This study investigates how workforce age diversity influences voluntary employee turnover and how training and development (T&D) practices moderate this relationship. It explores whether complex and costly T&D programs reduce or, paradoxically, increase turnover by enhancing employees' external employability.
Drawing on data from 5,899 organizations across 38 countries participating in the CRANET Survey, this research applies multiple and moderation regression analyses to examine the direct and interactive effects of workforce age composition, complex T&D systems, and training costs on voluntary turnover.
Results reveal that a larger age gap correlates with higher voluntary turnover. However, this relationship is moderated by training complexity and costs. Complex T&D programs reduce turnover among older employees but increase it among younger staff by strengthening their marketability. High training expenditures intensify turnover among younger workers, while the stabilizing effect of older employees diminishes under extensive training schemes.
Managers should align development strategies with generational needs, combining skill enhancement with tailored retention initiatives to balance learning and stability in multigenerational organizations.
This paper extends HRM and HRD theory by demonstrating that age diversity and training interact in paradoxical ways. It contributes cross-national evidence on how generational composition and T&D investments jointly shape workforce stability, offering insights for strategic HRD in diverse labour markets.
1. Introduction
Employee turnover remains a major challenge for organizations globally, affecting productivity and performance. Generational differences within the workforce significantly influence turnover rate. Diverse age groups exhibit different work values and communication styles, leading to conflict and reduced job satisfaction. These disparities challenge human resource management (HRM) by balancing the retention of experienced staff with younger employee development. The workforce has become increasingly age-diverse. Research shows that generational gaps affect engagement and organizational commitment (McCarthy et al., 2020; Chen et al., 2023). Training programs help mitigate generational differences while improving skills, though they may increase younger staff marketability. Targeted programs such as mentoring facilitate intergenerational knowledge sharing (Ju and Li, 2019; Miao et al., 2020).
Although there is an increasing amount of research on employee turnover and generational diversity, significant gaps still exist in the literature. Many studies focus on factors affecting turnover, such as job demands, compensation, mentoring, and managerial quality (Chong et al., 2025; Brunt et al., 2025; Södergård et al., 2025; Bäker et al., 2025), while research on workforce demographics emphasizes the intricate effects of age diversity on turnover outcomes (De Meulenaere et al., 2022). However, most research tends to examine either the age composition of the workforce or training and development (T&D) practices in isolation, with limited focus on their interaction. While HR practices and training initiatives are often discussed as predictors of turnover (Li et al., 2024; Baydili and Tasci, 2025; Veglio et al., 2024), the combined effects of these factors across various workforce age structures are not thoroughly explored. Additionally, empirical evidence is predominantly derived from single-country samples, which restricts cross-national insights into how demographic workforce structures and training investments together influence voluntary turnover. Although training programs are generally anticipated to decrease turnover by enhancing employee commitment, their potential contradictory effect, boosting employee mobility through increased employability, has not been extensively examined in prior research (Cai et al., 2024).To address these gaps, this study seeks to examine age differences and voluntary turnover using the global CRANET survey data to investigate how training programs moderate these relationships.
Accordingly, the research question of this study is how workforce age diversity interacts with T&D practices to influence voluntary employee turnover across organizations. The primary objective of the research is to investigate the influence of age diversity within the workforce on voluntary employee turnover and to evaluate the role of T&D initiatives in moderating this relationship. The research specifically explores whether age disparities between younger and older employee cohorts contribute to elevated turnover rates and whether the complexity of training programs and the financial investment in training can mitigate these effects. By analyzing data from a large, cross-national sample of organizations, the study aims to provide a comprehensive understanding of how demographic workforce structures and investments in human resource development (HRD) collectively affect employee retention.
This research makes substantial contributions to the existing literature by introducing the age-gap paradox, demonstrating how age diversity and training investments interact to influence voluntary employee turnover. Utilizing data from 5,899 organizations across 38 countries, the study provides a broader perspective than typical single-country samples. By examining how the complexity of training programs affects the relationship between workforce age composition and turnover, the research integrates workforce demography with HRD practices.
2. Theoretical framework and literature review
The aim of the study is to investigate the correlation between age disparities among employee groups within an organization and the voluntary turnover that may arise from resultant friction and conflict. We propose that when the ages of employees within an organization are relatively uniform (i.e. there are no significant differences), these issues are less pronounced. Conversely, substantial age disparities are more likely to manifest in problems related to work styles, attitudes, life stages, core values, and other generationally pertinent factors, which may collectively contribute to increased turnover. Furthermore, we hypothesized that turnover can be mitigated if organizations implement comprehensive T&D programs as part of their diversity management initiatives. In the subsequent section, we examine the potential moderating effects of complex T&D programs and the associated training costs. Based on our preliminary assumptions, increasing age gaps between younger and older employee groups is expected to increase turnover most significantly in the absence of targeted development programs. Specialized programs tailored to age groups can enhance satisfaction and retention. The initial model is illustrated in Figure 1.
The flowchart begins with two factors: the proportion of employees under 25 and the proportion of employees aged 50 and above. These factors contribute to an age gap within the workforce. The age gap leads to complex training and development programs, which in turn result in increased training costs. These training costs are shown to influence voluntary employee turnover.Conceptual model of the research. Source: own edition
The flowchart begins with two factors: the proportion of employees under 25 and the proportion of employees aged 50 and above. These factors contribute to an age gap within the workforce. The age gap leads to complex training and development programs, which in turn result in increased training costs. These training costs are shown to influence voluntary employee turnover.Conceptual model of the research. Source: own edition
2.1 Contextual and theoretical foundations of employee turnover and training practices
Mayrhofer et al. (2024) emphasized internal and external factors in understanding global HRM practices. Their research shows companies in international markets invest more in T&D to enhance competitiveness. Farndale et al. (2017) observed that T&D practices are less context-dependent than other HRM practices. Employee turnover significantly affects organizational performance and competitiveness. Common causes include job stress, low satisfaction, poor work environments, limited advancement, inadequate pay, and work-life imbalance (Al-Suraihi et al., 2021). High turnover leads to recruitment costs, reduced productivity, and knowledge loss. Younis et al. (2023) found that central roles in organizational networks do not necessarily increase turnover, influenced by negative sentiments and low group efficacy. Incorporating social network parameters into HRM strategies can improve retention (Younis et al., 2023). The relationship between age differences and employee turnover, along with T&D programs' impact, can be examined using multiple theories. Generational Theories propose that variations in values and work styles among age groups influence turnover intentions (Parry and Urwin, 2011; Westerman and Yamamura, 2007; Kapoor and Solomon, 2011; Saorín-Iborra and Cámara-Campos, 2025). Human Capital Theory suggests T&D enhances employees' skills and affects retention (Becker, 1964). Social Exchange Theory posits that organizational investment in T&D fosters commitment across generations (Blau, 1964, 2017; Cropanzano and Mitchell, 2005). Generational theory, human capital theory, and social exchange theory collectively provide a framework for examining the effects of age diversity within the workforce and the influence of training investments on voluntary turnover.
2.2 Training and development strategies for age-diverse retention
Age disparities between employees influence turnover through organizational and interpersonal factors. Research shows that job satisfaction and work environment impact turnover across age groups (Lambert and Hogan, 2008). Workplace innovation and family policies affect retention based on life stages (Bass et al., 2018). Training initiatives mitigate turnover, with supervisor support correlating with reduced turnover intention (Ashar et al., 2013). Age-sensitive training programs must align with generational cohorts. Organizational strategies that enhance satisfaction can reduce age-related impacts. Diverse training strategies increase engagement across age groups (Al-Suraihi et al., 2021). A comprehensive approach combining age-sensitive T&D policies with HR strategies addresses generational differences. Kreamer et al. (2025) advocate a ‘diversity-general’ strategy embracing generational values. Climek et al. (2022) identified common and unique turnover factors across generations. Research shows millennials have similar turnover rates to earlier generations, with distinct workplace preferences (Cronin, 2018).
Generational differences influence employee turnover through organizational justice, burnout, and job satisfaction. Moon et al. (2024) found Generation MZ values equitable treatment, with procedural justice affecting turnover intentions. Lu and Gursoy (2013) showed emotional exhaustion impacts job satisfaction across generations. Research indicates millennials' lower job satisfaction affects turnover intention (Moreno et al., 2022; Chen et al., 2023). Age, marital status, and tenure affect turnover, with older and married employees showing lower turnover intention (Croes et al., 2024). Job security and team cohesion are essential for retention (Shinde, 2025). For Generation Z, emotional exhaustion reduces job satisfaction, while engagement enhances it (Adelia et al., 2024). Generational differences affect turnover through gaps in work values and communication styles. Age differences influence turnover intentions among younger staff (Mccarthy et al., 2020), creating less inclusive environments (Chen et al., 2023). Based on these arguments, the following hypothesis is proposed:
A greater age gap between younger and older employee groups is associated with higher voluntary employee turnover.
Conflict resolution and emotional intelligence bridge generational divides, while mentorship mitigates cognitive conflicts (I.Ilavarasi, 2024; Appelbaum et al., 2022; Doucet et al., 2009). Millennial turnover is influenced by intergenerational conflicts and labour market factors. Coaching helps millennials manage careers and reduce turnover (Minzlaff et al., 2024). For Generation Z, resonant leadership improves work performance through self-efficacy and organizational identification (Gaan and Shin, 2022). Younger generations prioritize work-life balance over employment stability, leading to ‘lie-flat' and ‘quiet quitting’ (Griep et al., 2025). Generational differences impact turnover trends, requiring targeted training programs using workforce analytics (Huselid et al., 2025). An inclusive HR strategy with engagement and tailored interventions maintains organizational stability across generations.
2.3 The impact of employee training and development on turnover rates
HRM practices enhance retention through training, development, engagement, job satisfaction, rewards, and employee participation (Al-Suraihi et al., 2021; Rabiul et al., 2021; Griep et al., 2025). Organizational commitment mediates the relationship between HRM practices and turnover intention (Kadiresan et al., 2015; Rawashdeh et al., 2022). Employee perceptions of managers correlate inversely with turnover intention, with training improving manager-employee relationships (Malek et al., 2018). Authentic leadership promotes knowledge sharing and supports retention in high-turnover sectors such as IT (Džambić and Hadziahmetovic, 2025). Retention strategies operate at multiple levels: job entry practices, employment practices, and organizational support reduce turnover intention (Wanyama et al., 2025). Data-driven talent management tools help to implement retention strategies (Nowak and Pawłowska-Nowak, 2024). Integrating T&D programs with HRM practices and organizational support enhances retention (Mallén-Broch et al., 2023). While training enhances skills and marketability, especially for younger employees, organizations must balance development with retention strategies, such as career development and competitive compensation. Management training improves satisfaction and reduces turnover (Choi and Dickson, 2009), whereas T&D programs can address age gaps (Ju and Li, 2019; Malek et al., 2018). Training influences the mobility of younger employees (Chen et al., 2023; Mccarthy et al., 2020; Veum, 1997). Programs that foster intrinsic motivation and align with career goals reduce turnover across age groups (Miao et al., 2020). Organizations that align development with employees achieve sustainable retention and improved performance (Kasdorf and Kayaalp, 2021). These considerations suggest that T&D programs may influence the relationship between workforce age diversity and employee turnover. Therefore, the following hypothesis is formulated:
The effect of the age gap on voluntary turnover is moderated by complex training and development programs.
T&D programs help reduce turnover and improve retention. Management training programs improve job satisfaction and commitment (Choi and Dickson, 2009; Malek et al., 2018). Gojeh et al. (2015) found limited career development increases turnover, whereas Vui Yee (2018) noted T&D alone doesn't reduce turnover intention. Performance management and compensation influence turnover. Training increases commitment because development opportunities enhance satisfaction and lower turnover intention (Ashar et al., 2013; Koster et al., 2011; Kasdorf and Kayaalp, 2021). Program effectiveness depends on design, delivery, and organizational culture (Duda and Žůrková, 2013; Bresk, 2023). Work design, job autonomy, and relationships influence retention, highlighting the need for tailored management practices (Moncarz et al., 2009; Al-Suraihi et al., 2021). Studies across the public and private sectors have revealed the impact of age gaps on turnover (Llorens and Stazyk, 2010). McGeachy and Daka (2025) suggested integrating T&D with retention strategies and competitive pay. However, training may also increase employees' external employability, particularly among younger workers. Accordingly, the following hypotheses are proposed:
A higher proportion of employees under 25 years of age predicts higher voluntary turnover.
The relationship between younger employees and voluntary turnover is moderated by complex training and development programs.
2.4 Training costs
Investments in training within organizations can affect employee turnover based on their alignment with employees' needs. Evidence shows that management training programs enhance job satisfaction and commitment, and reduce turnover (Choi and Dickson, 2009; Ashar et al., 2013). Management training strengthens manager-employee relationships and decreases turnover intention (Malek et al., 2018). Across sectors, various training types have been shown to decrease turnover by fostering satisfaction and engagement, while protecting organizational knowledge (Duda and Žůrková, 2013; Koster et al., 2011). However, high training costs can also increase employee turnover. Significant development investments create pressure for immediate returns, potentially increasing workload and stress, leading to departures if employees feel overwhelmed (Chakraborty et al., 2021; Tracey and Hinkin, 2008). Training expenditures can reduce compensation, diminish job satisfaction, and increase turnover intention (Al-Suraihi et al., 2021). Limited development opportunities may indicate restricted career growth, prompting employees to seek alternatives (Malek et al., 2018). The burden of justifying training expenses can affect work-life balance and increase turnover risks (Porter and Rigby, 2020). When employees view training as supportive and aligned with personal growth, it enhances satisfaction and commitment, mitigates costs, and reduces turnover (Kasdorf and Kayaalp, 2021). Integrating development programs into an organizational culture maximizes retention benefits. Considering the stabilizing role of older employees and the potential influence of training investments, the following hypotheses are proposed:
A higher proportion of employees aged 50 years and above predicts lower voluntary employee turnover.
The stabilizing effect of older employees is moderated by complex training and development programs.
The relationship between the age gap and voluntary turnover is moderated by training costs.
Training costs moderate the relationship between the proportion of younger employees and voluntary turnover.
Training costs moderate the relationship between the proportion of older employees and voluntary turnover.
3. Sample and methods
A methodology is adopted to test the hypotheses and examine the relationships in the literature. Descriptive statistics were used to outline the company characteristics. Multiple linear regression analysis was used to test the hypotheses.
3.1 Study population and sampling
The data utilized in this analysis were derived from the 2021 CRANET Survey, a cross-national investigation of HRM policies and practices (www.cranet.org). An international consortium designed and distributed the questionnaire, which was then translated and back-translated to ensure accuracy. Senior HR officials from organizations across 38 countries participated, with data compiled into an international dataset (for further details, see Parry et al., 2021). The author, a CRANET member, facilitated the data access. Only organizations with over 100 employees were selected, resulting in a sample of 5,899 organizations. CRANET, established in 1989 at the Cranfield School of Management (UK) and relocated to Penn State in 2021, now includes over 40 countries. The 2021–2022 survey covered 38 countries using a harmonized instrument with translations. The data collection methods varied by country, primarily using electronic surveys with supplemental personal visits or telephone interviews.
To develop the conceptual model, we used multiple linear regression to examine the influence of the age gap on employee turnover and the moderating effect of training programs. To mitigate confounding variables, we included controls for organizational structure, HR department presence, industry, sector, and firm size. We employed multiple linear regression to model the relationship between the dependent and independent variables, enabling the assessment of the impact of the age gap on turnover while accounting for other factors. Regression analysis was performed using ordinary least-squares estimation. Diagnostic tests confirmed the assumptions of linear regression, including normality, homoscedasticity and absence of multicollinearity.
The regression model is detailed in Equation (1) (Pituch and Stevens, 2015).
Y = Annual employees voluntary turnover
X = Age-gap
C1-5 = control variables: group membership, HR department, industry, sector, and number of employees
ε = random error
In our investigation of the influence of age disparities on voluntary employee turnover across 38 countries, we employed a moderation analysis to examine the potential moderating role of T&D methods in this context. Moderation analysis is crucial for determining whether the relationship between an independent variable (age gap) and a dependent variable (turnover) is affected by a moderator variable. To estimate these models, we utilized PROCESS v4.1, a computational tool compatible with SPSS, SAS, and R that employs ordinary least squares regression. By employing PROCESS v4.1, we evaluated the moderating effect of complex T&D techniques on the interaction between age gap and employee turnover.
In the next phase, we evaluated the impact of advanced T&D methods (W) on the correlation between age disparity (X) and employee turnover (Y). Multiple sub-models were developed for the analysis. Initially, we focused on the age-related differences. In the subsequent phase, we evaluated the effects of younger and older employees separately to gain a more nuanced understanding of the interrelationships among the factors. Finally, we analyzed the model with training costs (W1) as a specific and possible moderating factor. Consequently, we defined the X1 and X2 variables within the moderation model.
The regression model is as follows (Equation (2)):
Y = Annual employees voluntary turnover
X = Age-gap
X1 = Workforce under 25 years
X2 = Workforce 50 years and above
W = Complex training and development techniques
W1 = Training costs
X × W = interaction term of age gap (or workforce under 25 years/50 years and above) and complex training and development techniques (or training costs)
C1-5 = control variables: group membership, HR department, industry, sector, and number of employees
ε = random error
The interaction term examines whether the influence of age disparities on turnover is contingent on the extent of the advanced T&D methods.
3.2 Construction of variables
The CRANET survey includes a question regarding employee turnover, defined as follows: “Turnover is calculated as the percentage of the total workforce that has voluntarily left the organization in the past year.” The respondents were asked to provide the percentage measured on a scale. In our model, the age gap served as the independent variable. Consistent with the methodology employed by Mayrhofer et al. (2024) to assess the distance between items in the CRANET dataset, we applied a similar technique to analyze the age gap. The questionnaire incorporated two variables pertaining to employee age: percentage of workforce under 25 years and percentage of workforce aged 50 years and above. These variables are measured on a numeric ordinal scale, where ‘0' represents 0 “%, ‘1' indicates 1–5%, ‘2' corresponds to 6–20%, ‘3’ is 21–50%, and 4” is above 50%. The age gap was calculated by summing these two items, with a larger age gap indicating a greater disparity between the younger and older age groups. For instance, an age gap of eight suggests the absence of employees aged between 25 and 50 years, whereas an age gap of zero indicates a workforce composed solely of employees aged 25–50 years. A higher value signifies increased polarization within the organization, reflecting more pronounced differences among the various age groups. In contrast, a lower score indicates a less pronounced presence of extreme age distributions, characterized by a predominance of either younger or older employees within the organization (see Supplementary Table 1).
Complex T&D includes various employee training strategies, such as programs for older and younger employees, job enrichment, off/on-job training, developmental assignments, networking programs, career plans, assessment centers, succession planning, lateral moves, high-potential programs, international assignments, coaching, mentoring, e-learning, and career counselling. Components were rated from 0 (not at all) to 3 (very great extent), except for training for younger and older employees, which used a binary scale (0 = not used; 1 = used). The binary scale was converted to 0–3 for uniform weighting. Complex training development included 17 elements, with a maximum score of 51. Training expenditures were measured as a percentage of annual payroll costs on a metric scale.
4. Results
Cross-tabulation analysis of the workforce age composition revealed distinct patterns in the distribution of younger and older employees (see Figure 2 and Supplementary Table 1 and Supplementary Table 2). The results indicate that most companies demonstrate a diverse age range rather than skewed demographics. In the prevalent pattern observed among 997 organizations, 6–20% of employees were younger than 25, while 21–50% were aged 50 or older, suggesting a balanced representation of both younger and older staff members. Organizations with either a very small or very large percentage of young employees are uncommon; only 170 organizations have no employees under 25, and only 236 organizations have a workforce where more than half are under 25. Older employees are significantly represented in many organizations: 2,225 firms have 21–50% of their workforce aged 50 and above, and an additional 420 organizations have more than half of their employees in this age group, collectively comprising approximately 48% of the total sample. These trends underscore that most organizations maintain an age-diverse workforce.
A heat map titled ‘Workforce distribution percentage of organizations Under twenty-five vs fifty plus N equals five thousand eight hundred ninety-nine’ displays the distribution of workforce ages across different organizations. The x-axis represents the percentage of the workforce aged fifty plus, ranging from zero percent to over fifty percent. The y-axis represents the percentage of the workforce under twenty-five, also ranging from zero percent to over fifty percent. The color scale on the right ranges from dark purple to bright yellow, indicating the percentage of organizations, with yellow representing higher percentages. Notable data points include a concentration of organizations with sixteen point one percent to eighteen point two percent of their workforce in the age groups of one to five percent under twenty-five and six to twenty percent fifty plus.Workforce distribution by age groups (% of organizations). Source: own edition
A heat map titled ‘Workforce distribution percentage of organizations Under twenty-five vs fifty plus N equals five thousand eight hundred ninety-nine’ displays the distribution of workforce ages across different organizations. The x-axis represents the percentage of the workforce aged fifty plus, ranging from zero percent to over fifty percent. The y-axis represents the percentage of the workforce under twenty-five, also ranging from zero percent to over fifty percent. The color scale on the right ranges from dark purple to bright yellow, indicating the percentage of organizations, with yellow representing higher percentages. Notable data points include a concentration of organizations with sixteen point one percent to eighteen point two percent of their workforce in the age groups of one to five percent under twenty-five and six to twenty percent fifty plus.Workforce distribution by age groups (% of organizations). Source: own edition
4.1 The relationship between age-gap and turnover
First, Pearson's correlation analysis was conducted to examine the relationship between the aggregated age gap and annual voluntary employee turnover rate (see Supplementary Table 3). The results indicate a statistically significant positive correlation (r = 0.081, p < 0.001), suggesting that organizations with larger age disparities among employees tend to experience slightly higher voluntary turnover rates. Despite the statistical significance of the correlation, the effect size is small, implying that the age gap alone is not a strong predictor of turnover, and that other organizational or contextual factors are likely to exert a more substantial influence.
The OLS regression findings explain small but statistical significant relationship, with a positive correlation between the age gap and voluntary employee turnover (R2 = 0.007, F(6, 3,730) = 5.96, p < 0.001). The standardized beta coefficient (β = 0.086, t = 5.210, p < 0.001) indicates that greater age disparity correlates with increased turnover, supporting H1. Control variables, including corporate group membership, workforce size, industry, sector, and HR department presence, showed no statistical significance, aligning with Farndale et al. (2017) that T&D functions independently of context. These results confirm the impact of age difference on turnover (see Supplementary Table 4).
4.2 The moderation effect of complex training and development programs
Multiple regression analysis assessed the impact of age gap and complex T&D programs on employee turnover. The primary effect of the age gap was not significant (β = −0.557, p = 0.292), while complex T&D showed a significant negative effect (β = −0.305, p = 0.003). The interaction between age gap and complex T&D showed limited explanatory power, but it was significant (R2 = 0.0071, β = 0.095, p < 0.001), indicating that the age gap's impact depends on training complexity; as complexity increases, the negative effect of the age gap diminishes (see Supplementary Table 5). This supports H2a. Control variables did not significantly predict the outcome, aligning with Farndale et al. (2017) that T&D functions independently of the context. These results highlight the importance of complex training programs to manage workforce age differences. An analysis of employees under 25 and over 50 revealed different responses to T&D initiatives through varied turnover outcomes (see Figure 3).
The image contains two line graphs. The first graph, titled ‘Predicted turnover patterns: Under-25 share x Complex T&D,’ shows the predicted turnover relative to the under-25 share. The x-axis represents the under-25 share, ranging from low to high, and the y-axis represents the predicted turnover relative, ranging from 7.00 to 8.75. Three lines represent different moderator levels: low (blue solid line), medium (orange dashed line), and high (green dotted line). The lines show an upward trend, indicating that as the under-25 share increases, the predicted turnover also increases. The second graph, titled ‘Predicted turnover patterns: 50+ share x Complex T&D,’ shows the predicted turnover relative to the 50+ share. The x-axis represents the 50+ share, ranging from low to high, and the y-axis represents the predicted turnover relative, ranging from 13.5 to 16.0. The lines show a downward trend, indicating that as the 50+ share increases, the predicted turnover decreases.The interaction between age groups, turnover and complex T&D programs. Source: own edition
The image contains two line graphs. The first graph, titled ‘Predicted turnover patterns: Under-25 share x Complex T&D,’ shows the predicted turnover relative to the under-25 share. The x-axis represents the under-25 share, ranging from low to high, and the y-axis represents the predicted turnover relative, ranging from 7.00 to 8.75. Three lines represent different moderator levels: low (blue solid line), medium (orange dashed line), and high (green dotted line). The lines show an upward trend, indicating that as the under-25 share increases, the predicted turnover also increases. The second graph, titled ‘Predicted turnover patterns: 50+ share x Complex T&D,’ shows the predicted turnover relative to the 50+ share. The x-axis represents the 50+ share, ranging from low to high, and the y-axis represents the predicted turnover relative, ranging from 13.5 to 16.0. The lines show a downward trend, indicating that as the 50+ share increases, the predicted turnover decreases.The interaction between age groups, turnover and complex T&D programs. Source: own edition
A modified moderation analysis examined the influence of employees under the age of 25 (X1) on voluntary turnover (see Supplementary Table 6) with complex T&D as a moderator. Younger employees predicted higher turnover (β = 1.60, SE = 0.64, t = 2.49, p = 0.013), while the direct effect of complex T&D was not significant (β = −0.07, SE = 0.06, t = −1.13, p = 0.257). The interaction between the young workforce proportion and complex T&D was significant (β = 0.074, SE = 0.029, t = 2.59, p = 0.010), showing that training levels affect turnover. The analysis revealed stronger positive relationships at higher training levels: low (16th percentile): β = 2.27, p < 0.001; median (50th percentile): β = 3.01, p < 0.001; and high (84th percentile): β = 3.89, p < 0.001. These findings support H3a and H3b, showing that complex T&D programs increase turnover rates because they enhance employees' skills and market value, facilitating their transition to other organizations.
In the following sub-model, in which older employees were analyzed, we can state that a higher proportion of employees aged 50 and above (X2) is generally correlated with decreased turnover (β = −2.77, p < 0.001), supporting Hypothesis H4a. However, this relationship is moderated by the extent of complex T&D. When T&D is minimal, the negative impact of older workforce on turnover is pronounced (β = −1.85, p < 0.001), with moderate T&D diminishing (β = −0.84, p = 0.0096), and with extensive T&D, it becomes negligible (β = 0.38, p = 0.408). While a larger proportion of older employees typically leads to reduced turnover, comprehensive T&D can mitigate this stabilizing effect. These results support Hypothesis H4b.
These findings indicate that workforce age composition and training complexity jointly influence employee turnover. Although age differences alone do not consistently predict turnover, the presence of both younger and older workers exerts distinct effects that depend on the scope and complexity of T&D initiatives. Younger workers were more inclined to leave, particularly in environments with more demanding training. Older workers generally contribute to workforce stability, although this stabilizing effect may be diminished through comprehensive training. These trends suggest that demographic factors influencing turnover should be considered alongside organizational measures such as training. Consequently, strategies for workforce planning and development should consider age-related responses to training programs to manage retention effectively and reduce voluntary turnover.
4.3 The moderation effect of training costs
We analyzed the moderating effect of training costs on the age gap and employee turnover, focusing on training expenses that increase with turnover due to new employee training needs. Moderation analysis examined how training expenditure (W) influences the relationship between workforce generational differences (X) and voluntary turnover (Y). The model showed statistical significance (R2 = 0.120, p < 0.001), explaining 12% of turnover variance (see Figure 4 and Supplementary Table 7). While generational differences alone did not significantly affect turnover, higher training expenditures were correlated with increased turnover. The interaction between generational differences and training expenditure is significant, supporting Hypothesis H2b. The analysis showed that generational differences' effects on turnover strengthened with higher training expenditures. These findings suggest that, in organizations with large training budgets, generational gaps may increase turnover, possibly because training programs are not equally effective across generations or fail to address value gaps. The results emphasize the need to customize training programs for a multigenerational workforce to reduce turnover risk.
Two line graphs showing predicted turnover patterns. The first graph illustrates the relationship between the under-25 share and predicted turnover relative to training costs, with three lines representing low, medium, and high moderator levels. The second graph depicts the relationship between age-gap and predicted turnover relative to training costs, also with three lines representing low, medium, and high moderator levels. In both graphs, higher moderator levels correspond to higher predicted turnover. The lines show a positive trend, indicating that as the under-25 share and age-gap increase, the predicted turnover also increases. All values are approximated.The interaction between age groups, turnover and training costs. Source: own edition
Two line graphs showing predicted turnover patterns. The first graph illustrates the relationship between the under-25 share and predicted turnover relative to training costs, with three lines representing low, medium, and high moderator levels. The second graph depicts the relationship between age-gap and predicted turnover relative to training costs, also with three lines representing low, medium, and high moderator levels. In both graphs, higher moderator levels correspond to higher predicted turnover. The lines show a positive trend, indicating that as the under-25 share and age-gap increase, the predicted turnover also increases. All values are approximated.The interaction between age groups, turnover and training costs. Source: own edition
An analysis of employee behaviours under the age of 25 within the training-turnover relationship framework reveals significant insights. The interaction between training expenses and the percentage of younger employees influences voluntary turnover rates. The model shows that a higher proportion of young employees increases turnover (B = 1.414, p < 0.0001), and that higher training costs correlate with increased turnover (B = 0.191, p = 0.0003). The interaction between these factors was significant (B = 0.089, p = 0.0001). When the proportion of young employees was low, training slightly increased turnover (B = 0.279, p < 0.001). At medium and high levels, the effect strengthened (B = 0.368 and B = 0.456, respectively; both p < 0.001). This supports H3c. These findings indicate that increasing training expenditure, particularly in organizations with many younger employees, may lead to higher turnover if not properly targeted.
The sub-model examines the impact of training expenses on employee turnover, focusing on employees aged over 50 years. The model was statistically significant (R2 = 0.1214, F(3, 3,287) = 151.41, p < 0.001), with independent variables explaining 12% of the turnover variance. The results showed that a higher proportion of older employees correlated with reduced turnover (B = −1.2231, p = 0.0002). However, increased training expenses (B = 0.3597, p < 0.001) contributed to higher turnover rates. The interaction effect was not significant (B = 0.0242, p = 0.2743), showing that employee age proportion does not affect the training costs-turnover relationship, thus not supporting H4c.
5. Discussion
This study investigated the relationship between workforce age disparities, and voluntary turnover, with a focus on complex training and development programs. Results showed a modest correlation between age differences and voluntary turnover, while their interaction with complex T&D programs significantly affects outcomes. In organizations with complex T&D programs, the impact of age differences becomes positive, influencing generational dynamics. The findings indicate that a higher proportion of employees under 25 increases turnover, particularly with complex training, as training enhances external market attractiveness. A larger proportion of employees aged 50 years and above have reduced turnover, although this effect weakens with comprehensive T&D programs. Higher training expenditures correlate with increased turnover, especially among young employees, revealing that training can increase the likelihood of departure if not aligned with generational needs. Table 1 illustrates the main impact of T&D on the relationship between age groups and employee turnover.
Summary of the T&D impact on different age groups and turnover
| Age group | Positive Impact of T&D | Negative Impact of T&D |
|---|---|---|
| Younger Employees (<25) | Improves skills, engagement, career prospects; supports mentoring and career planning | Increases marketability and risk of voluntary turnover if training intensity is high |
| Older Employees (50+) | Enhances loyalty, knowledge transfer, and adaptability to new technologies | Complex training may reduce retention benefits; adaptation challenges and high costs possible |
| Employee Turnover (Overall) | Well-designed T&D reduces turnover, strengthens cohesion, and fosters retention when integrated with HR policies | Misaligned or overly intensive T&D may increase turnover, especially among younger employees |
These findings support the hypotheses and are consistent with generational workforce theory, which suggests that different age cohorts respond differently to organizational practices. The training costs significantly moderate the relationship between younger employees and voluntary turnover. Higher training expenditures intensify turnover among younger workers. While training costs showed a positive main effect on turnover, there cannot be found significant interaction with older employees. These results show that training costs have age-contingent effects, suggesting a need for targeted generational development investments. Overall, this study responds to the research objectives by examining how workforce age diversity and training investments jointly influence voluntary employee turnover across organizations. By addressing the previously limited empirical evidence on the interaction between demographic workforce structures and HRD practices, the study highlights that training investments do not automatically reduce turnover. Instead, their impact depends on the generational composition of the workforce and the design of development programs. The results indicate that employee turnover is not directly determined by a single factor, rather, it emerges from the simultaneous configuration of multiple variables. Among these, the age composition of the workforce and the availability of T&D opportunities play a particularly significant role. These insights contribute to both theory and practice by demonstrating the importance of aligning HRD strategies with workforce demographics in order to enhance organizational stability and employee retention.
5.1 Theoretical contributions
This study addresses the research gap by offering cross-national empirical insights into how workforce age diversity and training investments interact to influence voluntary employee turnover. Unlike previous studies, which have mainly focused on either generational workforce structures or T&D practices in isolation, this research integrates these elements within a cohesive analytical framework. This approach enriches the HRM and HRD literature by showing that the impact of training investments depends on the age composition of the workforce, thereby providing a more nuanced theoretical understanding of the interconnections between workforce demographics, employee development, and retention. Our findings contribute significantly to HRM and organizational behaviour. It expands the workforce demography understanding by showing that age composition interacts with training practices and cost structures. These findings demonstrate that training investments do not always improve retention rates. This study advances generational workforce theory by revealing different patterns between age groups: younger employees show increased mobility after training, while older employees generally promote stability, although training can disrupt this. By integrating the age gap, training complexity, and costs into a moderation framework, this study provides a comprehensive model for understanding turnover, showing how organizational investments interact with workforce characteristics to influence retention strategies.
5.2 Practical implications
In response to the research gap identified in the study, the results hold considerable importance for HR experts, managers, and policymakers engaged in workforce development. Companies should consider employee age composition when designing training strategies. Training can boost mobility among younger workers while fostering stability among older employees. To retain trained staff, businesses should align training with retention-oriented practices like internal career paths and mentoring. Organizations can use workforce analytics to track age diversity and turnover trends to tailor development programs. These approaches balance skill enhancement with workforce stability, especially in large firms where demographics and training influence retention. The findings warn against assuming training reduces employee turnover. In settings with many young employees, extensive training might encourage departure by providing marketable skills. Organizations should implement retention strategies like career progression and financial incentives. The stabilizing effect of older employees contributes to organizational stability, though their training should be tailored. Programs facilitating knowledge transfer between generations can harness both groups' strengths. These findings show the need for varied training strategies accounting for generational differences. Organizations should adopt targeted methods recognizing unique motivations of younger and older employees to improve effectiveness and minimize turnover.
5.3 Limitations and future research
The cross-sectional design limits the establishment of causality among age composition, training practices, and turnover. Longitudinal studies are required to understand the temporal evolution of turnover decisions. Reliance on organizational-level data restricts insights into individual factors such as employees' perceptions of fairness and satisfaction. Future research could integrate organizational data with interviews to explore turnover motivations. The generalizability of the findings is limited because the sample includes specific organizations within one national context. Studies across countries can reveal how cultural and market factors influence age composition, training, and turnover relationships. While this study examined age and training variables, future models could include compensation schemes and leadership practices to better understand the turnover dynamics. While the CRANET dataset offers cross-national organizational data, it captures HR practices at a specific point in time. The 2021 CRANET survey represents the most recent iteration, with the subsequent round scheduled for 2026. Since 2021, labour markets have undergone significant transformations due to remote work, digitalization, and evolving employee expectations, which may influence turnover, particularly among younger workers. Nevertheless, current trends indicate that the relationship between workforce age, training opportunities, and turnover remains pertinent. The dynamics examined continue to provide valuable insights into employee retention. Future research could employ more recent datasets to investigate how evolving labour conditions impact age diversity, training, and voluntary turnover.
Ethical considerations
The study utilized data from the international CRANET survey, which collects information on human resource management practices at the organizational level. Participation in the survey was entirely voluntary, and all respondents were informed about the purpose of the research, the anonymity of their responses, and the organizational (not personal) nature of the data collected. No personally identifiable information was obtained, and all data were aggregated and anonymized prior to analysis.
The research was conducted in compliance with international research ethics standards, as well as the relevant international procedures and the applicable European Union and national guidelines and regulations. Based on the Research Regulations of the University of Pécs and the Code of Ethics of the Faculty of Business and Economics of the University of Pécs, and considering the method, content, and data used in the study, the research qualifies as minimal risk. Therefore, according to these regulations, a formal research ethics procedure and a separate ethics approval were not required.
Consent to participate
Participation in the CRANET survey was voluntary. All respondents were informed about the purpose of the research and the anonymous, organizational-level nature of the data collection. By completing the questionnaire, participants provided implied informed consent to participate in the study.
Consent for publication
The study used anonymized, organizational-level data from the international CRANET survey. No personal or identifiable information about individual participants was collected or published. Therefore, informed consent for publication was not required.
The author would like to thank the CRANET research network and all country coordinators who contributed to the 2021 dataset used in this study.
The supplementary material for this article can be found online.
