Are developmental performance management systems inclusive? An intersectional analysis for age and gender Open Access

The progressive feminisation of the labour market together with the ageing process of the working population, constitutes a topic of great interest for organisational scholars and practitioners in human resource management. These labour market changes generate relevant implications for one of the core practices of human resource management (i.e. performance management), rising issues of how performance management systems might influence occupational wellbeing for employees with different sociodemographic characteristics. Against this background, the present research aims to enrich the growing body of literature on the gendered nature and outcomes of apparently neutral organisational practices, following Acker’s footsteps in the study of organisational gender inequalities through an intersectional framework (Acker, 2006 and, 2012; Healy et al., 2019).

Performance management is a central practice in human resource management, designed to align individual and team performance to organisational priorities (Brown et al., 2019). To this end, systems are recognised in literature, and constructed in this paper, as being composed of the following components: goal setting, evaluation criteria, evaluation process, feedback practices and finally rewards.

According to a part of literature, performance management represents an elective field for the study of gendered organisations (Acker, 1990; Van Dijk et al., 2020; Zanoni et al., 2010) as it might obscure and legitimise systems of inequalities by giving them a semblance of meritocracy, and ultimately negatively affecting employee occupational wellbeing (Wilks and Neto, 2012).

This perspective can be challenged by the fact that the renewal of performance management systems’ designs is becoming a priority for an increasing number of companies in their process of moving away from traditional performance management practices, which tend to adopt as evaluation criteria the attainment of individual and short-term performance targets, towards development based approaches, whose focus is employees’ skills and competencies enhancement as a tool to achieve a fairer distribution of training and career opportunities. While research on developmental performance management systems (hereafter DPMS) constitutes a literature gap in itself, as research on competency based performance management systems is still in its prodromal state (Brown et al., 2019). DPMS are regarded by recent literature (Ganji et al., 2023) as a possible inclusive performance management practice, as pursuing the development of competencies and skills (Hing et al., 2023) could lead to the promotion of women to senior leadership, thereby increasing the level of equity (Hing et al., 2023).

Moreover, considering how women have traditionally been withheld from career opportunities (Hoobler et al., 2011), development programmes might considerably improve their occupational wellbeing (Hopkins et al., 2008).

However, the examination of a rather complex phenomenon, such as imbalances in wellbeing outcomes fostered by DPMS, might require a deeper investigation than a single-axis perspective on gender, as prompted by feminist research (Acker, 2011; Collins, 1992; Crenshaw, 1991) which brought attentiveness towards the notion of intersectionality: the concept that processes of gender inequality do not stand in isolation, but are rather intersected and influenced by other forms of exclusion. The introduction of intersectionality suggests the joint analysis of one or more intersecting segments of difference (i.e. gender and age, as in Martin et al., 2019) may reveal dynamics not found by the analysis of a single segment of identity (Thrasher, 2022). However, organisational studies that look at several inequality processes simultaneously are rare (Acker, 2011).

Among the possible dimensions considered to adopt an intersectional perspective (i.e. gender identity as in Suarez et al., 2020; nationality as in Abane and Phinaitrup, 2019 and disability as in Pagan and Malo, 2020), this study considers age, a traditionally neglected dimension in the analysis of gender inequality (Acker, 2011). Age, however, constitutes a central variable in the experience of occupational wellbeing and in the type of feedback workers receive with respect to the level of enjoyment of a system with emphasis on the development of their skills (Brown et al., 2019; Kooij et al., 2012; Innocenti et al., 2013).

In doing so, we address Cleveland et al.’s (2017) call for the development of intersectional human resource management as they claim that “Programs for employee development might be practices where intersectionality matters substantially” (p. 133) extending on intersectional research by focusing on age, a variable potentially able to unveil the gendered nature of DPMS and its wellbeing outcomes.

The paper is structured as follows: Section 2 presents the theoretical background and the development of the hypothesis of the study. Then, we describe the data and methods in Section 3, while empirical results are presented in Section 4. Section 5 provides the discussion of theoretical and managerial implications, including limitations and avenues for further research.

There is recognition in organisational and managerial literature that performance management might not be gender neutral. For instance, during appraisal processes, the assessor might be influenced by potential biases and translate behavioural performance into gendered scores (Acker, 2006; Fenech et al., 2021; Van den Brink et al., 2010). The universal gender stereotypical norms and the tendency to penalise individuals who violate them may result in a worse distribution of opportunities, systematically disadvantageous towards women within organisations (Heilman et al., 2004), leading them to adapt to male stereotypes and expectations (Brinck et al., 2019). Recognising the gendered dimension of performance management is therefore pivotal in analysing equity in organisations to prevent female marginalisation and exclusion (Ely and Meyerson, 2000). In this paper, we argue that a key analytical lens to do so is framing gendered performance management analysis in Acker’s work. To the best of the authors’ knowledge, in fact, Acker’s framework is one of the main perspectives to analyse the issue of gender discrimination as an organisational concern rather than an individual one (Britton and Logan, 2008). In Acker’s perspective, inequalities are built into job design, wage determination, distribution of decision-making and supervisory power, workplace design and rules, both explicit and implicit, for behaviour at work, conveyed to workers through performance management systems (Acker, 2012).

In depicting her framework, Acker uses four major notions “gendered organisation”, “inequality regimes”, the “ideal worker” and “intersectionality” (Nkomo and Rodríguez, 2018) to illustrate how masculinity constitutes one of the cornerstones on which organisations are constructed.

Firstly, to say that organisations are gendered is to say that: Advantage and disadvantage, exploitation and control, action and emotion, meaning and identity, are patterned through and in terms of a distinction between male and female, masculine and feminine (Acker, 1990, p. 146). To understand organisations as gendered is therefore to recognise how gender is an embedded and constitutive element of all organisational practices (Acker, 2006), rendering the gendered analysis of organisational processes essential to understand the status of women in organisations. The outcome of gendered processes within organisations are inequality regimes, defined as: Systematic disparities between participants in power and control over goals, resources, and outcomes on workplace decisions such as how to organize work; opportunities for promotion and interesting work; security in employment and benefits; pay and other monetary rewards; respect; and pleasures in work and work relations (Acker, 2006, p. 443).

A supplementary concept useful for understanding Acker’s framework is represented by the ideal (or universal) worker, as in most cases work is arranged on the basis of expectations that incorporate the image of an unencumbered worker (Acker, 2009) totally dedicated to work without any responsibility for family care, and thus most likely to be a man, whose partner takes care of everything else (Acker, 1992). The masculine image of the ideal worker is crucial in the analysis of discrimination because its legitimacy results in privileges for men, marginalisation for women and the maintenance of gender segregation in organisations. In this context, performance management can act as significant mechanisms to reinforce gender inequality (Bear et al., 2017), especially traditional performance management systems where employees’ performance is evaluated against a predetermined set of individual short-term performance targets, reflecting the achievements of the ideal worker, conceived as a white, heterosexual, able-bodied, cisgender man (Vance et al., 1992).

Lastly, Acker herself recognises the necessity of intersectional research, as she argues how a gender analysis is incomplete if it ignores racial and class processes, both of which are also central to the ongoing reproduction of inequality (Acker, 2012). Nevertheless, in his work, Acker largely concentrates on the intersection of gender with class and race, overlooking the role of age, which she herself recognises as a possible basis for discrimination (Acker, 2011), despite having generated less research. In adopting Acker’s framework, we further fill a literature gap, as according to Nkomo and Rodríguez (2018), Acker’s framework has rarely been fully applied.

Recently, a small set of studies on performance management is focusing on the developmental approach whose main focus is on employees’ experiences, knowledge and skills (Boswell and Boudreau, 2002; Brown et al., 2019). To depict such systems, existing research draws attention to the following components: goal setting and evaluation criteria, appraisal and feedback practices and rewards.

DPMSs are characterised, according to Roberts (2003), by a participatory nature, which characterises the goal-setting phase, the evaluation phase (through worker self-assessment) and the feedback phase, through two-way communication in the evaluation interview. In DPMS, criteria for performance evaluation are centred on team results, which reinforces interdependencies between workers, knowledge sharing and therefore leads to the development of new skills, ultimately fostering improved perceptions of occupational wellbeing. Moreover, evaluation criteria in DPMS focus on employee skills and competencies (Bayo-Moriones et al., 2020) relating to three main domains: job-related competencies, competencies not required by job description and finally, human qualities or personality (Kubiak, 2022). Concerning appraisal and feedback practices, in DPMS, evaluation is iterative, and frequent (Hajnal and Staronova, 2021; Kubiak, 2022), so as to foster employee’s self-improvement and development at work (Bayo-Moriones et al., 2020) and wellbeing (Kubiak, 2022). Moreover, with the final objective of ensuring more fairness in evaluations, DPMS feature 360° feedback systems, where employees are evaluated by multiple assessors (Hopkins et al., 2008; Roberts, 2003) Finally, rewards in DPMS are represented by both career advancement opportunities or training paths (Hajnal and Staronova, 2021).

DPMS are accordingly represented by some scholars as a possible inclusive measure (Ganji et al., 2023) because by obtaining new context specific competencies, not acquirable outside given organisational settings, women might gain a better position to negotiate improved working conditions, job opportunities and incentives, more in line with their wellbeing needs (Hing et al., 2023). Consequently, the impacts of DPMS might be more favourable to women because, by focusing on skills development, these systems can remove the obstacles that women, more than men, face in accessing attractive career opportunities (Hind and Baruch, 1997). However, critical literature warns that theoretically inclusive development systems may result in non-inclusive practices (Gedro and Mizzi, 2014), due to the fact that gender-related effects can be further complicated by intersections with other dimensions. The need for intersectional analysis is also called out by scholars (Dennissen et al., 2020; Holvino, 2008; Rodriguez et al., 2016) to address simultaneously gender, race, ethnicity, age, class, disability and sexuality in organisations (Woods et al., 2021). Prior studies on the relationship between performance management and wellbeing already provide evidence of the key role of several individual identity characteristics (see, for example, Suarez et al., 2020 for gender identity, Abane and Phinaitrup, 2019 for nationality or Pagan and Malo, 2020 for disability). In this research, we particularly addressed the issue of age. In fact, age constitutes a central variable in the experience of occupational wellbeing (Warr, 1992; Van Dijk, 2020), and, more importantly, it is a fundamental identity category in shaping employee’s reaction to DPMS, a system whose main focus is the development of their skills (Brown et al., 2019; Innocenti et al., 2013; Kooij et al., 2012). Furthermore, literature indicates that men born more recently tend to do more housework than men of other cohorts (Leopold et al., 2018). Rebalancing care burdens between genders could have significant implications in this context, as it could allow women to engage more in educational activities, thus improving their effective access to development opportunities.

Against this background, Wilks and Neto (2012) take a step further in intersectional research showing that the impact of DPMS on occupational wellbeing is explained by the intersection of age and gender rather than by gender itself. More specifically, they provide evidence that older women’s wellbeing is jointly affected by age and gender, making them doubly disadvantaged and thus victims of what they define as a double jeopardy effect. Hence, the results question whether DPMS may generate different wellbeing outcomes on people pertaining to different intersections of gender and age.

In conclusion, the above studies suggest that the effect of DPMS on workers’ wellbeing may vary across different intersections of gender and age and lead us to formulate the following hypothesis:

The conceptual model of the relationship between the dimensions considered is depicted in Figure 1.

To test the research hypothesis, we collected data by administering an online survey between the end of June and the end of July 2022 to a sample of Italian employees working in private businesses. The sampling strategy was based on the convenience sampling method (Etikan et al., 2016), a non-probabilistic sampling widely used among scholars (Etikan et al., 2016; Mehta, 2021), particularly useful when dealing with limited resources. In particular, the main selection technique was accidental sampling (Etikan et al., 2016), where the target population is selected for meeting certain practical criteria such as accessibility, geographical proximity or willingness to participate.

The survey was composed of two main sections, one referring to demographic characteristics of respondents, and the second, concerning dimensions of interests. Data were collected through the platform Alchemer, and respondents were invited to complete the survey through:

• email contacts (a reminder was sent every 10 days);

• snowball techniques (by sharing the link with researchers’ private contact lists); and

• targeted social media campaigns.

Participants were required to meet two main criteria: being an employee and working in the private sector.

Overall, the survey was sent to a database of roughly 10,000 employee’s contact. Out of these, we received 1,052 responses, of which 647 were usable (61.50% of the total sample) since employees declared that performance management systems were used in their organisation.

Regarding the individual characteristics of the respondents, 75.00% identified as male, 17.90% of the surveyed were younger than 34 years of age, while 57.20% were between 35 and 49 and the remaining 24.90% were older than 49 years of age. Concerning the educational level, the majority of respondents (53.80%) hold a master’s degree or a PhD. Rolewise, the majority of respondents (58.60%) worked as clerks mainly with a low level of seniority (46.40%) working in large companies (62.00%) operating in the service sector (52.90%). The full description of the sample’s characteristics is summarised in Table 1. The descriptive statistics highlight how performance management systems are more often adopted in medium and large organisations, confirming prior literature (Curzi et al., 2019) and national statistics, which recognise how large and medium companies are more often located in northern Italy.

Nonetheless, we adopted a post hoc weighting procedure to make the results more robust and tackle the issue of potential biases related to the use of convenience sampling, which could either under- or over-represent certain demographic subgroups, such as male and female or young, adult and senior employees (see Section 3.3).

We adopted as independent variable a DPMS index on the basis of how close the purpose of the company performance management system was to employees’ development (Kubiak, 2022), with the aim to enhance the workers’ skills and competencies (Brown et al., 2019; Innocenti et al., 2013; Kubiak, 2022; Kuvaas, 2006). According to literature, we selected nine different dimensions of DPMS to indicate whether the performance management system adopted was in fact developmental, with regard to three main areas: goal setting and evaluation criteria, appraisal and feedback practices and rewards. Among goal setting and evaluation criteria, we considered employee involvement in goal setting (Roberts, 2003) as well as the use of the following evaluation criteria: collective results (Brown et al., 2019 and Roberts, 2003), employee competencies, either required or not by the job description (Bayo-Moriones et al., 2020) and finally human qualities or personality (Kubiak, 2022).

Among appraisal and feedback practices, we considered the use of 360° feedback (Hopkins et al., 2008; Kubiak, 2022; Roberts, 2003), as well as frequent feedback (Hajnal and Staronova, 2021; Kubiak, 2022). Finally, rewards included both career advancement opportunities and training paths (Hajnal and Staronova, 2021). The dimensions of the index are summarised in Figure 2.

Each dimension was measured through a dummy variable, indicating with 1 if such dimension was used in the organisation, or 0 if it was not. The final index was created as a sum of the different dimensions, and it ranged from 0 (no dimension) to 9 (all dimensions).

We adopted as dependent variable the level of employee occupational wellbeing. The variable was measured following Warr’s (1990) scale on employee’s perception of positive and negative feelings at work. Hence, the wellbeing construct was based on the analysis of 12 items ranging from 1 (never) to 5 (always) with questions such as “To which extent do you associate happiness, optimism, calmness to your work?” or negative feelings, using items such as “To which extent do you associate anxiety, depression or sadness to your job?”. The final construct was computed as a mean of the items (Alpha 0.891).

In the analysis, we also considered gender and age as potential moderators of the relationship between DPMS and work-related wellbeing. Gender was measured through a dummy variable with “1” indicating male respondents and with “0” female respondents.

Following Warr (1992), age was categorised into three classes:

1. Youngsters: 34 years old or younger;

2. Adults: between 35 and 49 years old; and

3. Seniors: over 49 years old.

We created a categorical variable with “0” for younger respondents, “1” for adult respondents and “2” for senior workers.

We also controlled the analysis for employees’ characteristics (role, seniority, education, contract’s type) and firms’ characteristics (size, sector and multinationality), following previous scholars (Curzi et al., 2019).

We used a convenience sample. Although they are widely used in literature (Haddad et al., 2022), they might under- or over-represent demographic subgroups, introducing biases that could distort the accuracy of the results (Royal, 2019). Accordingly, we performed a post hoc weighting procedure using the two moderating demographic variables (i.e. gender and age) to correct the sample following Royal (2019) and Haddad et al. (2022), chosen because of their centrality in the study. The weighting was built on data taken from the Italian National Institute of Statistics (ISTAT) considering gender and age workforce distribution in private companies operating in northern Italian regions at NUTS 2 regional level (Nomenclature of Units for Territorial Statistics). As outlined by Table 2, after the correction, the workforce distribution by gender and age in the weighted sample is closer to the one of the reference population. Therefore, the weighting procedure allowed to have more balanced subgroups, making the analysis more accurate.

Moreover, two robustness checks were carried out. Firstly, we checked for gender and age distribution of workforce across sectors. Tables 3 and 4 show that, when weighted, the distribution of the sample is closer to the distribution of the reference population also across sectors.

Secondly, we carried out T-test and analysis of variance (hereafter ANOVA) to look for significant differences in terms of average levels of occupational wellbeing and access to DPMS across genders and ages.

Both T-tests and ANOVA highlighted that significant differences can be found in terms of gender and ages. For instance, women report a significant lower level of wellbeing (M_W = 3.22, SD_W = 0.70) than men (M_M = 3.44, SD_M = 0.66), t (645) = −3.93, p ≤ 0.01.

At the same time, ANOVA shows that significant differences can be found in terms of access to DPMS across ages: young workers (M_Y = 4.52, SD_Y = 1.97), adult (M_A = 3.99, SD_A = 2.03) and senior (M_S = 3.83, SD_S = 2.00), F(2,644) = 4.21, p ≤ 0.05.

Finally, we checked for common method bias using Harman’s test (Podsakoff et al., 2003). The results indicated that significant common method bias was not present in our data, as the variance extracted by a single component was 34.67%, which was far below the threshold of 50%.

To perform the analysis, we built three different regression models to test our hypothesis. In the first model, we regressed the DPMS index on employee wellbeing (controls included) to analyse the effect that DPMS have on occupational wellbeing. The second model included gender as interaction term to test whether there were significant differences in the effect of DPMS between men and women on occupational wellbeing. In the third model, we finally performed a three-way interaction by including age (using two dummies variables with youngsters as the reference category) and its interactions with DPMS and gender to test whether gender differences in the relationship between DPMS and occupational wellbeing were contingent upon age. To conclude the analysis, as an additional robustness check, we carried out a bootstrap parametric test with 1,000 resampling at 95% confidence interval for all three models. Bootstrapping is a resampling technique which involves the repeated sampling with replacement from the original data set to create multiple “bootstrap” samples. The procedure generates more accurate and robust results by providing confidence intervals estimation, and mimicking large samples (He et al., 2024). The bootstrap test confirmed the finding of the study as outlined in the result section.

Table 5 presents the descriptive statistics of the main variables of the model. Most variables are significantly correlated, and the descriptive statistics exclude the multicollinearity issue.

Table 6 shows the models of the regression analysis. In Model 1, the positive and significant effect of DPMS on employee wellbeing (β = 0.117; p ≤ 0.01) is evident. The fit of the model is adequate, since the adjusted R² is explained at 12.75%.

In Model 2, we tested whether the positive association between DPMS and perceived level of occupational wellbeing differ between male and female employees. The interaction is not significant (β = 0.026; p ≥ 0.10), therefore gender per se does not moderate the relationship between DPMS and perceived occupational wellbeing. What we do find in Model 3 instead, is a significant interaction between DPMS, gender and different classes of ages, in particular that of adults (β = 0.167; p ≤ 0.05) and seniors (β = 0.169; p ≤ 0.05). Therefore, there is a three-way interaction between DPMS, gender and age on perceived wellbeing at work, hence our hypothesis is supported. The findings were also confirmed by the bootstrap test, which highlights how in Model 1 we find DPMS β = 0.117 [C.I. 0.092–0.143], in Model 2 DPMS*Gender β = 0.026 [C.I. −0.026 to 0.075] and in Model 3 DPMS*Gender*Adult β = 0.0167 [C.I. 0.026–0.316] and DPMS*Gender*Senior β = 0.0169 [C.I. 0.020–0.323].

To understand how age influences the effect of gender in the relationship between DPMS and occupational wellbeing, we estimated the conditional effect for all combinations of genders and ages, as shown in Table 7. In addition, following Paternoster et al. (1998) we tested whether those conditional effects were significantly different across different combinations of age and gender, as shown in Table 8.

As pointed out by Table 8, younger women experience significantly higher levels of occupational wellbeing than both young and adult men, due to the use of DPMS, while adult women experience significantly lower levels of occupational wellbeing associated with the use of DPMS when compared to senior men.

This study applied an intersectional perspective to provide a new point of view on how DPMS influence employee wellbeing. Specifically, we investigated the interaction between gender and age in shaping the relationships between DPMS and wellbeing. Our results showed that gender by itself does not significantly affect the positive association between DPMS and wellbeing. However, a significant three-way interaction revealed that the relationship between gender and DPMS outcomes was influenced by age. In the next section, we delve deeper into these findings and discuss their implications for both theoretical understanding and practical application.

The results of this research contribute to two strands of literature: the one on the inclusive potential of DPMS, and the other on inequality regimes in organisations.

Regarding the first matter, the adoption of an intersectional perspective allows us to shed new light on the difference between developmental and traditional performance management systems. The latter have, in fact, already proved to be significant mechanisms to obscure and legitimise organisational inequalities (Acker, 1990; Van Dijk et al., 2020; Zanoni et al., 2010) capable of negatively influencing women’s occupational wellbeing (Wilks and Neto, 2012). The effects are, however, not so unequivocal in the case of DPMS. Our results show that these systems produce positive outcomes for employee wellbeing, but with significant differences across different combinations of age and gender. Specifically, young women (under 35 years of age) are particularly favoured, reporting significantly higher levels of wellbeing not only compared to any other group of women, but also to both men of the same age and older. Women, traditionally excluded from career opportunities (Hoobler et al., 2011), may perceive chances provided by DPMSs as a solution to gendered obstacles, thus explaining gender differences among young employees.

At the same time, women younger than 35 are less likely than both men and women aged 35–50 (the adults of our sample) to have caring responsibilities (Del Boca et al., 2020), making younger women more willing to prioritise participation to developmental opportunities over familiar commitment (Cleveland et al., 2017).

Adult women (between the ages of 35 and 49) experience a smaller effect on wellbeing from adopting DPMS. In particular, such effect is not significant when compared to men of the same age group, but it is significantly smaller than the one experienced from both senior men (aged 50+) and younger women (aged under 35).

In essence, unlike traditional performance management systems, DPMS have an inclusive potential (Ganji et al., 2023; Hoobler et al., 2011; Hopkins et al., 2008), which, however, favours in particular young women above other intersections such as adult women. This shows how the use of age in a gendered intersectional analysis is crucial to reveal inequality regimes on the use of DPMS which would otherwise remain invisible. We believe this is a major contribution to extant literature claiming gender single axes analyses (Collins, 1992) do not allow to identify systems of inequalities, confirming the centrality of intersectionality in organisation and management analysis (Acker, 2006; Cleveland et al., 2017; Thrasher, 2022).

A second theoretical contribution is to the literature strand of inequality regimes by Acker (2006). In particular, it advances the understanding of the so-called “ideal worker” in the specific context of DPMS and the evolving expectations of employers in contemporary work organisations.

Since Acker’s seminal work (1990), the ideal worker has been conceived as a white, heterosexual, able-bodied, cisgender man (Vance et al., 1992). This image appears to be well reflected by traditional performance management systems, in which workers are expected to pursue individual short-term goals, thereby reflecting the masculine norm emphasising assertiveness, competitiveness, success and individualism at work. However, Acker (2006, 2012) highlights that the category of ideal worker is not established once and for all, but tends to shift across organisational contexts and over time. In this sense, our research makes a contribution to this line of investigation, outlining how in developmental systems the ideal worker is shifted to reflect the changing expectations of organisations at least for jobs requiring intellectual and administrative activities. In these systems, the focus is in fact no longer only on the performance of the job and the achievement of results, but it also becomes important how the job is performed and the results are achieved.

Our analysis of the impacts on wellbeing across different intersections shows that such expectations of employers are more likely to fit with employees’ needs and preferences only when it comes to young women and senior men. These categories, traditionally excluded from corporate investments in terms of skills development and enhancement, are more likely to fit with DPMS that offer them opportunities to fulfil unmet aspirations or recognition needs. In other words, young women and older men are more likely than young and adult men, as well as adult women to conform to the changing employers’ expectations conveyed by the adoption of DPMS. Thus, by adopting a gender and age intersectional perspective our study shows how DPMS reflect a conceptualisation of the ideal worker whose boundaries are progressively blurred and broadened to include women, at least young women.

The research allows for the unveiling of possible practices through which organisations create an inclusive environment. In particular, we might indicate the following practical implications:

• Developmental nature of performance management practice does not imply this practice to be inclusive to all, as DPMS may still result in discriminatory outcomes.

• DPMS are particularly favourable for younger women and older workers who might feel like the organisation is not prone to invest in development initiatives catered to these intersections.

• DPMSs need to be tailored to meet the needs of people belonging to different age and gender intersections, so as not to favour merely those who most resemble the ideal worker. Failure to do so means implicitly favouring one category of personnel over the others.

• DPMS have a lower impact on the wellbeing of adult men and women, therefore different practices for inclusion are needed to improve the wellbeing of adult workers. A first initiative in this direction could be to concentrate all activities, including training activities, within the paid hours of the working day.

• Valuing on-the-job training experiences, by incentivising the use of mentoring, coaching and peer support programmes that can result in the establishment of development plans, tailored to employees’ needs, which could ensure a better balance between professional work and family commitments outside work.

The current study has a number of limitations. In the study, we find that the relationship between DPMS and employee wellbeing is moderated by intersections of gender and age. This difference might be influenced by the role of parenthood and care burden. Future research should examine the role of care burden, in the relationship between DPMS and wellbeing, especially in the categories of adult men and women, as suggested by our results.

Furthermore, our research is based on a convenience sample of Italian employees of companies located in Northern Italy. Hence, although the corrections of the sample, our results are not generalisable given, on the one hand, the limited area they refer to and, on the other hand, the nature of the sample which has limitations in relation to some dimensions such as education and employee’s role. Therefore, it would be of interest to extend the current analysis on a larger representative data set, to evaluate the robustness of the current findings and provide additional empirical evidence about intersectionality in DPMS.

Another limitation of this study is that results might contain spurious relationships among the variables because of the study’s cross-sectional nature. That is, each variable was only measured on one occasion for each participant. Therefore, a longitudinal analysis, with data collected at different time points, might add to the provided results, by evaluating the relationship analysed in an intersectional life span perspective.

Finally, this article suggests new lines of inquiry for future investigations on the impact of DPMS on wellbeing among groups at the intersection of different types of diversity-based inequalities, such as disability or nationality. We encourage further examination in this sense to evidentiate the intricacy of systems of inequality.