one-way ANOVA between the three different organizations. The mean differences between the organizations are significant for nearly all HR instruments. This means that the three organizations vary in the differences between the ratings of the employer and employee on the employment of the HR practices. The public government body scored significantly lower on all the HR instruments (except flexible working options) than the public government body and the transport company. This means that the employer and employee of the public government body agree more on the employment of the HR practices than the other organizations. The employer and the employees of the transport company agreed more on the HR instruments training and development and compensation practices than the electricity company. The employer and the employees of the transport company agreed more on the HR instruments job design practices and performance evaluation practices than the electricity company.
Table 6: Degree of agreement on the employment of HR instruments
Variable Degree of the agreement between employer and employee
Highest Lowest
Flexible working options Electricity company Transport company Public government body Training and development
practices
Public government body Transport company Electricity company
Job design practices Public government body Electricity company Transport company Recognition and respect
practices
Public government body Electricity company Transport company
Performance evaluation practices
Public government body Electricity company Transport company
Table 7: One way between groups ANOVA analysis, multiple comparison Variable Organization (I) Organization (J) Mean difference
(I-J) Confidence interval Lower bound Upper bound -0.16 -0.44 1 2 3 -1.52* -1.93 -1.130.13 2 1 3 -1.37*0.16 -0.13-1.74 -1.010.44 Flexible working options 3 1 2 1.53*1.37* 1.131.01 1.931.73 0.61* -0.90 -0.31 1 2 3 2.36* 1.38 2.13 -0.61* -2.77 -1.94 2 1 3 1.50* -2.13 -1.38 -2.36* -1.40 -0.76 Training and development practices 3 1 2 -1.75* 0.96 1.86 -1.08* 0.76 1.40 1 2 3 1.41* 2.08 2.90 1.08* -1.86 -0.96 2 1 3 2.49* -2.90 -2.08 -1.41* -0.64 0.04 Job design practices 3 1 2 -2.49* 0.22 1.18 -0.30 -0.04 0.64 1 2 3 0.70* 0.57 1.43 0.30 -1.18 -0.22 2 1 3 1.00* -1.43 -0.57 -0.70* -1.54 -0.83 Recognition and respect practices 3 1 2 -1.00* 1.28 2.26 -1.19* 0.83 1.54 1 2 3 1.77* 2.51 3.40 1.19* -2.26 -1.28 2 1 3 2.96* -3.40 -2.51 -1.77* 0.21 0.86 Performance evaluation practices 3 1 2 -2.96* 2.48 3.40 .54* -0.86 -0.21 1 2 3 2.94* 1.99 2.80 -.54* -3.40 -2.48 2 1 3 2.41* -2.82 -1.99 -2.94* -0.90 -0.31 Compensation practices 3 1 2 -2.41* 1.38 2.13
Note. * The mean difference is significant at the .01 level
3.3. Analysis per hypothesis
3.3.1. Hypothesis 1: The discrepancy between employer and employee on the perceived implementation of HR practices (flexible working, training and development, job design, recognition and respect, performance evaluation and compensation) will have a negative relationship with career opportunities, autonomy, social support, work-life balance and job satisfaction. Career opportunities, autonomy, social support, work-life balance and job satisfaction will have a positive relationship with desired retirement age.
The discrepancy between employer and employee on the perceived implementation of HR practices (flexible working, training and development, job design, recognition and respect, performance evaluation and compensation) will have a negative relationship with desired retirement age.
3.3.1.1. The discrepancy between employer and employee on the perceived implementation of HR practices will have a positive relationship with desired retirement age.
Support for the relationship between the discrepancy between employer and employee on the perceived implementation of HR practices and desired retirement age is not found (r = 0.01, n.s.). The regression analysis (table 8) shows that the employer-employee implementation of HR practices discrepancy does not significantly contribute to predicting variance in desired retirement.
Job satisfaction Organizational commitment
Autonomy Support from managers
Desired retirement age The discrepancy between employer and employee on the perceived implementation of HR practices The discrepancy between employer and employee on the perceived implementation of HR practices Desired retirement age
Table 8: Results of the hierarchical regression analysis of the predictors on desired retirement age R R² Adjusted R² Δ R² Δ F Sig. Demographic variables 0.08 0.01 -0.1 0.01 0.06 .98 Employer-employee implementation of HR practices discrepancy 0.24 0.06 -0.16 0.05 0.50 .69 Dependent variable:
3.3.1.2. The discrepancy between employer and employee on the perceived implementation of HR practices will have a negative relationship on career opportunities, autonomy, social support, work-life balance and job satisfaction.
Table 3 shows that the variable the discrepancy between employer and employee on the perceived implementation of HR practices is significantly related to autonomy (r = -.33, p < .01), support from managers (r = -.42, p < 0.1), job satisfaction (r = -.16, p <
.05) and organizational commitment (r = .24, p < .01). When the discrepancy
between the HR director/representative and the employee on the perception of use of HR practices is high the score on autonomy, support from managers, job satisfaction and organizational commitment is small.
The discrepancy between employer and employee on the perceived implementation of HR practices Job satisfaction Organizational commitment Autonomy Support from managers
3.3.1.3. Career opportunities, autonomy, social support, work-life balance and job satisfaction will have a positive relationship on desired retirement age.
Table 3 shows that the dependent variable desired retirement age is significantly related to autonomy (r = .15, p < .01), support from managers (r = .15, p < 0.1), job satisfaction (r = .11, p < .05) and organizational commitment (r = .17, p < .01). This means that when employees score high on job satisfaction, organizational commitment, autonomy and support from managers they also score higher on the desired retirement age.
The regression analysis (table 9) shows that the predictors seem to contribute significantly to the predicted variance on desired retirement age. Organizational commitment seems to contribute most to the predicted variance in desired retirement age with a β of .12 (p < .10) and a percentage of predictable variance of 37%.
Job satisfaction Organizational commitment
Autonomy Support from managers
Desired retirement age
Table 9: Results of Regression analysis, Relative weight analysis and Hierarchical regression analysis of the predictors on desired retirement age
Full Model Regression Hierarchical
regression Independent variables Standardized regression coefficients (β) Relative weights (ε) Percentages of predictable variance (%) R² Δ R² 1.Demographic variables 0.60 0.01 0.01 Gender -0.04 .00 2.2 Years of service 0.06 .00 5.9 Job position 0.02 .00 3.1 2. Predictors 4.5 0.05* 0.04* Autonomy 0.10 .01 23.1 Support from managers 0.07 .01 22.2 Job satisfaction -0.01 .00 6.5 Organizational commitment 0.12 .02 37.0
Note. * Correlation is significant at the .05 level (1-tailed). ** Correlation is significant at the .01 level (1-tailed).
3.3.2. Hypothesis 2: The discrepancy between employer and employee on the perceived implementation of HR practices will have a positive influence on Job stress, Routine, high Workload and Pressure from the employer. Job stress, Routine, Workload and Pressure from the employer will have a negative influence on the desired retirement age.
3.3.2.1. The discrepancy between employer and employee on the perceived implementation of HR practices will have a positive influence on Job stress, Routine, high Workload and Pressure from the employer.
Table 3 shows that the variable discrepancy between employer and employee on the perceived implementation of HR practices is significantly related to routine (r = .26, p < .01), however it is not related to; job stress (r = .02, n.s.), workload (r = .03, n.s.) and pressure from the employer (r = .08, n.s.). When the discrepancy between the HR director/representative and the employee on the perception of the use of HR practices is higher the score of routine is also higher.
The discrepancy between employer and employee on the perceived implementation of HR practices - Job Stress - Routine - Workload - Pressure from employer The discrepancy between employer and employee on the perceived implementation of HR practices - Job Stress - Routine - Workload - Pressure from employer Desired retirement age
3.3.2.2. Job stress, Routine, Workload and Pressure from the employer will have a negative influence on the desired retirement age
Table 10 shows that the dependent variable desired retirement age is significantly related to job stress (r = -.17, p < .01), pressure from the employer (r = -.20, p < 0.1) and workload (r = -.17, p < .01) Routine is not related to desired retirement age (r =
.02, n.s.).
As hypothesized the variables job stress, pressure from the employer and workload result in negative intentions about desired retirement age. The predictors seem to contribute for 7% (p < .01) to the predicted variance on the desired retirement age.
Table 10: Results of the hierarchical regression analysis of the predictors on desired retirement age
R R² Adjusted R² Δ R² Δ F Sig. Demographic variables 0.08 0.01 -0.01 0.01 0.49 .69 Job stress, Pressure from employer, Workload, Routine 0.28 0.08 0.05 0.07** 5.15 .00 - Job Stress - Routine - Workload - Pressure from employer Desired retirement age
