Does supporting cultural diversity benefit only Māori? A study of Māori and Pākehā employees
ABSTRACT
Māori are the Indigenous people of Aotearoa but experience workplace disadvantages including high discrimination, lower pay, and greater unemployment. This study uses psychological contracts theory to explore employee perceptions of a set of mutual obligations and implicit promises from their employer around supporting Māori employees. Cultural diversity promise fulfilment (CDPF) focuses on the way firms provide Māori cultural representation, seeks broad inputs from Māori, and work to eliminate bias against Māori. We empirically test CDPF on 165 Māori and 729 Pākehā. We include Pākehā because theoretically, under social exchange theory, all employees might react positively to employer support for a disadvantaged group. We test a moderated mediation model and find support with CDPF being significantly related to job satisfaction, cultural wellbeing, and turnover intentions, with the former mediating CDPF effects to turnover. Next, moderation effects are found (Māori versus Pākehā) but with mixed support. However, moderated mediation effects are supported with Māori employees reporting a stronger indirect effect from CDPF than Pākehā, through both mediators (job satisfaction and cultural wellbeing). The paper establishes the importance of CDPF and helps build the arguments for diversity support.
Introduction
There has been a growing focus on Common Good Human Resource Management HRM (Aust et al. 2020), with the field looking to understand how HRM might enable organisations to achieve the United Nations Sustainable Development Goals. The Sustainable Development Goals outline 17 goals with many applicable to workplaces including good health and wellbeing (#3), decent work and economic growth (#8), and reduced inequalities (#10). The present study focuses on Māori, the Indigenous people of Aotearoa (New Zealand). Māori make up 17.4% of the population and 14.2% of the workforce (Statistics NZ 2023a, 2023b) and there is growing interest in how HRM can enhance the work experience of Indigenous workers (see Spiller and Stockdale 2012; Spiller et al. 2017, 2020; Haar et al. 2020; Manganda et al. 2023). The focus on Māori is especially appropriate with Sustainable Development Goal #10 (reducing inequalities), because Māori appear to face challenges in Aotearoa society especially in the workplace.
There are many unique challenges Māori face in the workplace. This includes high levels of workplace discrimination (Haar 2023a), lower (10%) income (Haar 2022), and 265% higher unemployment than Pākehā, the dominant ethnic group of Aotearoa (Statistics NZ 2023c). Further, Māori have unique workplace experiences. For example, Haar and Martin (2022) identified the double cultural shift, whereby Māori are expected to be navigators and facilitators for others of Māori cultural factors in addition to their normal workloads. Mika et al. (2020) suggest Māori bring unique values to the workplace, such as a greater focus on responsible business and helping others outside the business. Māori reported similar levels of organisational-based self-esteem as Pākehā but far greater benefit on job outcomes and wellbeing (Haar and Brougham 2016).
The present study suggests Aotearoa organisational fulfillment to cultural diversity promises for Māori aligns well with Common Good HRM as well as Te Tiriti o Waitangi (the Treaty of Waitangi), which ensures Indigenous rights (Oda and Rameka 2012). Building off Chrobot-Mason (2003) who focused on ethnic diversity, we offer cultural diversity promise fulfillment (CDPF), which relates to employee perceptions around whether their employer has delivered their promises around cultural diversity. These promises focus on core diversity issues specifically ‘diverse representation in the workforce, minority input is considered at all levels, different opinions, ideas, and perspectives are valued, racial bias and prejudice eliminated, and support/understanding of unique issues for minority employees’ (Chrobot-Mason 2003, p. 31). Testing the potential benefits of CDPF is important because there are critiques around Diversity, Equity, and Inclusion (DEI). Banerjee (2023) argues that DEI efforts are a colonial project, which represent a ‘smokescreen’ for mainly white neoliberalism institutions to basically ‘feel good about themselves’. Banerjee (2023) warns DEI approaches might be unlikely to transform organisations in any meaningful way. As such, he calls for decolonising that involves structural dismantling not diversity projects. To et al. (2023) similarly argue that the success of diversity initiatives is reliant on people in positions of structural power who need to be genuine champions to ensure progress and change.
This study uses social exchange theory (SET) (Blau 1964) to understand how fulfilled promises encourage employees to react positively. Coyle-Shapiro and Parzefall (2008) note that psychological contract theory draws on the foundations of SET making it a highly relevant theoretical lens. We test CDPF towards turnover intentions, with two mediators, extending the literature by including a culturally relevant factor for Māori (Haar and Brougham 2013). Our study also extends the CDPF focus by including not only Māori but the majority employees of Aotearoa (Pākehā). Theoretically, we argue Pākehā might also react positively under SET to promise fulfillment and we include ethnicity as a moderator (Māori versus Pākehā). Finally, a moderated mediation model is tested to further examine the indirect effects of CDPF and how ethnicity may play a boundary role. We show our study model in .
Literature review: psychological contracts and CDPF
Psychological contracts are defined as employee perceptions of the set of mutual obligations and implicit promises of the employer (Rousseau 1990), reflecting employee perceptions of what their employer owes to employees and in turn, what employees owe their employers (Robinson et al. 1994). Chrobot-Mason (2003) notes that when employees believe promises and obligations are unfulfilled by their employer, this results in unfulfilled promises resulting in detrimental outcomes. Examples of fulfilled promises would be an employer stating they provide their workers with great autonomy and freedom, and having such promises being met will have employees reacting positively to these promises. The evidence (e.g. Jayaweera et al. 2021) suggests that promise fulfillment still plays a fundamentally important role on employee attitudes and behaviours, and the approach of Chrobot-Mason (2003) has been used to provide a base of understanding diversity climates (Buttner et al. 2012). A review of measures for climates on inclusion and diversity, Park et al. (2023) noted that Chrobot-Mason (2003) was useful but limited in scope, which the present study extends. Ultimately, if employers do not fulfill their promises, such as (using the example above), providing less freedom and greater workplace surveillance, then employees would consider employer promises unfulfilled and thus breached. A psychological contract can be positive, where promises are met/fulfilled, although a negative orientation (unmet or breached employer promises) dominates the literature.
Strong evidence shows that breaches are detrimental to employee job attitudes and behaviours (e.g. Robinson et al. 1994; Birtch et al. 2016). Indeed, psychological contract literature enjoys meta-analytic support towards job attitudes (Bal et al. 2008) and work behaviours including performance and turnover intentions (e.g. Zhao et al. 2007; Jayaweera et al. 2021). Further, given the psychological contracts literature focuses on promises, this does raise the ability of researchers to extend the focus beyond traditional elements of psychological contract formation around pay, job security, career development, work responsibility, and support (Rousseau 1990; Robinson et al. 1994; Chrobot-Mason 2003). We suggest CDPF aligns well with psychological contracts and our focus on Māori employees.
Chrobot-Mason (2003) focused on psychological contracts towards ethnic diversity promise fulfillment and found them distinct from traditional psychological contracts. That measure focused on the way firms provide ethnic minority representation, seek broad inputs from minorities, and works to eliminate bias against minorities. These are all issues that Māori workers face (see Brougham and Haar 2013; Haar 2022) and thus apply well to Māori. Chrobot-Mason (2003) noted that expectations amongst minority employees have changed overtime, including ‘their expectations with regard to what they are willing to provide employers, as well as what they believe organisations are obligated to provide in return’ (p. 23). For example, organisations publicly talking about the importance of greater diversity might be seen as signalling these are important factors for their workplaces. This aligns with greater Te Tiriti o Waitangi attention in Aotearoa workplaces, which should see greater minority (Māori) rights supported. For example, recognising Māori data sovereignty principles, and firm obligations to consult with Māori due to legislative requirements (Came et al. 2017). We next examine the theoretical mechanism for understanding how psychological contract promise fulfilment can lead to enhanced employee outcomes.
Theory: SET
Cropanzano and Mitchell (2005) state SET ‘is among the most influential conceptual paradigms for understanding workplace behavior’ (p. 874). SET suggests that employees reciprocate with enhanced work attitudes and behaviours when the treatment they receive from their employer is positive or valued (Blau 1964; Haar and Spell 2004). Fundamentally, when employees perceive valued treatment, they become psychologically conditioned via felt obligations, to respond in a similarly favourable manner (Haar and Spell 2004). In turn, this reciprocation (e.g. enhanced attitudes/behaviours) act as an indicator to employers their efforts are recognised and beneficial. This reinforces employer actions and can encourage further giving (Blau 1964). Aligned with SET, is the related norm of reciprocity (Gouldner 1960), which Haar and Spell (2009) state ‘is a cultural universal based on give and take, which leads to the mutual reinforcement by two parties of each other’s actions’ (p. 1829). Under SET, we see employer actions leading to employee reciprocation, which then leads to more employer actions. Thus, norm of reciprocity captures the mutual reinforcement of each parties’ actions under SET.
Cropanzano et al. (2017) extended SET offering a quadrant model considering both hedonic value and activity, with the latter referring to ‘the extent to which an entity actively exhibits the behaviour in question’ (Cropanzano et al. 2017, p. 497). They suggest two poles: (1) exhibit, and (2) withhold, and this creates a quadrant model with employees perceiving their organisation (our focus here) as administering desirable behaviour (desirable hedonic value and active exhibit), which aligns with CDPF. Hence, an employer who meets their promises around ethnic minority engagement (a desirable value for Māori and Pākehā employees) is expected to reciprocate with enhanced behaviours (see Cropanzano et al. 2017), which has meta-analytic support (see Kurtessis et al. 2017). Birtch et al. (2016) states ‘psychological contract fulfillment is important to the employment exchange because it reflects employee beliefs, expectations and perceptions about the extent to which mutual obligations (implicit promises) between an employee and employer have been satisfied’ (p. 1217).
Hypothesis development
Psychological contracts align with the SET mechanisms of felt obligation and reciprocation (Gouldner 1960; Blau 1964) and empirical studies of Māori employees support SET effects (e.g. Haar and Brougham 2011, 2013). Māori have distinct work experiences compared to the majority workforce (Pākehā), with higher unemployment and lower pay. Aligned with meeting Sustainable Development Goal #10 specifically, under SET we expect CDPF to be linked to important work factors, specifically higher job satisfaction and lower turnover intentions. This builds on existing evidence around the importance of psychological contract fulfillment and how it shapes employee attitudes and behaviours (see Zhao et al. 2007; Jayaweera et al. 2021). In the meta-analysis by Zhao et al. (2007), psychological contract fulfillment was found to have a strong effect on job satisfaction (r = .45) and a significant but moderate effect towards turnover intentions (r = -.34). The meta-analysis by Jayaweera et al. (2021), which focused on behaviours, found a similar effect size towards turnover intentions (r = -.32) with almost three times the number of studies. Ultimately, the literature supports that fulfilling promises made to employees is a valuable way for organisations to extract value from their workforce with higher job satisfaction and lower turnover intentions.
Overall, the evidence around promise fulfillment influencing outcomes aligns with SET mechanisms around workplace support (e.g. Kurtessis et al. 2017). In addition, we include cultural wellbeing, defined as how ‘employees feel about the way their cultural values and beliefs are accepted in the workplace’ (Haar and Brougham 2013, p. 877). We believe that under SET, employees are likely to report greater cultural wellbeing when CDPF is higher. As such, we expect employees to respond positively to a high CDPF reporting higher job satisfaction and cultural wellbeing, and lower turnover intentions. This leads to our first set of hypotheses.
Hypothesis 1: CDPF will be positively related to (a) job satisfaction and (b) cultural wellbeing.
Hypothesis 2: CDPF will be negatively related to turnover intentions.
Mediation effects
Beyond direct effects of CDPF, we extend that literature by including job satisfaction and cultural wellbeing as mediators of CDPF on turnover intentions. Meta-analysis shows job satisfaction can be key to turnover intentions (e.g. Griffeth et al. 2000). Further, meta-analysis (Zimmerman and Darnold 2009) support job satisfaction playing a mediating role of other work factors towards turnover intentions. However, while turnover intentions have been explored amongst Māori workers (e.g. Haar et al. 2012), that study found some unique effects, with family-work conflict being dominant, a finding atypical of the literature. Similarly, towards career satisfaction, Haar and Brougham (2013) found cultural wellbeing was the dominant predictor over several factors (e.g. human capital, organisational sponsorship, individual differences) that dominate the largely western literature. We suggest here that cultural wellbeing might be especially useful for understanding the effects of CDPF on turnover intentions, potentially mediating effects. We posit the following.
Hypothesis 3: (a) job satisfaction and (b) cultural wellbeing will be negatively related to turnover intentions, mediating the influence of CDPF.
Moderating effects
Finally, we test ethnicity as a moderating effect. One element of CDPF that remains unexplored is whether it encourages a felt obligation (Haar and Spell 2004) in the majority ethnicity. Given workplace evidence of differences for Māori workers (Haar and Brougham 2013) especially when compared to Pākehā employees (Haar and Brougham 2016), we suggest this is an important factor to consider. Haar and Spell (2009) state SET is a cultural universal, thus it should operate beneficially across cultures and ethnicities. Chrobot-Mason (2003) stated that ‘organizations are coming to realise that managing diversity means challenging the assumption that all employees are the same and should be treated the same way’ (p. 22). We argue this places workplace diversity as an issue for all workers, not just minorities, albeit it clearly focuses on minorities.
Theoretically, under SET, Māori employees should respond positively when promises around Māori representation are fulfilled. Further, we argue that aligned with inclusion (Shore et al. 2011), Pākehā employees should react positively as well, because this type of support shows their employer is fair and just, specifically towards a disadvantaged group (Māori). Here, we extend the literature by exploring ethnicity to explore whether majority ethnicity respond similarly positively. Finally, given our model includes mediation, we extend the moderating effect and test a moderated mediation model. We hypothesise that while CDPF will benefit all employees, it will be more beneficial for Māori, with ethnicity acting as a boundary condition. We posit the following.
Hypothesis 4: Māori ethnicity (versus Pākehā) will moderate the effect of CDPF leading to superior effects on outcomes for Māori.
Hypothesis 5: The indirect effect of CDPF on turnover intentions through the mediators (a) job satisfaction and (b) cultural wellbeing, will be stronger for Māori employees (moderated mediation).
Methods
Data was collected via a Qualtrics survey panel of New Zealand employees, targeting both Māori and Pākehā employees, with full ethics. Respondents self-identified as either Māori or New Zealand Europeans/Pākehā, with other ethnicities not utilised for this study. Participants are anonymous and paid (by Qualtrics). For more details on Qualtrics see Haar (2023a). The qualifiers in the present study were: (1) being in paid employment; (2) working a minimum 20 hours/week; (3) being aged 18 years and over; and (4) self-reporting ethnicity of Māori and Pākehā. Bernerth et al. (2021) offered recommendations around online sampling which we followed including removing fast and slow respondents and including an instructed response item (‘For this question, answer strongly agree only’). Failure in either option resulted in the respondent being removed.
Overall, data was received from 165 Māori and 729 Pākehā, with 58.8% females, and an average age 38.9 years (SD = 13.0). On average, respondents worked 35.8 hours per week (SD = 10.9), which aligns well with the New Zealand average of 37.98 hours per week (Statistics New Zealand 2023b) and had tenure of 5.48 years (SD = 5.51), which also aligns well with the New Zealand average of 6.0 hours per week (Statistics New Zealand 2023d). By education, 30.2% had a university degree and 16.4% a postgraduate qualification, which is broadly over-representative of the New Zealand workforce which averages 21.6% (university degree) and 11.1% (postgraduate). While roughly half came from the private sector (56.3%), there was spread across government (19.9%), not-for-profit (10.2%). Community (8.7%) and other (5.0%). By work hours,
Measures
All measures were coded 1 = strongly disagree, 5 = strongly agree.
Turnover Intentions was measured using the four-item measure by Kelloway et al. (1999), which is validated in New Zealand (Haar 2023b) including with Māori employees (Haar et al. 2012). A sample item is ‘I intend to ask people about new job opportunities’ (α = .93 Māori, α = .94 Pākehā, α = .94 combined).
CDPF was measured using five-items based on Chrobot-Mason (2003), shaped to focus specifically on Māori due to Te Tiriti o Waitangi legislation. Because this is a new measure, Exploratory Factor Analysis (principal components, oblimin rotation) was conducted to test the measures psychometric properties. See for all details.
Table 1. Exploratory factor analysis cdpf measure.
Job Satisfaction was measured using three items by Judge et al. (2005), similarly well validated with New Zealand and Māori employees (Haar et al. 2014). A sample item is ‘I feel fairly satisfied with my present job’ (α = .90 Māori, α = .86 Pākehā, α = .87 combined).
Cultural Wellbeing was captured with four-items by Haar and Brougham (2013). The original sample was Māori employees only, and has additional validation (e.g. Brougham et al. 2015). Here, we extended this to both Māori and Pākehā, with items expanded from ‘I find real enjoyment in my Māori culture in my workplace’ to ‘I find real enjoyment in my Māori/Pākehā culture in my workplace (α = .86 Māori, α = .84 Pākehā, α = .84 combined).
Ethnicity was coded 1 = Māori, 0 = Pākehā.
Control Variables. We control for demographic factors that might influence turnover intentions including Age (in years), Tenure (in years), and Gender (1 = male, 2 = female) with these demographic factors having meta-analytic links towards turnover (Griffeth et al. 2000; Ng and Feldman 2010). Next, we controlled for Career Stage (item ‘What stage of your career are you at?’, responses coded 1 = early-career, 2 = mid-career, 3 = late career), due to links between career advancement and turnover intentions (Lee 2020). Finally, Job Mobility was controlled due to evidence indicating its links to turnover intent (e.g. Haar et al. 2021). The three-item scale by Haar (2023b), with a sample item ‘I know there are similar jobs available to me outside my organisation’ (α = .86 Māori, α = .79 Pākehā, α = .81 combined).
Measurement model and metric invariance test
Using AMOS (v.28) we conducted a global CFA on all measures with the hypothesised measurement model being a good fit for the combined sample: χ2(142) = 397.9 (p < .001), CFI = .97, RMSEA = 0.05 and SRMR = 0.04. Alternative CFAs were also run but they confirm our hypothesised CFA model being all inferior fits to the data (all p < .001). Further, we conducted metric invariance to confirm that Māori and Pākehā respondents answered items in similar patterns. We conducted a multi-group CFA (see Haar et al. 2014) and followed thresholds form Cheung and Rensvold (2000) who suggest comparing the RMSEA of models (unconstrained model versus measurement weights model). The difference was.01 being below the established critical value (Cheung and Rensvold 2000). This finding supports measurement invariance, meaning comparative analysis is possible.
Analysis
Relationships were tested using PROCESS 4.0 (in SPSS v. 28), specifically model 4 (mediation) and model 8 (moderated mediation). Control variables were entered in Step 1 with CDPF as the independent variable, job satisfaction and cultural wellbeing as dual mediators, and turnover intentions as the dependent variable. Ethnicity was the moderator and products were mean-centred, with bootstrapping (5,000 times), confidence intervals and indirect effects. We checked the data for normality (skewness and kurtosis statistics), and these were found to be within acceptable limits (Hair et al. 2010).
Results
Descriptive statistics for the study variables are shown in .
Table 2. Correlations and descriptive statistics of study variables.
shows that all major study variables are significantly correlated with each other (p < .05) in the expected direction.
Results of the mediation analysis are presented in .
Figure 2. Results of mediation analysis.
The results show that CDPF is significantly related to job satisfaction (β = .44(.04), p < .0001 [LL = .37, UL = .51]), cultural wellbeing (β = .52(.03), p < .0001 [LL = .46, UL = .57]), and turnover intentions (β = -.31(.05), p < .0001 [LL = -.41, UL = -.21]), supporting Hypotheses 1a-1b, and 2. Job satisfaction is significantly related to turnover intentions (β = -.53(.04), p < .0001 [LL = -.62, UL = -.45]) as is cultural wellbeing (β = -.14(.06), p = .0209 [LL = -.25, UL = -.02]). Further, when the mediators are included in the model it fully mediates the effect of CDPF on turnover intentions, with the direct effect dropping (to β = -.01(.05), p = .8019 [LL = -.12, UL = .09]). This supports Hypotheses 3a-3b. Even with both mediators, CDPF has a significant indirect effect (β = -.29(.04), p < .0001 [LL = -.37, UL = -.22]).
The moderation and moderated mediation results are shown in .
Table 3. Moderated mediation results.
Regarding the moderator, Māori (versus Pākehā) had a significant direct effect on job satisfaction (β = -.15(.07), p = .0453 [LL = -.22, UL = -.00]) only. Significant interaction effects were found between CDPF and Māori towards job satisfaction (β = .22(.09), p = .0185 [LL = .04, UL = .40]) and cultural wellbeing (β = .17(.07), p = .0137 [LL = .03, UL = .30]), and turnover intentions (β = .33(.11), p = .0043 [LL = .10, UL = .55]). This supports Hypothesis 4. We graph the interactions to illustrate effects in .
Figure 3. 2-way interaction of cdpf x ethnicity towards job satisfaction.
Figure 4. 2-way interaction of cdpf x ethnicity towards cultural wellbeing.
Figure 5. 2-way interaction of cdpf x ethnicity towards turnover intentions.
shows at low levels of CDPF, there are significant differences in job satisfaction, with Pākehā reporting higher job satisfaction than Māori employees. At high levels of CDPF, all respondents report significantly higher job satisfaction and there are no differences between Māori and Pākehā. Both Māori and Pākehā report high job satisfaction, providing partial support for the hypothesis. shows at low levels of CDPF little differences between Māori and Pākehā employees towards cultural wellbeing. At high levels of CDPF, all respondents report significantly higher cultural wellbeing, but Māori report a greater level of increase, with highest cultural wellbeing. This supports the hypothesised effect. shows at low levels of CDPF there are little differences between Māori and Pākehā towards turnover intentions. At high levels of CDPF, Pākehā respondents report stable levels of turnover intentions while Māori report significantly higher turnover thoughts, counter to the hypothesised effect.
Finally, the index of moderated mediation is significant for CDPF on turnover intentions through both job satisfaction (index = -.12(.05), p = .0170 [LL = -.23, UL = -.01]) and cultural wellbeing (index = -.03(.02), p = .0495 [LL = -.06, UL = -.00]). This means the indirect effect of CDPF on turnover intentions through job satisfaction and cultural wellbeing (as mediators) differs by ethnicity (Māori versus Pākehā). These are presented in and (graphed moderated mediation effects).
Figure 6. Indirect effects of cdpf on turnover intentions through job satisfaction conditional on ethnicity.
Figure 7. Indirect effects of cdpf on turnover intentions through cultural wellbeing conditional on ethnicity.
and show similar moderated mediation effects, so these are presented together. We follow Wayne et al. (2017) around probing conditional indirect effects by examining the magnitude and significance of the indirect effect of CDPF on turnover intentions through both mediators. While typically at three levels of a continuous variable (-2SD/Mean/+2SD) because our moderator (ethnicity) is dichotomous, we graph it only at Māori and Pākehā. The analysis shows that for Pākehā respondents, the indirect effect of CDPF is significant and negative on turnover intentions through job satisfaction (β = -.21(.03), p < .0001 [LLCI = -.27, ULCI = -.14]) and through cultural wellbeing (β = -.08(.03), p = .0067 [LLCI = -.14, ULCI = -.02]). Māori respondents reporting report a significant, negative, and stronger indirect effect on turnover intentions through job satisfaction (β = -.32(.06), p < .0001 [LLCI = -.44, ULCI = -.21]) and through cultural wellbeing (β = -.11(.04), p = .0076 [LLCI = -.19, ULCI = -.02]). The index of moderated mediation shows the indirect effect of CDPF is significant across the full 95% confidence intervals, indicating the indirect effect of CDPF is stronger for Māori than Pākehā although both are significant. These effects support Hypothesis 7a-7b.
Overall, the models are significant for job satisfaction (F = 23.3285, p < .0001), cultural wellbeing (F = 51.7727, p < .0001) and turnover intentions (F = 32.7977, p < .0001), and account for moderate amounts of variance towards job satisfaction (18%), but larger amounts for turnover intentions (28``) and cultural wellbeing (32%). Regarding the control variables towards job satisfaction, age is significant (β = .05(.01), p = .0022 [LL = .02, UL = .07]), as is career stage (β = .09(.04), p = .0129 [LL = .01, UL = .17]), and job mobility (β = .14(.03), p < .0001 [LL = .08, UL = .20]). Towards cultural wellbeing, significant control variables are age (β = -.02(.01), p = .0398 [LL = -.04, UL = -.00]), and job mobility (β = .06(.02), p = .0092 [LL = .01, UL = .10]). Finally, significant control variables towards turnover intentions are age (β = -.07(.02), p = .0001 [LL = -.11, UL = -.04]), career stage (β = -.11(.05), p = .0275 [LL = -.21, UL = -.01]), and job mobility (β = .39(.04), p < .0001 [LL = .31, UL = .47]).
Additional analysis
We conducted additional analysis on CDPF to better understand similarities and differences. A t-test (t = 1.898(df = 892), p = .058) found non-significant differences on CDPF for Māori respondents (M = 3.78, SD = .79) versus Pākehā (M = 3.65, SD = .76). However, ANOVA did find significant differences across sectors (F = 4.120, p = .003). Overall, respondents working in the not-for-profit sector reported the highest CDPF (M = 3.92), which was significantly higher than community sector (M = 3.76), which was significantly higher than government (3.73), which in turn was significantly higher than private sector (M = 3.61). The lowest score (3.54) is in ‘other’ sector.
Discussion
The paper makes two important contributions. First, CDPF is measurable in the Aotearoa context including with an Indigenous workforce population, and the psychometric properties of CDPF indicate the measure is a useful way to capture workplace experiences from both Māori and Pākehā around cultural diversity promises for Māori. Importantly, the high mean scores show good progress is being made in Aotearoa workplaces and both Māori and Pākehā agree that Aotearoa workplaces provide good support for cultural diversity (mean scores above the mid-point), although unsurprisingly, these experiences do vary. These findings differ from Chrobot-Mason (2003) who found only modest support for CDPF being fulfilled (below the mid-point). This might reflect the two decades since Chrobot-Mason (2003) original scale, and growth in support for cultural diversity in Aotearoa, perhaps influenced by Te Tiriti o Waitangi obligations. It is important to acknowledge that our CDPF measure is based and used within a different cultural setting than Chrobot-Mason (2003), which was the US, and the focus was on university employees. Thus, our differences might reflect a different cultural setting and a broader examination of the workforce.
Second, CDPF was beneficial for both Māori and Pākehā employees, a new addition to the psychological contracts’ literature focusing on ethnic diversity. This provides an important insight into how addressing ethnic diversity in organisations can be beneficial for all employees–even when minorities are targeted. This universal beneficial finding also aligns with SET around promise fulfillment benefiting all employees. As such, reducing barriers for Māori in the workplace are acknowledged and valued by Pākehā, represent a fair and just place to work that elicits felt obligations under SET.
We found strong support for CDPF influencing turnover intentions and job satisfaction, which aligns with meta-analyses of psychological contracts (Zhao et al. 2007; Jayaweera et al. 2021). The present study extended the outcomes influenced by CDPF (and by association, psychological contracts) by finding positive effects towards cultural wellbeing, including for Pākehā employees. While research has found this is an important factor for Māori (e.g. Brougham et al. 2015; Haar and Brougham 2022), this study extends these findings and shows it may be more generally applicable beyond Indigenous or minority employees. Indeed, both job satisfaction and cultural wellbeing were significantly related (negatively) to turnover intentions.
Finding the beneficial effects of CDPF differed by ethnicity (Māori versus Pākehā) further adds new insights. While we hypothesise that Māori would respond more favourably than Pākehā, this was only clearly supported towards cultural wellbeing, where Māori did report higher levels compared to Pākehā when CDPF was high. The effects towards job satisfaction show similar effects for Māori and Pākehā when CDPF is high. However, the moderator (ethnicity) is significantly and negatively related to job satisfaction, showing that Pākehā have higher job satisfaction directly, and thus high CDPF might be viewed as enabling Māori employees to record similarly high levels of job satisfaction.
Further, the moderating effects highlight that high CDPF lead Māori to consider turnover more and not less as hypothesised, which encourages examination. While tempting to suggest that CDPF has driven Māori to consider leaving more than Pākehā, the moderated mediation effects do not support this. Indeed, we find that CDPF is negatively related to turnover intentions indirectly for both Māori and Pākehā through either mediator. The indirect effect is stronger for Māori, which does align with our hypothesised argument. Hence, we do find that Māori report a greater felt obligation from CDPF likely due to the focus on enhancing Māori diversity in their workplaces. These findings support arguments around conducting moderated mediation analysis to gain deeper insights (e.g. Haar et al. 2021) and highlight the potential complexity of relationships. Specifically, while the effects are stronger for Māori employees, we draw attention to the point that through both mediators, the indirect effect of CDPF is also significant for Pākehā. This reiterates the direct effects of CDPF and encourages firms to adopt a greater focus on cultural diversity and delivering on promises.
Our findings are also important towards the critical views of DEI. Banerjee (2023) suggested such efforts are unlikely to transform organisations although our evidence here suggests there are employee benefits–both for the DEI targets and the majority ethnicities. Perhaps our findings align with To et al.’s (2023) point around requiring people in positions of structural power to create change through being genuine champions. It might be through fulfilling the diversity promises that organisational leaders are indicating their genuine commitment to diversity. Extending this line of critical inquiry, we note that built into the notion of diversity is the idea is an implicit normal and diverse (the different). Labelling those who differ from the norm is likely counterproductive, not just for Māori, but similarly other minorities globally including indigenous peoples and black Americans.
Implications and future research
The key implication for organisations is delivering on any promises around cultural diversity. The findings show that both minority and majority employees react beneficially when promises are fulfilled around cultural diversity and thus following through on promises around representation and growing numbers of minorities and disadvantaged groups is key. CDPF does include eliminating biases and prejudice, and this is timely given research showing workplace discrimination for Māori is high in frequency, albeit low in intensity (Haar 2023a). An organisation promoting a CDPF but failing to reduce discrimination is in effect breaching their promises, which is known to be detrimental to worker outcomes (Bal et al. 2008) including performance (e.g. Zhao et al. 2007).
Organisations should recognise that in today’s Aotearoa, Pākehā employees also benefit when organisations deliver on the promises of enhancing workplace equity for Māori, who we know are disadvantaged on average (see Haar 2022; Haar and Martin 2022; Haar 2023a; Statistics NZ 2023c). So, workplaces might see greater equity for Māori not only aligning with legislative support via Te Tiriti o Waitangi (Came et al. 2017) but also being good for business, by enhancing key work outcomes for Māori and Pākehā. Businesses should understand that focusing on a disadvantaged group shows the majority workforce ethnic group (Pākehā) that they are working for a fair and equitable employer. This benefits them also and is beneficial as this enhances job satisfaction and encourages retention.
New Zealand researchers are encouraged to utilise the CDPF more and provide additional data around the measure including replicating the findings here and extending outcomes to include attitudes like meaningful work for example. Researchers might also extend the study of CDPF to look at all minorities, which in Aotearoa includes pan-ethnic groups, particularly Pacific peoples, Asians, and MELAA (Middle Eastern/Latin American/African). Understanding the levels of CDPF and how these might differ across minority ethnicities might provide useful insights. We do encourage researchers to include the majority ethnicity on studies of CDPF to determine whether our findings are generalisable. Researchers might take suggestions from Chrobot-Mason (2003) and explore trust (moderator/mediator), to explore potential differences from CDPF. We also suggest exploring other workplace factors like citizenship behaviours and wellbeing outcomes like mental health. Indeed, there is much less research in the psychological contracts field in general towards employee wellbeing, especially with CDPF. Future studies might extend the examination of boundary conditions, for example, testing moderated moderated mediation models (see Haar et al. 2019) to provide more in-depth analyses.
Limitations
Given the data used here is cross-sectional, there is the potential concern around common method variance. Thus, collecting CDPF at the same time as our outcome measures might influence the effectiveness. In response to the potential issue around common methods, we follow suggestions from Podsakoff et al. (2003) regarding post hoc testing. First, we conducted the Harman’s One Factor test with the resulting unrotated factor analysis yielding a first factor accounting for 24.1% of the variance, well below the 50% threshold (Podsakoff et al. 2003). Second, we conducted the Lindell and Whitney (2001) test, where we conducted a partial correlation while controlling for an unrelated construct. We controlled for industry type, which offers 20 options, and the analysis showed no change in the strength of correlations. Overall, these post hoc analyses suggest that common method variance is not modifying the strength of relationships. In addition, Evans (1985) conducted Monte Carlo simulations and concluded that significant moderation effects are unlikely if common method variance is present. The present study found multiple moderation and moderated mediation effects. From these, we suggest common method variance is not a critical factor in these findings. Overall, our sample included a broad range of employees across sectors, industries, and occupations, providing confidence in the findings.
Conclusion
The present study finds strong support for CDPF influencing key work outcomes of turnover intentions and job satisfaction and extends this to include cultural wellbeing. While the study found stronger effects for minority employees (Māori) there were similar beneficial effects for the majority (Pākehā). This is an important contribution because it shows that fulfilling promises around ethnic diversity likely benefits all. This should provide fewer barriers for firms looking to take proactive decisions around minorities. Overall, the model showed while minority ethnicity clearly benefits from CDPF these benefits do spillover to other employees, encouraging organisations to actively embrace cultural diversity.
Disclosure statement
No potential conflict of interest was reported by the author(s).