Effect of Diversity and Inclusion-oriented HR Practices and Strategic Leadership on Perceived Team/Club Performance: Testing Perceived Organisational Alignment with Vision 2030 Sports Objectives as a Moderator

2.5. Literature Gap

The available literature offers a coherent amount of evidence that D&I-oriented HR practices and strategic leadership have a positive effect on the performance of organisations and teams; however, there are a number of shortcomings.

Although unprecedented national investment is being made under Vision 2030, a range of sports organisations still face uneven organisational capability, imbalanced HR maturity, and leadership issues that can undermine the transfer of national investments into sustainable organisational performance. This means that infrastructure growth alone might not be adequate unless internal organisational means and institutional fit evolve concurrently. Other studies, including (Singh et al., 2024; and Saha et al., 2024), establish that structured D&I systems improve performance; however, the data are limited by the use of HR professionals or secondary firm-level data to understand perceived team-level results, especially in a sports context. Similarly, sport related research (e.g., Clare et al., 2025; Liqiang et al., 2024) supports the relationship between leadership and performance, and the effect is often small or indirect, pointing to a need to take further steps to explain the variations in performance with regard to the concept of leadership. Further, the relationships are largely external circumstance dependent as is now evident. (Kiplagat et al., 2024; and Arora, 2024) indicate that leadership effectiveness is based on contextual governance and environmental alignment. Nevertheless, such studies are mainly set in the context of the public sector or corporations, where they are not applicable to sports organisations that experience national change. In general, the lack of comprehension of the perceived role of national strategic frameworks, such as Saudi Vision 2030, in pre-conditioning the efficacy of D&I-oriented HR practices and strategic leadership regarding perceived team/club performance is clear and prompts the proposed hypotheses to be tested.

3. METHODOLOGY

3.1. Research Design

This research employed a quantitative study design, which can be characterised as a methodological enquiry involving the use of numerical data and statistical methods to study the connections between variables (Taherdoost, 2022). Hypothesis testing and theory validation are two specific areas where quantitative methods are most appropriate because they can be objectively measured and generalised to the findings. This approach was appropriate for the present study, as it aimed to test the relationships between D&I-oriented HR practices, strategic leadership, and perceived team/club performance, as well as the moderating role of perceived organisational alignment with Vision 2030 sports objectives, using statistical modelling.

The organisations engaged included professional and community sports clubs, governing bodies, and sports development organisations within Saudi Arabia. Such organisations are involved in the development of athletes, management of competitions, community involvement programs, and execution of national sports policies. As Vision 2030 has greatly increased public investment and governance reforms in this sector, workers in these organisations are directly exposed to strategic initiatives that facilitate organisational change, making them a suitable group to study HR practices, leadership, and institutional alignment.

3.2. Data Collection Tool

Data was collected using a structured, closed ended questionnaire, created through Google Forms. An online survey was chosen for the research rather than face to face interviews for a number of reasons: firstly, it was effective in terms of speed of completion and secondly, the potential reach was much greater and thirdly, it would allow for the sampling of people that were geographically spread out in the sports industry. In addition, Internet tools eliminate interviewer bias and enable respondents to take the survey at a time of their own choosing. The questionnaire used a 5-point likert scale from 1 (Strongly Disagree) to 5 (Strongly Agree) and included four constructs: D&I oriented HR practices, strategic leadership, perceived team/club performance, perceived organisational alignment with Vision 2030 sports objectives (Appendix A).

The study constructs were measured using items that had been previously tested in the literature to ensure their validity. Inclusive policies and practices were captured in D&I-oriented HR practices based on (Zhao & Ren, 2025b). The Strategic Leadership Scale (SLS) was adapted from (Pisapia, 2006) to include items that focus on strategic leadership, which involve vision and support of leadership. The Perceived Performance in Team Sports Questionnaire (PPTSQ) is a sport-specific questionnaire developed by (Gershgoren et al., 2021) that was used to assess perceived team/club performance. Perceived organisational alignment with Vision 2030 sports objectives was based on the literature on strategic alignment because no standardised scale was available.

Since the research was carried out on staff members of Saudi sports organisations, the questionnaire was designed in English and translated into Arabic using the back-translation technique that is suggested in cross-cultural research. The questionnaire was translated into Arabic by a bilingual scholar with knowledge of management research for the first time Two independent bilingual persons translated the Arabic version into English. All discrepancies between the original and back translation versions were discussed, and it was decided that the versions should be as semantically equivalent and contextually appropriate as possible. A pilot study with 30 sports organisation workers was first undertaken to check the clarity of the items, wording and flow of the questions in the questionnaire. Minimal changes were made to words to enhance readability and the conceptual meaning of constructs did not change. Results of the pilot study were not used in the final analysis.  However, formal measurement invariance testing (e.g. MICOM) was not conducted between the Arabic and English versions. Therefore, the study lacks cross-language comparability and cannot confirm that the constructs were measured equivalently across both language versions.
.
Table 1 shows the adaptation process for all the measurement items, including the original source, change in wording, rationale for adaptation, method of translation, and changes after the pilot period. The actual questionnaire used to collect information from all respondents is provided in the Appendix. No items on the questionnaire were changed after data collection was completed.

While the constructs are conceptually related, the measurement items have been purposefully operationalised to capture different theoretical domains. D&I-oriented HR practices measured employees’ perceptions of inclusive HR policies, communication, and fairness, while Strategic Leadership measured leadership vision, strategic direction, and managerial support. Perceived Team/Club Performance included the respondents’ assessment of organisational performance outcomes. All items were examined during the adaptation process to reduce conceptual overlap but ensure consistency with the theoretical definition of each construct. Additionally, the empirical results of discriminant validity (HTMT< 0.85) showed that the constructs remained statistically different.

3.3. Population and Recruitment

The target group comprised club staff and administrators associated with different sports organisations. This group was chosen because of its direct participation in implementing HR practices, interacting with leadership, and team performance, and is therefore well positioned to offer informed perceptions of the study variables. Professional networking sites such as LinkedIn and social networking sites such as Facebook, where sports professionals and organisational members are also active, were used as recruitment methods to secure participants. The survey link was sent in invitations through direct messages and to the appropriate groups. However, the use of digital platforms can leave less digitally literate people behind. To address this, the survey was designed to be simple, mobile-ready, and sent out via a number of channels to make it more accessible.

3.4. Sampling Technique and Sample Size

Respondents with experience in sports organisations were selected using purposive sampling. Purposive Sampling is a non-probability sampling technique in which the subjects are selected intentionally based on their characteristics and perhaps their ability to provide an appropriate and informed contribution in the study (Campbell et al., 2020). The criteria used for eligibility were: (1) member must be currently employed in a sports organisation (e.g., club staff, club administrator, management team), (2) at least 1 year of employment within the organisation, and (3) directly or indirectly involved in the organisation’s team, HR practices, or leadership process. With these criteria, the respondents were assured of knowledge and exposure to adequately assess the constructs under study.

The final sample comprised employees from ten sports organisations. The range of respondents per organisation was 15-50 (Mean = 30.0), showing that multiple respondents were recruited from each organisation. However, the unit of analysis remained individual employees, and no aggregation was carried out as responses indicated individual perceptions and not organisational-level measures. However, clustering effects cannot be entirely excluded and must be investigated in future studies using multilevel approaches. Data collection was conducted from June to July 2025. Approximately 400 staff members were invited via organisational networks and staff social media. A total of 323 questionnaires were returned, resulting in a response rate of 80.75. After data screening, 300 valid responses were used for data analysis, as questionnaires with missing values were excluded. To prevent duplicate submissions, a Google form with a one-response setting was used.

G*Power (version 3.1.9.7) was used to determine the minimum sample size and appropriate procedure recommended in social science research. The a priori power analysis was done using three main-effect predictors (D&I-oriented HR practices, Strategic Leadership, and Perceived Organisational Alignment with Vision 2030 Sports Objectives), assuming a medium effect size (f² = .15), α = .05, and statistical power of .80, which resulted in a minimum required sample of 119 respondents. In addition, the complete structural model was tested with two interaction terms (D&I-oriented HR Practices × Perceived Organisational Alignment with Vision 2030 Sports Objectives and Strategic Leadership × Perceived Organisational Alignment with Vision 2030 Sports Objectives). Thus, a sample size of 300 was well above this requirement.

3.5. Data Analysis

Partial Least Squares Structural Equation Modelling (PLS-SEM) was used to analyse the data. As mentioned by (Hair et al., 2021) as cited in (Ghouri, 2023), PLS-SEM can be used to test theories, evaluate models, and validate hypotheses, to explore the complex relationship between the constructs of a theory and the indicators of the constructs of a theory. PLS-SEM was appropriate for this study because it can address a complex model with moderation effects and non-normal data. This analysis consisted of two steps: (1) Measurement Model (reliability and validity) assessment and (2) Structural Model (path coefficients, significance, and moderation effects) assessment. The structural model was estimated using SmartPLS version 4.0. Two-tailed significance testing and bias-corrected confidence intervals were used with 5,000 bootstrap resamples. The moderation effects were computed using a path weighting scheme and product indicator approach.

3.6. Common Method Bias, Non-response Bias, and VIF

This study adopted both procedural and statistical techniques to assess common method bias (CMB). First, Harman’s single-factor test was used, which, according to Rodríguez-Ardura and Meseguer-Artola (2020), tests whether a single factor explains more than 50% of the variance in the data. The results showed that no single factor had more than 50% variance, implying that there was no significant bias. Second, following the full collinearity assessment approach proposed by (Kock, 2015), construct-level variance inflation factors (VIFs) were examined to assess both collinearity and potential common method bias. A VIF less than the conservative value of 3.3 suggests that the model does not exhibit any significant collinearity issues and that there is no likelihood that common method bias issues would compromise the validity of the results. Table 2 reveals that none of the construct-level VIFs were higher than this value, meaning that collinearity could not be an issue and that the likelihood of common method bias influencing the study results was low. The two diagnostics indicated that the results did not indicate a common method bias.

To ensure there was no non-response bias, independent-sample t-tests were conducted to compare the first half of the responses (150) with the second half (150) for each construct. The results showed no statistical differences (p > 0.05) suggesting that the results were unlikely to be affected by non-response bias.

3.7. Ethical Considerations

This study was conducted in compliance with internationally accepted ethical principles for conducting research. Participation was voluntary, and the purpose of the study was explained to the respondents. Anonymity and confidentiality were assured, and no personally identifiable information was gathered. Before participation, informed consent was obtained from all respondents. The study was independent and did not require a formal institutional ethics committee review, as it was an anonymous, minimal-risk questionnaire.

4. RESULTS

4.1. Descriptive Analysis

Table 3 presents the demographics of the 300 respondents. The vast majority of them were male (66.0%) and female (30.7%), which is indicative of the gender gap in the sports industry. The majority of the respondents were 25-34 (42.0%), indicating a relatively young workforce. In terms of education, the sample of the population was well educated, with the majority being bachelor’s degree holders (46.0%) and master’s degree holders (22.0%). Regarding positions, the highest percentage was the administrative/operational staff (28.0%), followed by the coaching staff (24.0%), which means that both technical and managerial levels were represented in the sample. The levels of experience were also fairly even, with 6-10 years of experience (28.0%) and 1-3 years of experience (26.0%) being the most frequent, indicating a fair amount of industry exposure.

4.2. Measurement Model Analysis

Table 4 presents the results for the measurement models. The reliability of all factor loadings were greater than 0.70 (e.g., DIHR2 = 0.886; PTP2 = 0.932) a good indicator of reliability. The values of Cronbach’s alpha ranged from 0.818 to 0.901 and composite reliability values were greater than 0.80, suggesting satisfactory internal consistency. Furthermore, the convergent validity of the AVE values was higher than 0.50 for both D&I-oriented HR practices and performance.

Note: Factor Loadings ≥ 0.70; Cronbach’s alpha ≥ 0.70; CR ≥ 0.70; AVE ≥ 0.50 (Hair et al., 2021; Mia et al., 2022)

4.3. Discriminant Validity Analysis

Table 5 shows the discriminant validity of the heterotrait-monotrait (HTMT) ratios. All values were below the threshold of 0.85 (for example, 0.706 between vision alignment and performance; 0.577 between D&I-oriented HR practices and strategic leadership) which proves that the constructs are differentiated and measure different things.

Note: HTMT threshold < 0.85 (Afthanorhan et al., 2021; Cheung et al., 2023)

4.4. Model Fit

Table 6 presents the model fit indices for both the saturated and estimated models. The SRMR value of 0.057 for both models is below the commonly accepted threshold of 0.08, indicating a good overall model fit. The d_ULS values of 0.257 and 0.256 and the d_G values of 0.186 and 0.185 are very similar across the two models, suggesting consistency between the saturated and estimated specifications. The NFI value of 0.846 indicates a reasonably acceptable fit, although it is slightly below the commonly recommended threshold of 0.90. Overall, the results suggest that the estimated model demonstrates an adequate level of fit to the observed data.

4.5. Explanatory Power and Predictive Relevance

The explanatory power and predictive relevance is reported in Table 7. The value of R² (0.436) was moderate, which means that the model could explain 43.6% of the variance in perceived team performance. The Q² was 0.400, which indicated that the model was relevant to predicting.

Note: R² = 0.25, 0.5, and 0.75 are weak, moderate, and substantial predictors, respectively; Q² > 0 indicates predictive relevance (Changalima & Chuwa, 2025).

4.6. Path Coefficient

Table 8 presents the results of the structural model and the hypothesis testing. Two complementary criteria were applied to determine the decisions regarding the hypotheses. Bootstrap p-values (p < 0.05) were first determined for statistical significance. Second, substantive significance was evaluated using Cohen’s effect size (f²), where effects below 0.02 were considered negligible. Therefore, hypotheses were considered well supported only if there was both statistical and practical substantive significance. Relationships that were statistically significant but had negligible effect sizes were considered statistically supported but not substantively supported.

Note: β = standardised path coefficient; SE = bootstrap standard error; t = bootstrap t-statistic; p = bootstrap significance level; CI = 95% bias-corrected bootstrap confidence interval; LL = lower limit; UL = upper limit; **p < 0.05; ***p < 0.01.

Regarding H1, that is the existence of a positive relationship between D&I-oriented HR practices and perceived team performance, the findings showed a significant positive relationship between these variables (β = 0.209, p = 0.003 and t = 2.989), but the result was small in size (f² = 0.043). However, the strategic leadership is not significantly associated with performance (β = 0.063, t = 1.055, p = 0.292), so that the contribution of strategic leadership will be weak (f² = 0.005) and does not support the hypothesis of H2.  The direct effect of perceived alignment with Vision 2030 sports objectives on perceived team performance was significant and strong β = 0.471, p = 0.001, f² = 0.220) which supports H5.

Although both interaction terms were statistically significant (H3: β = -0.016, p = 0.001; H4: β = -0.006, p = 0.035), they had a very small effect size (f² = 0.001), suggesting that they were statistically significant but negligible in practice. Both interaction terms were statistically significant, but the size of the effect was negligible (f²=.001). This means that the interaction effects are statistically significant but not very meaningful in practice. Thus, the results do not demonstrate meaningful moderation. Perceived Organisational Alignment with Vision 2030 Sports Objectives is better interpreted as a direct organisational predictor of perceived performance than as a substantive boundary condition.

4.7. Conditional Effects

Table 9 shows the conditional effects, where the positive associations of D&I-oriented HR practices and strategic leadership with perceived team performance weaken slightly as perceived organisational alignment with Vision 2030 sports objectives grows. However, the differences between low, medium, and high levels of alignment are extremely small and represent little real difference in the strength of these relationships. This pattern is consistent with negligible interaction effect sizes (f² = 0.001), indicating that there is no meaningful moderating effect of perceived organisational alignment with Vision 2030 sports objectives on the effect of diversity and inclusion-oriented HR practices or strategic leadership on perceived team performance.

5. DISCUSSION

The research question aimed to understand how internal organisational capabilities or institutional alignment is a better predictor of perceived team performance in Saudi sports organisations working under the transformation agenda of Vision 2030. The results show that the factors have varying levels of influence on perceived performance, with perceived institutional alignment appearing to be a more powerful factor of explanation than the internal organisational capabilities initially hypothesised. Instead of the hypothesised predictors having similar effects, the findings show a pattern of differentiation in the relationships. This trend must be explained by the fact that the data are perceptual, cross-sectional, and non-probability, which means that the subjective assessment of employees is involved, not objective organisational performance. In general, the results indicate the asymmetric incorporation of the theoretical framework, in which the Resource-Based View can be used to justify the minimal contribution of internal resources, and the Contingency Theory can be manifested more through direct contextual impact than meaningful moderation.

H1 is supported, according to which D&I-oriented HR practices are positively and significantly related to perceived performance. The results show a positive association between D- and I-oriented HR practices and perceived performance (β = 0.209, t = 2.989, p = 0.003). This is consistent with earlier research (Singh et al., 2024; Saha et al., 2024), which points out that well-structured and formalised inclusion processes have the potential to enhance collaboration, psychological safety, and decision quality. Nevertheless, the effect size in this study is rather small (f² = 0.043), which implies that D&I-oriented HR practices are significant, but their effect on perceived team performance is relatively small when considered individually. This is in line with the claim made by (Okatta et al., 2024), who observed that D&I efforts are more likely to have better results when they are backed by adjunctive processes such as leadership commitment and implementation strength. In this regard, the current results support a more moderate understanding of the effects of D&I: it is important, but it does not prevail in performance perception. This can be suggestive in the context of sports team settings, in particular, that inclusion can enhance interpersonal functioning but not necessarily lead to more robust perceptions of competitive success that are frequently determined by the result of matches, the level of skills, and coaching policies outside HR systems.

The findings do not confirm H2, which states that strategic leadership is positively and significantly related to perceived performance (β = 0.063, t = 1.055, p = 0.292). This is contrary to a large portion of the literature on leadership performance (e.g., Clare et al., 2025; Liqiang et al., 2024), which usually show positive correlations. The absence of a strong direct impact in the present study may imply that employees or team members do not directly correlate performance gains with leadership. Leadership can be a background condition that facilitates processes rather than a performance driver. This interpretation aligns with critiques of sports research that have identified that leadership effects may be less salient in high-stakes, result-oriented settings where performance is largely influenced by external competition conditions and athlete competence, and less by management (Tadesse et al., 2024). The small direct effect of strategic leadership implies that its performance impact might not be direct but indirect and process-based, as strongly suggested by earlier literature. The processes of leadership may be reinforced by team cohesion, psychological safety, motivation, or quality of communication, which were not investigated in this study. Future studies should embrace mediation models to include these indirect pathways and provide a more comprehensive picture of the effectiveness of leadership in sports organisations. The use of multistage/longitudinal designs can further elucidate how leadership is translated into performance over time, especially in institutionally aligned contexts.

H3 states that perceived organisational alignment with Vision 2030 sports objectives significantly moderates the relationship between diversity and inclusion (D&I)-oriented HR practices and perceived team/club performance. The interaction term is statistically significant (β = -0.016, p = 0.001), but its effect size is insignificant, which means that there is no substantive moderating effect. Similarly, H4 explains that perceived organisational alignment with Vision 2030 sports objectives plays a significant role in moderating the relationship between strategic leadership and perceived team/club performance. The outcome is statistically significant (β = -0.006, p = 0.035), but the effect is practically non-significant. H3 and H4 attained statistical significance but were not substantively supported because of the negligible size of the interaction effect (f² = 0.001). Hence, there is no practical significance in the interaction as a meaningful moderator, even though it is statistically present. Therefore, H3 and H4 are not supported from a practical perspective. Since the moderation effects had no substantive practical significance, they should not be considered to possess meaningful boundary conditions. Instead, the data show that perceived organisational alignment with Vision 2030 sports objectives is a direct predictor of perceived team performance and not a moderator.

These results offer a weak justification for contingency theory in its classical moderation form, as the statistically significant interaction terms do not have substantial effect sizes. In line with (Kiplagat et al., 2024), performance perceptions seem to be influenced by contextual alignment, which provides legitimacy and institutional guidance. However, the current findings indicate that this influence acts mostly via a direct effect rather than a real interaction process.

Moreover, the negative coefficients should not be understood as a negative impact but as a slight weakening of the relationship between the internal factors (D&I-oriented HR practices and strategic leadership) and perceived performance in the upper levels of alignment. This trend can indicate a possible substitution effect, where high perceived institutional compatibility acts as the primary reference point in assessing performance, and thus lessens the dependence on internal organisational qualities. This has been witnessed in the governance literature, whereby powerful institutional setups can overshadow internal processes (Iqbal et al., 2025).

This study does not negate the RBV but rather introduces an important limitation to the RBV in policy-oriented organisational settings. RBV assumes that valuable internal capabilities create a competitive advantage, but the value of these capabilities can be contingent upon employees’ understanding of the organisational context within which they are set. From this perspective, D&I-oriented HR practices and strategic leadership continue to be strategic tools, with employee perceptions of value contingent on their awareness of the organisation’s strategic priorities. This builds upon the RBV by stating that internal resource effectiveness does not exist in isolation but rather in conjunction with employees’ perception of the organisation’s strategic direction. Therefore, the combination of RBV and Contingency Theory offers a more context-sensitive account of organisational performance, as strategic resources and perceptions of the organisational context both affect performance assessments.