Psychological career resources , career adaptability and work engagement of generational cohorts in the media industry

Employees within the contemporary, ever changing world of work are faced with numerous challenges, such as diminished employment opportunities, uncertain career paths, reduced job security, fast-changing technology and an increasing personal responsibility for frequently re-evaluating and making adjustments to their careers (Chabault, Hulin & Soparnot, 2012; Potgieter, 2012). In light of this context, organisations are required to prioritise their strategic mandate in attracting, retaining, engaging and developing key employees in order to increase their business performance and remain relevant (Del Corso & Rehfuss, 2011; Obschonka, Silbereisen & Wasilewski, 2012; Savickas, 2012; Savickas & Porfeli, 2012; Schuler, Jackson & Tarique, 2011). Orientation: The global skills crisis coupled with the aging workforce, rapid technological advances and changing nature of work have infringed various challenges upon organisations and employees. Media organisations in particular are affected by these trends, with retention further at risk because of the specialised and scarce skills sought and the versatility and ambiguity inherent in the nature of careers within the media industry, therefore resulting in engagement and skills retention being high on the agenda.


Background to the study
Restructuring, the re-organisation of production processes, the need to ensure economic viability and the use of increasingly advanced technology have led to drastic changes in the nature of careers and employment relationships in the media industry (International Labour Office, 2014).This has resulted in the workforce increasingly facing instability and ambiguity throughout their careers, as they are no longer guaranteed job security, loyalty and linear career paths by their employers (Lent, 2013;Santilli, Nota, Ginevra & Soresi, 2014;Segers & Inceoglu, 2012;Zacher, 2014).These experiences may have an adverse impact on individuals' energy, dedication and interest to the organisation, that is, impacting the extent to which employees are engaged.Furthermore, this might influence the retention of valuable talent within organisations (Festing & Schafer, 2014;Mohlala, Goldman & Goosen, 2012;Olckers & Du Plessis, 2012).
Media organisations often depend on talent, human creativity and a highly specialised and skilled workforce in order to drive their core strategic mandate (International Labour Office, 2014).Therefore, employees in the media industry are required to be increasingly flexible, adaptable and continuously updated on their skills and knowledge.The main challenge pertaining to the participating organisation is the lack of organisational development knowledge and practices pertaining to career development and engagement strategies, which once embedded may positively contribute towards the career success, employability and retention of an organisation's high-potential employees (Festing & Schafer, 2014;Mohlala et al., 2012).Individuals' psycho-social career meta-capacities include a variety of self-regulatory adaptive capacities (Savickas & Porfeli, 2012) such as career adaptability and active consciousness of one's career preferences, values, motivations, problem solving and relational skills, career self-management competencies, self-esteem and emotional literacy (Coetzee, 2014).These capacities enable individuals to become selfsufficient, adaptive learners in managing their own careers (Bezuidenhout, 2011;Coetzee & Roythorne-Jacobs, 2012;Savickas & Porfeli, 2012).Researchers emphasise that psychosocial career meta-capacities may affect the general employability attributes of an individual, which is viewed as an alternative to job security and thus a crucial enabler for career success in the 21st century workplace (Guest, Isaksson & De Witte, 2010;Naute, Van Vianen, Van Der Heijden, Van Dam & Willemsen, 2009;Savickas & Porfeli, 2012).Therefore, more researchers and practitioners are arguing that organisations need to prioritise and integrate employees' career meta-capacities into their strategic talent retention strategies and interventions as it is believed to facilitate positive outcomes such as high level of work engagement, high-performance organisational culture, lifelong learning skills, career competencies, coping strategies and talent retention (Coetzee & Roythorne-Jacobs, 2012;Coetzer & Rothmann, 2007;Mendes & Stander, 2011).
Additionally, it is noted that diversity is an important development and challenge within South African media organisations.Employers in South African organisations should seek to understand how individuals' sociodemographic characteristics, such as generational cohort (age), may influence their psycho-social career meta-capacities (psychological career resources and career adaptability) and their work engagement (Potgieter, 2012).The current work context requires individuals to realise and rely on their psycho-social resources or meta-capacities in order to manage and own their careers, achieve a sense of integration with their self and the environment, adapt to job demands and unplanned events and strive for personal and professional development (Coetzee, 2014;Ebberwein, Krieskok, Ulven & Prosser, 2004;Hobfoll, Johnson, Ennis & Jackson, 2003;Rottinghaus, Buelow, Matyja & Schneider, 2012;Savickas & Porfeli, 2012).

Psychological career resources
The concept of psychological career resources was developed by Coetzee (2008Coetzee ( , 2014) ) and refers to all competencies and subjective experiences that significantly influence the promotion and facilitation of individuals' proactive career behaviour and career development over time.Other researchers state that psychological career resources include attributes and abilities such as behavioural adaptability, career orientation consciousness, self-insight, sense of purpose, self-esteem and emotional literacy (Coetzee, 2008;Ferreira, 2012;Ferreira, Basson & Coetzee, 2010;Hall & Chandler, 2005).
Coetzee's (2014) psychological career resources model is of specific relevance to this study, as it has been developed and tested for adults in the South African context (Bezuidenhout, 2011).The psychological career resource model encapsulates an integrated career framework, in order to provide a holistic picture of an individual's career enablers, career drivers, career harmonisers, career preferences and career values, which can be linked to an individual's extrinsic or intrinsic lifetime career experience that leads to career success (Gunz & Heslin, 2005): • Career preferences and values represent an ongoing cognitive roadmap of the career pathways that an individual perceives as interesting and meaningful to them (Coetzee, 2008(Coetzee, , 2014)).However, although career preferences and career values are intertwined, they are distinct from each other.Career preferences refer to the career actions or activities that guide long-term career growth and direction, while career values represent cognitive expressions of the ideas, needs or philosophies that are deemed as significant by the individual (Coetzee, 2014;Coetzee & Esterhuizhen, 2010;Lyons et al., 2006).Therefore, an individual's preference can be influenced by his or her values and is therefore necessary to understand the concept of value in order to better understand the concept of preference.According to Coetzee's (2014) psychological career resources model, the four career preferences comprise of stability or expertise, managerial, creativity or variety and autonomy or independence and two overarching values, which include growth or development and authority or influence (Coetzee, 2008(Coetzee, , 2014)).• Career enablers are seen as individuals' portable skills and abilities that can be conveyed across a variety of situations in order to help them achieve career success (Coetzee, 2008(Coetzee, , 2014)).Career enablers include an individual's practical or creative skills and self or other skills.Practical or creative skills are necessary for executing career options in original and inventive ways (e.g.conceptual thinking and problem-solving skills, goal setting, management skills and skills related to innovation and change, etc.), while self or other skills, often linked to one's emotional intelligence, are the skills necessary for self-reflection and effective interaction with others (e.g.self-awareness, empathy, self-management) (Coetzee, 2014;Ferreira et al., 2010).• Career drivers compose of an individual's career purpose, career directedness and career venturing.Career drivers are considered to be the motivational enablers that stimulate experimentation, exploration and risk-taking behaviour with possible careers, based on a selfassessment of one's abilities for present and future work roles (Coetzee, 2008(Coetzee, , 2014)).These drivers, in return, provide a view of the possible self and roles that an individual can define, create or express within the career construction or design process.• Career harmonisers are the psychological attributes that not only act as promoters of flexibility and resilience, but also as controls (because they keep career drivers in balance so that people do not burn themselves out in the process of pursuing and reinventing their careers) (Coetzee, 2008(Coetzee, , 2014)).Coetzee (2008Coetzee ( , 2014) labels these career harmonisers attributes as self-esteem, behavioural adaptability, emotional literacy and social connectivity.
Individuals with a well-developed psychological career resources profile are also more likely to be engaged in their work and committed to their job, work, career, occupation or the organisation (Converse, Pathak, DePaul-Haddock, Gotlib & Merbedone, 2012;Fugate, Kinicki & Ashforth, 2004;Savickas & Porfeli, 2012).A study by Tladinyane (2013) showed that individuals who reported high levels of selfesteem also exhibited high vigour and absorption to their work.Coetzee and Bergh (2009) reported that individuals' psychological career resources and sociodemographic variables predict their life satisfaction, job and career satisfaction, sense of happiness and the meaning they attach to the importance of work in their lives.Ferreira et al. (2010) indicated that the younger age group (25 and younger) reported higher on self-esteem and ability to form meaningful social relationships than older individuals (56 and older) age group.Their study further highlighted that individuals in the late career life stage (56 and older) appear to have a significantly higher need to venture out in search of new career opportunities (Ferreira & Coetzee, 2010).
Psychological career resources underpin the key concepts at the core of the new economy career arrangement as they equip individuals with capacities to actively seek opportunities, update transferrable competencies and skills and transform into agentic and continuous learners (Coetzee, 2014;Potgieter, 2012).Therefore, the use of psychological career resources is likely to promote proactive career behaviour and empower the individual in controlling his or her employment journey and being relevant within a highly competitive environment, all of which will ultimately lead to higher levels of personal fulfilment and satisfaction.

Career adaptability
Career adaptability is a psycho-social construct that indicates an individual's readiness and resources for coping with current and anticipated occupational development tasks, vocational transitions as well as personal dramas (Ferreira, 2012;Guan et al., 2014;Ismail, Ferreira & Coetzee, 2016;Öncel, 2014;Savickas, 2012).Various scholars have characterised career adaptability as a set of transactional resources (i.e.attitudes, competencies and behaviours) that facilitate employment seeking options and career improvement behaviour, assist with person -environment congruence and empower positive immersion into one's work (Klehe, Zikic, Van Vianen, Koen & Buyken, 2012;Savickas, 1997Savickas, , 2012;;Tolentino, Garcia, Restubog, Bordia & Tang, 2013).Career adaptability is underpinned by selfregulatory and agentic processes which are viewed as preparing individuals to plan and cope with current and anticipated occupational changes and transitions (Savickas & Porfeli, 2012;Tolentino et al., 2013).Savickas and Porfeli (2012) explain career adaptability as a multidimensional, hierarchical structure which comprises four subscales, namely, career concern, control, curiosity and confidence: • Concern refers to an interest with aligning oneself to one's future goals, and acknowledgement that present actions (i.e.awareness, involvement and preparedness) are linked to attainment of future career aspirations.Individuals draw from these four self-regulatory resources in order to craft solutions for adapting within a context of complexity and unfamiliarity pertaining to career development transitions, unplanned events and obstacles (Johnston, Luciano, Maggiori, Ruch & Rossier, 2013;Maggiori, Johnston, Krings, Massoudi & Rossier, 2013;Savickas & Porfeli, 2012;Tolentino et al., 2013;Zacher, 2014).
Although studies tend to suggest no differences between age groups on career adaptability levels (Rossier, Zecca, Stauffer, Maggiori & Dauwalder, 2012;Shunmugum, 2016), career adaptability is seen as an important set of psycho-social career meta-capacities that shape the problem-solving strategies and coping behaviours that individuals' use to synthesise their vocational self-concepts with work roles.Furthermore, as an important set of individual resources, career adaptability resources influence several other career-related variables and outcomes to various extents, such as work engagement, job satisfaction, career anxiety, successful job transitions, workstress and tenure (Brown, Bimrose, Barnes & Hughes, 2012;Ismail et al., 2016;Rossier et al., 2012).

Work engagement
Work engagement includes experiencing a deeper dimension of well-being, emotional and behavioural responses, such as joy, energy and fulfilment at work (Hallberg & Schaufeli, 2006;Maslach, Schaufeli & Leiter, 2001;Schaufeli, 2013;Schaufeli, Bakker & Van Rhenen, 2009).Schaufeli et al. (2009) define work engagement as a positive, fulfilling, workrelated state of mind that comprises three dimensions, namely, vigour, dedication and absorption: • When employees are engaged, they exude high-energy levels and want to devote time and effort to their work (vigour).• They feel enthusiastic, inspired and view their work tasks as a significant and meaningful pursuit (dedication).• They are fully concentrated, giving sufficient attention to their work to the point whereby they experience time moving quickly (absorption).
Engaged employees harness themselves to their work by fully investing their heads, hearts and hands in performing their role (Breevaart, Bakker, Demerouti & Hetland, 2012;Kahn & Heaphy, 2014;Schaufeli, Bakker & Salanova, 2006).Individuals' will become more or less interested (i.e.engaged) in activities as a function of the degree to which they experience and need satisfaction while engaging in those activities (Linley, Nielsen, Gillett & Biswas-Diener, 2010).
According to Saks (2006), work engagement is associated with an individual's attitudes, intentions and behaviours.Hence, engaged employees are likely to be more attached to their organisation and would have a lower propensity to leave the organisation (Schaufeli & Bakker, 2004).This view is supported by several researchers who found that work engagement is negatively related to turnover intention (Bakker & Demerouti, 2008).
Given the significance of work engagement for both the individual and the organisation, various researchers have explored the influence of certain biographical variables on the process of work engagement (Bezuidenhout, 2011;Schaufeli & Bakker, 2004).Within South Africa's multicultural context, it may be beneficial to also explore the sociodemographic differences in relation to work engagement, as this information may assist organisations in customising their career development practices to suit the needs of each unique group.More specifically, older employees seem to be more engaged in their work (Schaufeli et al., 2006).Macey and Schneider (2008) indicated that full-time employees are more engaged than part-time employees because full-time individuals experience higher empowerment in terms of decision-making affecting their job or work and thus have high levels of engagement.Work engagement is related to a range of positive organisational outcomes such as job satisfaction, motivation, a decline in intention to quit, employee well-being and health (Rothmann & Rothman, 2010;Schaufeli & Bakker, 2004).
In conclusion, work engagement could be understood as the underlying energy of the organisation which might be utilised towards organisational success and individual benefits (Schaufeli et al., 2006).As personal resources are believed to contribute towards the employees' work engagement, attention needs to be given to the potential influence of psycho-social career meta-capacities (as explained by psychological career resources and career adaptability) on work engagement.In recent years, organisations have leaned on financial rewards and other benefits to uplift engagement levels (Schaufeli et al., 2009).However, the emerging trend that has come to the forefront is the notion that personal resources, or psycho-social career meta-capacities (rather than taking the approach solely on external rewards and factors) could be employed as the driver of engagement (Hoole & Bonnema, 2015).

Generational cohorts
Researchers have long argued whether differences exist among generational cohorts in terms of what motivates and drives them to perform optimally (Drake, 2012;Ismail et al., 2016;Kapoor & Solomon, 2011)

Research objective
The present study aimed to investigated (1) whether individuals psycho-social career meta-capacities predict their work engagement and (2) the differences between generational cohort (age) in terms of their career metacapacities (psychological career resources and career adaptability) and employee work engagement.The study may potentially contribute to career and retention theory by shedding light on psychological attributes that may influence frameworks and strategies in retaining diverse age (generational) groups of employees within the South African multicultural work environment.
The next section of the article will focus on the research design, which comprises the research approach and method, followed by the presentation of the results and a discussion of the findings.The article concludes with a brief summary of the key conclusions, implications for practice and recommendations for potential future research.

Research approach
A cross-sectional quantitative research approach was followed in order to achieve the research objective.

Research method Participants
A stratified random sample of 248 (N = 248) employed adults participated in the study.The participants were employed within a media organisation situated in Gauteng.They were mostly employed on skilled level (75%) and comprised predominantly 54% black people and 63.3% females in their early career stage (80% = < 45 years), 63.3%.The representations from the different generational cohorts were Gen Y (18-40 years) 63.3%, Gen X (40-49 years) 17.7% and Baby Boomers (50 years and older) 19.0%.

Measuring instruments
The Psychological Career Resources Inventory: The Psychological Career Resources Inventory (PCRI) developed by Coetzee (2008) is a self-rated multifactorial measurement tool.The PCRI contains 64 items that include five subscales: career preference (17 items), career values (8 items), career enablers (8 items), career drivers (8 items) and career harmonisers (21 items).A six-point Likert-type scale was used for subject responses to each of the items.An exploratory factor analysis provided evidence of construct validity as indicated by the five-factor model, which supports the underlying dimensions of employability described by Coetzee (2008).In terms of reliability (internal consistency), Cronbach's a coefficients for each subscale range from 0.79 (moderate) to 0.91 (high).In terms of the present study, the overall PCRI scale obtained a reliability coefficient of 0.95.The internal consistency reliability coefficients for the five subscales ranged between 0.79 (career values) and 0.91 (career harmonisers).
The Career Adapt-Abilities Scale: The Career Adapt-Abilities Scale (CAAS) developed by Savickas and Porfeli (2012) is a self-rated multifactorial measurement tool.The CAAS contains 24 items that include four subscales: concern (six items), control (six items), curiosity (six items) and confidence (six items).A five-point Likert-type scale was used for subject responses to each of the items.An exploratory factor analysis provided evidence of construct validity as indicated by the four-factor model, which supports the underlying dimensions of employability described by Savickas and Porfeli (2012).In terms of reliability (internal consistency), Cronbach's a coefficients for each subscale range from 0.83 (moderate) to 0.90 (high).
In terms of the present study, the overall CAAS scale obtained a reliability coefficient of 0.94.The internal consistency reliability coefficients for the five subscales ranged between 0.83 (control) and 0.90 (confidence).
The Utrecht Work Engagement Scale: The Utrecht Work Engagement Scale (UWES) developed by Schaufeli and Bakker (2004) is a self-rated multifactorial measurement tool.The UWES contains 21 items that include three subscales: vigour (eight items), dedication (five items) and absorption (eight items).A seven-point Likert-type scale was used for subject responses to each of the items.An exploratory factor analysis provided evidence of construct validity as indicated by the three-factor model, which supports the underlying dimensions of employability described by Schaufeli and Bakker (2004).In terms of reliability (internal consistency), Cronbach's a coefficients for each subscale range from 0.88 to 0.90 (high).In terms of the present study, the overall UWES scale obtained a reliability coefficient of 0.96.The internal consistency reliability coefficients for the five subscales ranged between 0.88 (dedication) and 0.90 (absorption).
A biographical questionnaire was used to determine the generational cohorts (age) of the participants.

Research procedure
Nine hundred questionnaires were distributed, with a total of 248 usable questionnaires returned (N = 248), yielding a response rate of 28%.The participants were invited to voluntarily participate in the study.The questionnaires were electronically distributed via an e-mail link.Each questionnaire included a cover letter inviting respondents to participate voluntarily in the study and assuring them that their individual responses would remain anonymous and confidential.The cover letter also stated that completing the questionnaires and returning them constituted agreement to use the results for research purposes only.

Statistical analysis
Preliminary statistical analyses assessed for common method variance and the construct validity of the three measuring instruments.Descriptive statistics (means, standard deviations and Cronbach's a coefficients), stepwise regression analysis and an analysis of variance (ANOVA) test with pairwise comparison were conducted to achieve the objectives of the study.In order to counter the probability of a type I error, it was decided to set the significance value for interpreting the results at a 95% confidence level ( p ≤ 0.05).

Ethical considerations
Ethical clearance was obtained from the University of South Africa and permission to conduct the study was obtained from the organisation.

Preliminary analysis: Common method variance and construct validity
Because of the three self-rating measures used and the crosssectional research design of this study, the preliminary data analysis involved testing for common method variance and construct validity of each measuring scale.This procedure entailed conducting a Harmann's one-factor solution and confirmatory factor analysis (CFA) by using the CALIS procedure in SAS ( 2013).The one-factor solution showed that the single factor that emerged for the PCRI accounted for only 16.Overall, in line with the guidelines of Podsakoff, Mackenzie, Lee and Podsakoff (2003), the one-factor results for the PCRI, CAAS and UWES suggested that common method bias did not pose a threat to the research findings.

Descriptive statistics
The reliability coefficients reported in Table 1 show acceptable internal consistency reliability of the three scales and the subscales.The overall reliability coefficient of the PCRI was very high (α = 0.95) as well as the reliability coefficient for the CAAS (α = 0.94) and the UWES (α = 0.96).These coefficients indicated strong overall internal consistency for the three scales.
Table 1 shows that all the PCRI variables were positively and significantly associated with all the CAAS variables (range: r ≥ 0.30 to r ≤ 0.72; p ≤ 0.001; moderate to large practical effect).Table 1 further shows that the overall PCRI scale significantly and positively correlated with the overall UWES scale (r = 0.22; p 0.001; small practical effect).With the exception of the career preferences and career values variables, all the other PCRI variables correlated significantly and positively with the engagement variables (range: r ≥ 0.15 to r ≤ 0.38; p ≤ 0.05; small to moderate practical effect).
The overall CAAS variable positively and significantly correlated with the overall UWES variables (r = 0.33; p ≤ 0.001; moderate practical effect).Furthermore, positive and significant correlations were observed between all four career adaptability variables and UWES variables (range: r ≥ 0.22 and r ≤ 0.39; p ≤ 0.001; small to moderate practical effect).Overall, the correlation range was below the threshold of < 0.90 for multicollinearity concerns.
Age (generational cohort) showed a significant and positive correlation with the overall UWES (r = 0.34; p ≤ 0.001; moderate practical effect) and all the UWES variables of vigour, dedication and absorption (range: r ≥ 0.31 and r ≤ 0.33; p ≤ 0.001; moderate practical effect).Age significantly and negatively correlated with the overall PCRI (r = -0.14;p = 0.02; small practical effect) and with the career preferences (r = -0.27;p ≤ 0.001; small practical effect), career values (r = -0.22;p ≤ 0.001; small practical effect) and career drivers (r = -0.13;p ≤ 0.05; small practical effect).No correlations were observed between the generational cohorts (age) and the CAAS variables.

Stepwise regression
A stepwise regression analysis was conducted using generational cohort or age, PCRI variables (career  http://www.sajhrm.co.zaOpen Access preferences, career values, career enablers, career drivers and career harmonisers) and CAAS variables (career concern, career control, career curiosity and career confidence) as the independent variables, and UWES (overall work engagement) as the dependent variable.Stepwise regression is a statistical technique that includes regression models in which the choice of predictive variables is carried out by an automatic procedure.The backward elimination procedure was applied which involves starting with all the variables and testing them one by one for statistical significance, deleting any that were not significant.Table 2 summarises the results.

Mean (SD
The results showed that the regression model was significant (F = 29.93;p = 0.000; = 0.27; ∆R² = 0.02; ∆F = 5.63; ∆Fp = 0.02).The adjusted R² value of 0.27 indicated that the model predicted approximately 27% (large practical effect) of the variance in the dependent variable (work engagement).Table 2 shows that only confidence; generational cohort (age) and career harmonisers acted as significant predictors of work engagement.Generational cohort (age) contributed the most towards explaining the variance in work engagement (ß = 0.32; p ≤ 0.000), followed by career confidence (ß = 0.29; p ≤ 0.000) and the career harmonisers (ß = 0.15; p ≤ 0.000).The collinearity statistics indicated that the tolerance values were all close to 1 and the variance inflation factor (VIF) is less than 2.5 (implying little or no multicollinearity concerns).

Test for significant mean differences
The tests for normality indicated that the data were normally distributed, a non-parametric statistical technique, namely a one-way ANOVA with pairwise comparison was applied to test for significant mean differences between the generational cohort groups.
For the purposes of statistical analysis and reporting, Baby Boomers were classified as participants 50 years and older, Generation X employees were classified as participants between the ages of 40 and 49 and Generation Y was classified as participants between the ages of 18 and 40 years.In order to detect the source of significant differences, the Fisher's least significant difference (LSD) test was used to allow for pairwise comparisons between the generational (age) groups and the overall psychological career resources, career adaptability and work engagement variables.

Discussion
The study explored individuals' age, psychological career resources and career adaptability in relation to their work engagement and how these manifest in the three generational cohorts: Generation Y (18-40 years), Generation X (40-49 years) and Baby Boomers (50+ years).
Overall, the findings of the present study suggest that work engagement is likely to be influenced by an individual's generational cohort (age), their confidence in engaging in career self-management behaviour (Savickas & Porfeli, 2012) and to a lesser extent their psychological career capacities (Coetzee, 2014) comprising their self-esteem, behavioural adaptability, emotional literacy and social connectivity (career harmonisers).The findings corroborate previous research that showed that individuals who feel more efficacious (confident) in their capacity to solve problems and surmount challenges and obstacles in the pursuit of constructing their careers will most likely feel more engaged in their work (Hartung, 2013;Potgieter, 2012) (Tladinyane, 2013;Venter, Coetzee & Basson, 2013).The proactive career behaviour associated with individuals' psycho-social career meta-capacities may translate into greater discretionary effort and energy invested in their work (Bakker & Demerouti, 2008;Rossier et al., 2012).
The Generation Y individuals had higher levels of psychological career resources (career preferences, career values and career drivers), while the Generation X individuals had higher career curiosity than the other two generational cohorts.The Baby Boomers showed higher levels of work engagement than the other two generational cohorts.Generational cohort research highlights that Baby Boomers value achievement, hierarchy and advancement fulfilment (Cennamo & Gardner, 2007;Kupperschmidt, 2000;Smola & Sutton, 2002;Zemke, Raines & Filipczak, 2000) which may explain their higher levels of engagement.However, research on age and work engagement has been highly inconsistent.The findings support some research that has found Baby Boomers to be the most engaged generational cohort (Park & Gursoy, 2012;Schaufeli et al., 2006Schaufeli et al., , 2009)).These findings highlight that Baby Boomers should not be assumed to be the 'dead wood' of the organisation, as they were found to possess significantly high levels of work engagement.The higher levels of psychological career resources (career preferences, career values and career drivers) observed for the youngest generational cohort (Generation Y: 18-40 years) could be attributed to them being in the early stages of their careers.The findings suggest that they have a higher need for clarifying their career motives, interests and values and are seemingly therefore more purposeful (driven) in clarifying the meaning of work for them, having clear career direction and exploring various avenues in which their career paths may unfold (Coetzee, 2014).The findings corroborate research that has reported that younger individuals' tend generally to be more adaptable to their career context because of their higher acceptance of change and job insecurity and higher inclination towards career exploration (Gursoy, Chi & Karadag, 2013;Ismail et al., 2016;Potgieter, 2012;Zemke et al., 2000).
The higher level of career curiosity of the Generation X participants (40-49 years) could be attributed to them being in the establishment phase of their careers, which could explain the stronger need for having clarity about their career path in the organisation (Savickas & Porfeli, 2012).This finding supports previous studies that found that respondents between the ages of 41 and 45 years were more curious about looking ahead and exploring future possibilities (Potgieter, 2012;Savickas & Porfeli, 2012).

Study limitations and future suggestions
The present study was limited to a stratified random sample of predominantly early career stage African female employees in a South African media company.The findings cannot therefore be generalised to other age contexts.Furthermore, given the cross-sectional nature of the research design, this study can yield no statements about causation.
Associations between the variables have therefore been interpreted rather than established.Longitudinal studies should be employed to establish the causal relationships among the variables.Further studies are needed with broader samples across various occupational, age and gender groups, cultural groups and economic sectors to replicate and extend the findings.

Conclusion Implications for practice
The findings may potentially inform interventions within organisations to enhance the work engagement of the different generational cohort groups for retention purposes.
The results of the study emphasise the importance of considering the influence of generational cohorts (age), psychological career resources and career adaptability on work engagement in order to assist organisations with the retention of valuable staff.The significant lower levels of engagement of the younger generations should be considered in the light of their higher concerns about their career development.Formal career discussions and clear career paths for the younger generations who are still in the early stages of their career may potentially influence their work engagement and retention.It is concluded that the findings contribute new knowledge to the field of career psychology and the media sector's career development practices.
Understanding how these three constructs are related can help initiate career paths that support employee development, career mobility, career management and retention practices.
71% of the covariance among the PCRI variables.The 15 PCRI variables were then loaded onto one factor to examine the fit of the CFA model.The PCRI single-factor model did not fit the data adequately (chi-square or df ratio = 3.68; p < 0.0001; RMSEA = 0.11; SRMR = 0.11; CFI = 0.42; AIC = 7446.88).The PCRI single-factor model fit indices were compared with the fit indices of a 15-factor solution measurement model for the PCRI (the 15 factors loading onto their respective meta-dimension and each of the five meta-dimensions loading on an overall latent variable: psychological career resources).The fit indices of the 15-factor measurement showed a more adequate model fit: (chi-square or df ratio = 2.23; p < 0.0001; RMSEA = 0.06; SRMR = 0.09; CFI = 0.94; AIC = 4613.07).A marginal value of RMSEA and SRMR for model acceptance is 0.10 and a value of 0.08 and lower is considered a good fit(Hamtiaux, Houssemand & Vrignaud, 2013;Park, Nam & Cha, 2012).The indices obtained for the 15-factor PCRI model indicate adequate fit of the data, implying acceptable construct validity of the scale.The one-factor solution for the CAAS showed that the construct accounted for only 10.24% of the covariance among the CAAS variables.When loading the four CAAS variables onto a single construct in the CFA model, the fit indices showed that the single factor did not fit the model well (chisquare or df ratio = 4.27; p < 0.0001; RMSEA = 0.12; SRMR = 0.09; CFI = 0.74; AIC = 1170.86).The CAAS single-factor model fit indices were compared with the fit indices of a four-factor solution measurement model for the CAAS (the four factors loading onto an overall latent variable: career adaptability).The fit indices of the four-factor measurement showed a more adequate model fit: (chi-square or df ratio = 2.36; p < 0.0001; RMSEA = 0.08; SRMR = 0.07; CFI = 0.90; AIC = 687.51).The indices obtained for the CAAS four-factor model indicate that the model represents the data adequately, implying acceptable construct validity of the scale.The one-factor solution for the UWES showed that the construct accounted for only 11.65% of the covariance among the UWES variables.When loading the three UWES variables onto a single construct in the CFA model, the fit indices showed that the single factor did not fit the model well (chisquare or df ratio = 4.19; p < 0.0001; RMSEA = 0.12; SRMR = 0.06; CFI = 0.84; AIC = 876.32).The UWES single-factor model fit indices were compared with the fit indices of a three-factor solution measurement model for the UWES (the three factors loading onto an overall latent variable: work engagement).The fit indices of the three-factor measurement showed a more adequate model fit: (chi-square or df ratio = 3.76; p < 0.0001; RMSEA = 0.09; SRMR = 0.05; CFI = 0.90; AIC = 790.22).The indices obtained for the UWES three-factor model indicate that the model represents the data adequately, implying acceptable construct validity of the scale.

TABLE 1 :
Descriptive statistics: Means, standard deviations, internal consistency reliability coefficients and zero-order correlations.

TABLE 2 :
Results of the stepwise regression analysis.

TABLE 4 :
Means and standard deviations of significant pairwise comparisons.