Theoretical underpinnings of employee engagement and gender disparities

Employee engagement is a critical component of organizational effectiveness, characterized by employees’ emotional, cognitive, and behavioral commitment to their work roles [12]. It has been proposed that employee engagement is influenced by the availability of job resources such as autonomy, support, and opportunities for career development, whereas excessive job demands may result in burnout and disengagement [1]. Furthermore, it was found in past research that this relationship can be explained through the lens of Social Exchange Theory, suggesting that employees tend to reciprocate perceived organizational support with higher levels of engagement [5]. However, gender disparities in engagement challenge these theoretical models, as women’s experiences in the workplace are shaped by social, cultural, and structural constraints. Social Identity Theory and Role Congruity Theory provide insights into these gender-based challenges, suggesting that women often struggle with identity threats and bias in male-dominated work environments [6,11]. Although these theories provide valuable insights, they primarily reflect Western corporate cultures, leaving a gap in understanding how engagement dynamics operate in developing economies like Bangladesh, where socio-cultural constraints play a larger role in shaping women’s working lives [4].

An integrated framework: Resources, reciprocity, and role constraint

The four perspectives used here are not alternatives but layers of one argument. The Job Demands-Resources model specifies the architecture: socio-cultural expectation operates as a chronic demand carried across the home and work domains, while organizational culture, leadership support, work-life integration, and career development operate as job resources [1]. Social Exchange Theory specifies the mechanism that turns a resource into engagement: employees reciprocate support they perceive as discretionary and genuine, so a resource is engaging only to the extent that it is enacted rather than declared [5]. Social Identity Theory and Role Congruity Theory specify the boundary condition on that mechanism: where the female professional role is incongruent with prevailing role prescriptions, women discount formal provisions and reciprocate only what is visibly enacted by proximate actors [6,11]. Combining the three yields a testable conversion logic. Resources predict engagement when they are enacted and role-congruent; formal but non-enacted resources should not convert. This logic generates the moderation hypotheses directly: perceived industry support governs role congruity (H6), and perceived size-related HR infrastructure governs whether career development is infrastructurally credible (H7). It also specifies why a Western-derived resource model requires adjustment rather than transplantation in this setting [4,12].

Women’s employee engagement across different socio-economic contexts

Comparative studies highlight significant variations in women’s engagement levels across different national and economic contexts. In several developed economies, gender-inclusive policies such as extended parental leave, flexible work arrangements, and strong anti-discrimination laws are associated with higher engagement levels among women [3,22]. In contrast, South Asian countries, including India and Bangladesh, experience lower engagement due to cultural expectations, gender biases, and limited workplace flexibility [8]. In the Middle East, engagement is further restricted by legal and structural barriers to women’s employment [13]. While Western literature predominantly focuses on corporate leadership and organizational policies as primary drivers of engagement, research from developing countries emphasizes the role of societal norms, workplace safety, and work-life balance [9,14]. This study integrates these diverse perspectives to create a holistic conceptual framework that examines women’s employee engagement in Bangladesh.

The influence of socio-cultural norms on women’s engagement

Socio-cultural factors significantly shape women’s engagement in the workforce, particularly in patriarchal societies. In many South Asian countries, traditional gender roles assign women the primary responsibility for household duties, limiting their career growth and workplace engagement [4]. Studies in India and Pakistan have shown that conservative societal expectations often prevent women from pursuing leadership roles, reducing their professional commitment and career aspirations [15]. In contrast, research in Western countries suggests that societal support for gender equality is associated with higher workplace engagement among women [3]. While organizational policies can mitigate some of these challenges, the deeply embedded cultural norms in Bangladesh still restrict women’s ability to fully engage at work.

Despite these socio-cultural constraints, some women in Bangladesh actively engage in the workforce due to economic necessity and changing social attitudes [14]. However, engagement levels remain lower in industries where rigid gender norms persist, such as manufacturing and public administration, compared to more progressive sectors like IT and telecommunications [8]. Studies in East Asian countries, such as Japan and South Korea, indicate that gradual shifts in cultural attitudes, coupled with corporate initiatives to support women, can improve engagement levels over time [16]. Therefore, this study hypothesizes:

1. H1: Socio-cultural influences are positively associated with women’s employee engagement.

The role of organizational culture and policies in women’s engagement

Organizational culture plays a crucial role in shaping employee engagement, particularly for women facing systemic workplace barriers. A culture that values inclusivity, gender equity, and diversity can foster higher engagement among female employees [9]. Research from European firms suggests that organizations with structured gender-sensitive policies, such as mentorship programs and bias training, report higher engagement levels among women [17]. Conversely, in many South Asian countries, organizations often lack formal policies supporting women’s career advancement, leading to disengagement and high turnover rates [18].

While global corporations have increasingly implemented diversity and inclusion programs, many Bangladeshi firms lag in this regard due to weak regulatory enforcement [10]. Research on gender-diversity HRM practice and transformational leadership in Bangladeshi organizations indicates that engagement among women employees is higher where gender-inclusive policies, such as flexible work arrangements and anti-harassment measures, are actually implemented [7,17]. Given the strong relationship between organizational policies and engagement, this study hypothesizes:

1. H2: Organizational culture and policies are positively associated with women’s employee engagement.

Leadership and managerial support as predictors of engagement

Leadership and managerial support have been widely recognized as key determinants of employee engagement [5]. Transformational leadership, characterized by mentorship, empowerment, and inclusive decision-making, has been found to enhance engagement levels among women in Western firms [19]. However, in South Asia, where hierarchical leadership styles dominate, women often report lower engagement due to a lack of supportive management and gender bias in leadership roles [14].

In Bangladesh, studies indicate that women’s engagement is significantly higher in organizations where leaders actively support gender diversity initiatives [20]. Research in India and Sri Lanka similarly highlights that managerial encouragement and mentorship programs contribute positively to women’s job satisfaction and engagement [21]. Based on these findings, this study proposes:

1. H3: Leadership and managerial support are positively associated with women’s employee engagement.

Work-life integration and its impact on engagement

Work-life integration is one of the most significant predictors of women’s employee engagement, particularly in societies where women juggle professional and domestic responsibilities [22]. Research in Europe and North America demonstrates that flexible work arrangements, such as remote work, part-time options, and extended parental leave, are positively associated with women’s engagement levels [3]. Conversely, studies in South Asia indicate that a lack of work-life balance contributes to burnout, job dissatisfaction, and disengagement [8].

In Bangladesh, flexible work policies remain limited outside multinational corporations, restricting women’s ability to maintain a sustainable work-life balance [23]. The absence of such policies in many traditional sectors forces women to exit the workforce prematurely, resulting in lower overall engagement levels [21]. Given these trends, this study hypothesizes:

1. H4: Work-life integration is positively associated with women’s employee engagement.

Career development opportunities and women’s engagement

Career development opportunities, including mentorship programs, training, and promotions, are critical to sustaining long-term engagement among women [24]. In Western contexts, organizations that offer structured career development programs see higher retention and engagement rates among female employees [5]. However, in Bangladesh, career growth opportunities for women are often constrained by gender biases in promotion and leadership selection [8].

Comparative evidence from Asian economies indicates that investment in female employees’ career advancement is associated with higher engagement, job satisfaction, and productivity [24,29]. Given the significant gap in career development opportunities for women in Bangladesh, this study hypothesizes:

1. H5: Career development opportunities are positively associated with women’s employee engagement.

Industry sectors as a moderator of socio-cultural influences on engagement

Industry sectors play a pivotal role in determining how socio-cultural influences affect women’s engagement. Research suggests that industries with a high degree of gender segregation, such as manufacturing and construction, impose greater socio-cultural constraints on women than service-oriented or knowledge-based sectors [14]. In Bangladesh, the garment industry, which employs a significant number of women, is characterized by rigid hierarchical structures, workplace safety concerns, and gender-based discrimination, leading to lower engagement levels [23]. In contrast, industries such as information and communication technology (ICT) have demonstrated higher engagement levels among female employees due to greater flexibility, skills-based career progression, and a more inclusive work culture [9]. This pattern aligns with global trends, where women in tech-driven and service sectors report better engagement due to structured HR policies and career growth opportunities.

However, studies from India and Pakistan indicate that even within progressive industries, gender biases persist, affecting women’s perceptions of inclusion and engagement [15]. Adiza et al. [14] found that women in the financial sector in South Asia faced barriers to career progression arising from unspoken bias and workplace politics, which were associated with lower engagement. In contrast, European industries have increasingly focused on gender-sensitive strategies, resulting in more equitable engagement patterns [3].

Sector matters here not as a control but as the setting that fixes role congruity. In highly segregated sectors, the female professional role is prescriptively incongruent, so socio-cultural norms should remain salient inside the workplace and should condition engagement [6,11,14,23,30]. In knowledge-based sectors, role prescriptions are weaker and organizational resources should dominate [9]. Sector is therefore expected to change the strength of the socio-cultural path, not merely to shift its mean. Accordingly, this study hypothesizes:

1. H6: Perceived industry gender-equity support significantly moderates the relationship between socio-cultural influences and women’s employee engagement.

Organizational size as a moderator of career development and engagement

The size of an organization significantly shapes its ability to provide career development opportunities, which in turn shapes the engagement of female employees. Large multinational corporations (MNCs) tend to offer structured career development programs, mentorship initiatives, and gender-inclusive policies that enhance engagement [10]. In contrast, small and medium-sized enterprises (SMEs) often face financial and administrative constraints, limiting their ability to invest in workforce development, particularly for women [20,25]. Studies in developed economies, such as the United States and Germany, indicate that women in larger firms benefit from leadership training, sponsorship programs, and skill enhancement opportunities, leading to greater engagement [26]. However, in developing economies, the relationship between career growth and engagement is less straightforward, as systemic barriers such as workplace discrimination and rigid promotion structures hinder women’s access to leadership roles [8].

In Bangladesh, corporate giants like Grameen Bank and BRAC have successfully implemented career development programs that have positively influenced female employee engagement [27,28]. However, SMEs, which constitute a significant portion of Bangladesh’s economy, often lack the HR infrastructure to support career progression for women [29]. Similar findings have been reported in South Asia, where women in smaller firms face limited mentorship opportunities, leading to disengagement [21].

Organizational size is used here as a proxy for HR infrastructure, and infrastructure is what makes a career development resource credible. Under Social Exchange Theory, a promotion pathway is reciprocated only if it is believed to be real [5]. Large organizations possess the formal promotion criteria, mentoring structures, and training budgets that make the pathway believable [10,26–28], whereas SMEs frequently announce development without the systems to deliver it [20,25,29]. Size should therefore condition whether career development converts into engagement, which is a statement about mechanism rather than about firm scale. Accordingly, this study hypothesizes:

1. H7: Perceived size-related HR infrastructure significantly moderates the relationship between career development opportunities and women’s employee engagement.

Women’s employee engagement is shaped by socio-cultural norms, organizational policies, leadership support, work-life integration, and career development. While Western studies emphasize workplace policies, research in Bangladesh highlights the added impact of cultural constraints [30]. Industry sectors and organizational size moderate these relationships, influencing engagement levels across different workplaces [8]. Existing research lacks insight into how these factors interact in developing economies. This study fills that gap, providing a framework to enhance gender equity in workforce engagement.

Conceptual model of women employee engagement

Fig 1 illustrates the key factors influencing women’s employee engagement, categorized into six main areas: Career Development Opportunities (training programs, promotion opportunities), Socio-cultural Influences (societal norms, cultural expectations), Leadership and Managerial Support (leadership style, managerial support), Organizational Culture and Policies (company values, policies, and procedures), and Work-Life Integration (flexible work hours, remote work options). Additionally, Moderating Effects such as industry sectors and organizational size influence these factors. The diagram visually represents how these elements contribute to enhancing women’s engagement in the workplace.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Factors influencing women’s employee engagement.

https://doi.org/10.1371/journal.pone.0359645.g001

Ethics statement

This study was conducted in strict accordance with the ethical standards and guidelines set forth by the university. Ethical approval was obtained from the University Research Committee (Ethics Number: IUS/Regi.Office/Letter/2024/53), and written informed consent was secured from all participants before data collection.

Sample and data collection

Sector and size coverage.

Because the study tests moderation by sector and by organizational size, coverage targets were set in advance rather than assessed after the fact: approximately 60 completed questionnaires per sector, and a minimum of 40 per size band (SME, large domestic, MNC), with fieldwork in each stratum continuing until the target was met. Organizational size was classified by employee headcount following national SME policy thresholds, consistent with the size-based comparisons reported in the Bangladeshi SME literature [18,20,29]. The achieved distribution is reported in Table 1. These quotas were set to secure adequate cell sizes for the interaction tests. They do not make the sample probabilistic, and no claim of population representativeness is made; purposive sampling is used here as appropriate to a predictive, theory-development design [32,37].

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Respondent distribution by industry sector and organizational size (n = 242).

https://doi.org/10.1371/journal.pone.0359645.t001

Measurements

The measurement items for all constructs were adapted from existing literature to ensure construct validity and reliability. The study examines the relationships between women’s employee engagement (WE) and its key determinants: socio-cultural influences (SC), organizational culture and policies (OC), leadership and managerial support (LM), work-life integration (WI), and career development opportunities (CD). Additionally, the study evaluates the moderating effects of perceived industry gender-equity support (IS) and perceived size-related HR infrastructure (OS) on engagement.

A structured five-point Likert scale (1 = strongly disagree to 5 = strongly agree) was used to collect responses. Measurement items were adapted from well-established scales, including the Utrecht Work Engagement Scale (UWES) for engagement [1], Socio-Cultural Influences [4,8,14], organizational culture [9,17,18,34], leadership and managerial support [19,20,35], work-life integration [8,21,22], and career development opportunities [8,17,24], perceived industry gender-equity support [3,23], perceived size-related HR infrastructure [10,20,29]. The measurement model was assessed for indicator reliability, internal consistency, convergent validity, and discriminant validity [39–42,44].

Table 2 presents key variables and measurement items related to women’s employee engagement in organizational settings. It includes Women Employee Engagement (WE) as the dependent variable, with Socio-Cultural Influences (SC), Organizational Culture and policies (OC), Leadership and Managerial Support (LM), Work-Life Integration (WI), and Career Development Opportunities (CD) as independent variables. Industry Sectors (IS) and Organizational Size (OS) serve as moderators. Each item is coded and sourced, providing a structured framework for analyzing factors influencing women’s workforce engagement.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Definition of the operational variables and measurements.

https://doi.org/10.1371/journal.pone.0359645.t002

Partial Least Squares Structural Equation Modeling (PLS-SEM) was used for hypothesis testing, as it is suitable for research involving complex models with latent variables [36]. PLS-SEM was selected over covariance-based SEM (CB-SEM) for four reasons. First, the objective is prediction and theory development in an underexplored setting rather than the confirmation of an established covariance structure [36], [37]. Second, the model contains two latent interaction terms, estimated here with the product-indicator approach, which PLS accommodates without the identification and convergence burden that CB-SEM would carry at n = 242 [32]. Third, the indicators are ordinal Likert items whose distributions depart from multivariate normality, violating the maximum likelihood assumptions on which CB-SEM depends [42]. Fourth, the aim is to maximize explained variance in the endogenous construct rather than to minimize the discrepancy between observed and implied covariance matrices [32,40]. Estimation used SmartPLS with the path weighting scheme and 5,000 bootstrap resamples with bias-corrected and accelerated confidence intervals [32].

Results of the measurement model

The results for item loadings, Cronbach’s alpha, composite reliability, and average variance extracted are shown in Table 3. The scales in the questionnaire were assessed against the generally established standards for validity and reliability. In order to confirm if each factor had a significant factor loading with the variables, we examined the factor loading values. Their value must be more than 0.40 for a sample size of more than 200 [38]. Cronbach’s alpha and CR have minimal acceptable limits of more than 0.70 [39,40].

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Convergent validity.

https://doi.org/10.1371/journal.pone.0359645.t003

Additionally, AVE needs to be greater than 0.50 [41,42]. All indicators satisfied the assumed requirements, according to the analysis’s findings. This indicates that the study’s use of the structural equation model was suitable.

Recent criticism has emerged regarding the Fornell-Larcker criterion, which is unreliable in detecting the absence of discriminant validity in everyday research contexts [43,44]. An alternative approach based on the multitrait-multimethod matrix, known as the heterotrait-monotrait ratio of correlations (HTMT), is recommended to assess discriminant validity [44]. The associated Monte Carlo simulation study demonstrated the superior performance of this method. Accordingly, we have employed this recommended approach to test discriminant validity, with the results presented in Table 4. HTMT values exceeding 0.85, or 0.90, indicate potential issues with discriminant validity [45]. As all HTMT values in this study were below the 0.90 threshold and the 0.85 threshold, as shown in Table 4, discriminant validity has been confirmed [45].

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Discriminant validity (HTMT ratio).

https://doi.org/10.1371/journal.pone.0359645.t004

Common method bias

Because all constructs were measured with a single self-report instrument, common method bias was assessed using the full collinearity approach. All full-collinearity VIF values ranged from 1.58 to 2.91, below the 3.3 threshold, indicating that common method bias is unlikely to distort the estimates [46].

Goodness of fit

The model’s effectiveness and fit were evaluated using the Coefficient of Determination (R²), the Standardized Root Mean Squared Residual (SRMR), and the Normed Fit Index (NFI). An R² value between 0.25 and 0.50 is considered optimal, with any value above 0.20 deemed acceptable [47]. Additionally, an R² value closer to 1 indicates more substantial explanatory power, with 0.25 serving as the threshold for meaningful effect sizes in path models [32]. The present study’s R² value of 0.882 indicates substantial explanatory power for the endogenous construct [32,47]. R² is a measure of explained variance and is not a significance test; the significance of individual paths is reported in Table 6.

Adjusted R², which accounts for the number of independent variables in a regression model, is also reported [48]. This adjustment enhances the robustness of the model, aligning with the comprehensive analytical approach used in this research. Regarding model fit, a model is well-fitted if the NFI value approaches one and the SRMR is below 0.08 [32,48]. The study’s findings include an NFI value of 0.915, close to the ideal of 1, and an SRMR value of 0.066, comfortably below the 0.08 threshold. These metrics, as outlined in Table 5, indicate that the model demonstrates an acceptable and reliable fit, consistent with the guidelines [47]. Because these indices are of limited diagnostic value in a composite-based framework, model quality is assessed primarily through R², f², and Q² [32,48].

Download:

• PNG
  larger image
• TIFF
  original image

Table 5. Model fitness.

https://doi.org/10.1371/journal.pone.0359645.t005

Hypothesis testing

The results of the hypothesis testing in Table 6 on women’s employee engagement in Bangladesh reveal significant insights into the factors associated with women’s employee engagement. Organizational culture and policies (H2, p < 0.001), leadership and managerial support (H3, p < 0.001), work-life integration (H4, p = 0.001), and the interaction between perceived size-related HR infrastructure and career development opportunities (H7, p = 0.017) are positively associated with women’s engagement. However, socio-cultural influences (H1, p = 0.724), career development opportunities (H5, p = 0.395), and the interaction between industry sector and socio-cultural influences (H6, p = 0.314) were not statistically significant, indicating that traditional societal expectations and career growth prospects are not directly associated with engagement levels in the Bangladeshi context.

Download:

• PNG
  larger image
• TIFF
  original image

Table 6. Structural model.

https://doi.org/10.1371/journal.pone.0359645.t006

Collinearity and suppression diagnostics

Notably, the standardized path coefficient for organizational culture and policies (H2) is greater than 1.0 (β = 1.416, p < 0.001). This indicates the presence of a suppression effect, which frequently occurs in PLS-SEM when two highly correlated predictors (OC and WI, HTMT = 0.780) are entered into the structural model simultaneously, causing the path coefficient of one variable to be mathematically amplified. To evaluate potential multicollinearity and verify the validity of our path estimates, the structural model’s inner Variance Inflation Factors (VIFs) were examined. The inner VIFs were: SC (1.58), OC (2.84), LM (2.45), WI (2.91), and CD (2.15). Because all inner VIFs are comfortably below the conservative threshold of 3.3 [46], critical multicollinearity is not present, confirming that our model estimates are stable and mathematically sound.

Effect sizes and predictive relevance

Effect sizes (f²) and predictive relevance (Q²) are reported in Table 7. Q² was obtained by blindfolding with an omission distance of seven, and Q²predict was obtained using PLSpredict [48]. The Q² value for women’s employee engagement is 0.548, exceeding zero and confirming the model’s predictive relevance [32,48]. Effect sizes are interpreted against the 0.02, 0.15, and 0.35 thresholds for small, medium, and large effects [32].

Download:

• PNG
  larger image
• TIFF
  original image

Table 7. Effect sizes and predictive relevance.

https://doi.org/10.1371/journal.pone.0359645.t007

The H7 interaction was probed at one standard deviation below the mean, at the mean, and at one standard deviation above the mean of perceived size-related HR infrastructure (OS) [32]. The simple slope of career development on engagement is β = −0.011, p = 0.793 (95% CI: [−0.093, 0.071]) in smaller organizations (low perceived size-related HR infrastructure, −1 SD), β = 0.037, p = 0.395 (95% CI: [−0.047, 0.121]) at the mean of perceived size-related HR infrastructure, and β = 0.085, p = 0.012 (95% CI: [0.018, 0.152]) in larger organizations (+1 SD), confirming that the positive association between career development opportunities and women’s employee engagement emerges only as the perceived organization-level size-related HR infrastructure increases and becomes highly credible.

Theoretical contributions

This study’s theoretical contribution is a gendered conversion argument rather than a contextual extension. The Job Demands-Resources model specifies which constructs are demands and which are resources [1]; Social Exchange Theory specifies why resources become engagement [5]; Social Identity and Role Congruity Theory specify when that conversion is blocked [6,11]. Our results indicate that in this setting resources convert only when they are relationally enacted and infrastructurally credible. Leadership support, inclusive culture, and workable schedules convert because they are experienced directly [9,17,19]. Career development converts only where perceived size-related HR infrastructure makes it believable [10,29]. Socio-cultural constraint does not appear at the engagement stage because it has already acted at the selection stage [4,50]. This specifies a mechanism that the four theories imply, but none states alone, and it is testable outside Bangladesh.

Additionally, this research contributes to Social Exchange Theory by illustrating that while employees reciprocate organizational support with higher engagement, this reciprocity differs for women in male-dominated or culturally restrictive work environments [5]. Women’s engagement is not solely a response to general organizational support but is also influenced by leadership inclusivity, gender-sensitive policies, and work-life integration strategies.

The study also extends Social Identity Theory and Role Congruity Theory by showing that women in professional settings often navigate identity threats and biases that are associated with their engagement levels [6,11]. This research demonstrates that workplace policies and leadership support can mitigate the adverse effects of gender-based biases, enabling women to engage more fully despite existing stereotypes and cultural expectations.

Furthermore, the study addresses a critical gap in the literature by contextualizing these theories within a developing economy like Bangladesh. While previous frameworks primarily reflect Western corporate cultures, this research emphasizes the role of socio-cultural constraints in shaping women’s engagement in South Asian workplaces. It highlights that women’s engagement is not solely driven by individual agency but is significantly influenced by gender norms, structural inequalities, and industry-specific challenges.

Practical implications

The findings of this research offer several practical implications for organizations, policymakers, and stakeholders seeking to enhance women’s employee engagement in Bangladesh and South Asia. First, our findings suggest that the positive associations of inclusive organizational culture (H2), leadership support (H3), and work-life integration (H4) with engagement are remarkably robust across different industry sectors, as perceived industry gender-equity support did not significantly moderate these relationships (H6). This implies that organizations across all sectors have powerful, direct leverage to enhance female engagement through local, firm-level HRM interventions. Organizations must prioritize leadership and managerial support, as leadership support shows the strongest association with women’s engagement, necessitating gender-sensitive leadership training and mentorship programs. Strengthening work-life integration policies, such as flexible work arrangements, childcare facilities, and parental leave, is crucial to supporting female employees in balancing professional and personal responsibilities. The study also highlights that career development opportunities alone are insufficient without addressing structural gender biases and the organizational infrastructure needed to make development credible, requiring HR policies to focus on transparent promotions, equitable pay, and anti-discrimination measures. Furthermore, perceived size-related HR infrastructure must be factored into engagement strategies, as the association between career development opportunities and engagement is significant only in environments with strong HR infrastructure. Lastly, stronger government and institutional policies are essential to enforcing gender-equality regulations, incentivizing women-friendly corporate policies, and fostering public-private collaboration for best practices in female workforce engagement. Addressing workplace and structural barriers, this study provides a roadmap for improving women’s engagement, ultimately contributing to economic development and gender equality in Bangladesh and South Asia.

Limitations and future research directions

First, the sample is non-probabilistic and drawn through purposive recruitment via organizational and professional channels; the findings describe associations within this sample and cannot be generalized to the population of employed women in Bangladesh [32,37]. Second, the design is cross-sectional, so causal direction and temporal ordering cannot be established, and reverse causality cannot be excluded: more engaged women may perceive their leaders and culture more favorably rather than the reverse. Third, the socio-cultural scale mixes internalized normative expectation with perceived organizational bias [4,8,14], and this heterogeneity may attenuate its estimated coefficient; disaggregating the construct is a priority for future work. Fourth, although quota targets secured adequate cells for the descriptive sample characterization, the continuous operationalization of our moderators in the PLS-SEM interaction tests means that categorical cell sizes did not limit the statistical power or precision of our estimates, which are reported with full confidence intervals. Fifth, the geographical focus on Bangladesh may restrict the generalizability of the findings to other regions or socio-cultural contexts. Sixth, the reliance on self-reported data introduces the potential for bias, as participants may have provided socially desirable responses or interpreted engagement differently based on individual experiences.

Future research should expand the scope beyond Bangladesh to explore women’s employee engagement in diverse geographical regions and industries. Conducting longitudinal studies would help capture the evolving nature of employee engagement, especially in response to changing organizational policies and socio-cultural influences. Additionally, exploring specific mediating factors, such as age, education level, or professional experience, could offer a more nuanced understanding of how different groups of women experience engagement. Disaggregating socio-cultural influence into internalized norms and perceived bias, and testing the selection-stage interpretation advanced here on a sample that includes women outside formal employment, would provide a direct test of the mechanism proposed in this study. The role of digital transformation and the increasing prevalence of remote work, particularly in a post-pandemic context, presents another promising avenue for future research, as these factors may significantly alter the landscape of women’s engagement in the workplace.