The nexus between financial literacy encompassing debt management, financial planning, and budgeting and MSE performance is well-documented. Research consistently demonstrates that entrepreneurs with superior financial management capabilities are better equipped to make informed decisions, optimize resources, and enhance profitability ( Lusardi & Mitchell, 2023). For instance, effective debt management ensures solvency and builds a positive credit history, which is crucial for business growth. However, the direct application of these skills can be severely limited in environments characterized by financial exclusion. As recently argued by Demirgüç-Kunt and Klapper (2023), the benefits of financial knowledge are often contingent on the availability of formal financial products; without access, even the most skilled entrepreneurs may find their growth aspirations curtailed, leading to a “capability-resource mismatch.”

This gap underscores the critical mediating role of access to finance. It is theorized that access to finance acts as a conduit, transforming latent financial competencies into tangible business outcomes. Recent empirical studies support this mediation hypothesis. For example, a study by Adomako et al. (2024) on SMEs in emerging economies found that access to finance significantly mediated the relationship between financial literacy and innovation, suggesting that financial resources enable the execution of knowledge-based strategies. Similarly, Uddin et al. (2023) highlighted that improved financial access amplifies the positive effects of financial management skills on SME resilience during economic crises. This mediating role is particularly salient for marginalized groups, such as women entrepreneurs, who often face dual barriers of limited financial literacy and systemic exclusion from formal credit markets ( Agyapong et al., 2022).

Within this broader discourse, this study focuses on a critical yet under-researched cohort: micro and small enterprises owned by Muslim women in Mbarara City. This context is unique due to the potential interplay of gender-specific challenges, religious principles influencing financial behavior (e.g., adherence to Islamic finance), and regional economic dynamics. While prior research, such as that by Bannò et al. (2021), has explored these relationships in Sub-Saharan Africa, a granular analysis focusing on this specific demographic is lacking. Therefore, this article investigates the mediating effect of access to finance on the relationship between key financial competencies debt management skills and the performance of Muslim women-owned MSEs. By employing a rigorous partial least squares structural equation modeling (PLS-SEM) approach, this research aims to provide robust, context-specific evidence that can inform targeted policy and financial inclusion strategies.

Results in Table 1 and Figure 1 show a partial mediating effect of access to finance on the relationship between debt management skills and performance of micro and small businesses owned by Moslem women in Mbarara City.

The Sobel test in Table 1 produced a z-value of 8.6661 (p < 0.000001), which is highly significant, indicating that the indirect effect of debt management skills on business performance through access to finance is statistically significant. The 95% confidence interval for the indirect effect (a*b) ranges from 0.39136 to 0.62012, which does not include zero, further confirming the presence of a mediation effect.

The unstandardized indirect effect (a*b) was 0.50574 with a standard error of 0.05836, suggesting that for every one-unit improvement in debt management skills, approximately 0.51 units of the improvement in business performance can be attributed to increased access to finance.

In terms of effect size, the standardized coefficients indicate that the total effect of debt management skills on performance was 0.988, of which the direct effect was only 0.465, while the indirect effect through access to finance was 0.523. This means that 52.9% of the total effect of debt management skills on performance operates through access to finance, while only 47.1% operates directly.

The R ² measures further highlight this mediation: the total variance explained in performance was 0.976 (97.6%), while the indirect pathway alone accounted for 0.971 (97.1%), compared to only 0.005 (0.5%) for the direct pathway. This shows that the majority of the explanatory power comes from the mediation effect of access to finance.

Overall, the results imply that while debt management skills directly improve the performance of Muslim women-owned micro and small businesses, their greater contribution comes through improving access to finance. This indicates that financial institutions and policymakers should prioritize interventions that enhance access to finance alongside strengthening women’s debt management skills to achieve sustainable business growth. The results were illustrated in Figure 1.

Figure 1 illustrates the mediating role of access to finance in the relationship between debt management skills and the performance of micro and small businesses owned by Muslim women in Mbarara City. From the model, the total effect of debt management skills on performance was 0.988 (p < 0.05), showing a very strong and significant relationship. When access to finance was introduced as a mediator, the effect split into a direct effect of 0.465 and an indirect effect of 0.523. This indicates that 52.9% of the influence of debt management skills on performance is explained through access to finance, while 47.1% is direct. The Sobel test confirmed the significance of this mediation with a z-value of 8.6661 (p < 0.05). The 95% confidence interval for the indirect effect ranged between 0.39136 and 0.62012, excluding zero, which further supports that the mediation pathway is statistically significant. The variance explained (R ²) by the total model was 0.976 (97.6%), which is exceptionally high. Of this, the indirect path through access to finance accounted for 0.971 (97.1%), while the direct path explained only 0.005 (0.5%) of the variance. This shows that access to finance is the dominant pathway through which debt management skills translate into improved business performance.

Thus, Figure 1 demonstrates partial mediation, where debt management skills still have a direct effect on performance (0.465), but their larger impact is realized indirectly via access to finance (0.523). Results show a partial mediating effect of access to finance on the relationship between debt management skills and performance of micro and small businesses owned by Moslem women in Mbarara City. We therefore do not support hypothesis that H ₀ 4e: Access to finance does not significantly mediate debt management skills and performance of micro and small enterprise owned by Muslim women in Mbarara city.

To confirm the relationships between study variables PLS SEM was conducted as shown in Figure 2.

From Figure 2, the structural model results reveal that access to finance partially mediates the relationship between debt management skills and business performance among Muslim women-owned MSEs in Mbarara City. Debt management skills exhibit a significant positive effect on access to finance (β = 0.313, p = 0.001 < 0.05) and also maintain a direct significant influence on performance (β = 0.342, p = 0.000 < 0.05). Access to finance, in turn, significantly affects performance (β = 0.223, p = 0.002 < 0.05), although the effect size is modest.

The persistence of a significant direct path from debt management skills to performance, alongside a significant indirect path through access to finance, indicates partial mediation. This suggests that effective debt management enhances a firm’s ability to secure external financing, which in turn improves overall business performance. However, debt management skills continue to exert a strong independent influence on performance, implying that entrepreneurs’ ability to plan, control, and utilize borrowed funds contributes directly to business success beyond the availability of external finance. Overall, these findings highlight that while access to finance strengthens the impact of debt management on performance, debt management skills remain a crucial determinant of performance even in the absence of external financial access.

As shown in Table 2, the predictive relevance results indicate that the model has strong predictive accuracy, with Q ² values of 0.940 for access to finance and 0.964 for performance, both well above the recommended threshold of 0.35, thus confirming high predictive validity. The low RMSE (0.246 for access, 0.192 for performance) and MAE values (0.197 and 0.149 respectively) further demonstrate that the model predicts the observed values with minimal error, supporting its reliability. Regarding model fit, the SRMR value of 0.016 is far below the acceptable cutoff of 0.08, signifying excellent fit. Similarly, the d_ULS (0.018) and d_G (0.351) values are consistent across the saturated and estimated models, suggesting stability in the model specification. The chi-square statistic (358.059) is identical for both models, while the NFI value of 0.901 exceeds the conventional threshold of 0.90, confirming good model fit. Overall, the results demonstrate that the performance model is both highly predictive and well-fitted, supporting its robustness in explaining the relationships among the study variables.

Table 3 show that the path coefficients reveal that access to finance has a positive and significant effect on performance (β = 0.223, t = 3.168, p = 0.002), underscoring its importance as a driver of business outcomes. Debt management significantly influences both access to finance (β = 0.313, t = 3.937, p < 0.05) and performance directly (β = 0.342, t = 5.235, p < 0.05), suggesting a dual role in enhancing financial accessibility and performance. Overall, the results highlight that debt management is the strongest predictor of both access and performance

Table 4 above show that the mediation bootstrap results confirm that access to finance serves as a significant mediator in the relationship between debt management and performance. The total indirect effects are significant for debt (β = 0.070, t = 2.305, p = 0.021), contribute to performance partly through their influence on access to finance. The specific indirect effects for each pathway through access to finance mirror these results, further validating the mediating role of financial access. Examining the total effects shows that debt (β = 0.412, p < 0.05) predictors of performance, both directly and indirectly, Overall, these findings suggest partial mediation, where debt management enhance performance both directly and indirectly through improving access to finance.

Table 5 show that the coefficient of determination results indicate that the model explains a substantial proportion of variance in both access to finance and performance. Specifically, the R ² values show that debt management explain 94.4% of the variance in access to finance and 97.0% of the variance in performance, with the adjusted R ² values (0.943 and 0.969 respectively) confirming the stability and robustness of these estimates. These very high R ² values demonstrate that the model has excellent explanatory power. The f ² effect sizes further reveal the relative contribution of each predictor. Debt exhibits a moderate effect on performance (f ² = 0.313, p = 0.023) and a small-to-moderate effect on access (f ² = 0.165, p = 0.083).

The construct reliability and convergent validity results in Table 6 confirm that all study constructs meet and exceed recommended thresholds, indicating strong internal consistency and validity of the measurement model. Cronbach’s alpha values range from 0.938 to 0.968, all well above the 0.70 benchmark, demonstrating excellent reliability. Similarly, the composite reliability coefficients (ρa and ρc) for all constructs fall between 0.938 and 0.979, surpassing the 0.70 threshold and further affirming measurement consistency. The Average Variance Extracted (AVE) values, ranging from 0.940 to 0.956, are significantly higher than the recommended minimum of 0.50, providing strong evidence of convergent validity by showing that each construct explains the majority of variance in its indicators.

The Fornell–Larcker criterion results in Table 7 confirm that discriminant validity is established among the constructs ACCESS, DEBT and PERFM. The square roots of the AVE values, shown on the diagonal (ranging from 0.970 to 0.978), are all greater than the correlations between the constructs in the off-diagonal cells. This means that each construct shares more variance with its own indicators than with other constructs, satisfying the rule of the Fornell–Larcker criterion. However, it is worth noting that the correlations among the constructs are very high, mostly above 0.93, which implies that while they are statistically distinct, there is still a strong conceptual relatedness among them. Overall, the measurement model demonstrates adequate discriminant validity, allowing the study to proceed with confidence to structural model testing.

The findings of this study provide robust empirical evidence that strongly aligns with and extends the existing body of literature on MSE finance. The confirmation of a significant partial mediating effect of access to finance resonates with the foundational work of Demirgüç-Kunt and Levine (2020), who emphasized that financial access is fundamental to improving productivity, not merely a tool for expansion but a buffer for risk management. Our results empirically validate their assertion by demonstrating that over half (52.9%) of the benefit derived from debt management skills is channeled through improved access to finance. This finding directly supports the conceptual framework proposed by Allen and Rahman (2018), which posits that a “capability gap” exists where skills remain underutilized without financial resources to activate them. The struggle faced by MSEs with strong skills but poor access, as noted by Rahman and Khanam (2017), is clearly reflected in our data, where the indirect effect vastly overshadowed the direct path. This underscores that financial access is the critical bridge converting internal financial discipline, as described by Kamil et al. (2018), into external performance gains.

Furthermore, the nuanced results across different financial constructs offer a more granular understanding of the mediation mechanism. The partial mediation for debt management and financial planning and budgeting suggests these are core competencies with inherent direct value, consistent with Njuguna’s (2023) findings that such skills directly improve loan performance and free up capital. However, the full mediation observed for Islamic financial practices indicates a more specialized pathway. This suggests that the benefits of Islamic principles, such as risk-sharing and ethical compliance, are realized primarily by enhancing eligibility and trust within specific financial ecosystems, thereby facilitating access. This finding echoes the cross-country evidence of Bannò et al. (2021), who found that hybrid interventions combining financial literacy (which would include Islamic financial principles) with access facilitation are most effective, particularly for marginalized groups. The exceptionally high predictive accuracy and model fit statistics confirm the robustness of this integrated model, highlighting that for Muslim women entrepreneurs in Mbarara City, the synergistic effect of developing financial skills and ensuring access to finance, as recommended in the literature, is paramount for achieving sustainable business performance.

Based on the findings, it is recommended that financial institutions, policymakers, and development partners adopt a dual-pronged approach to support Muslim women entrepreneurs. Firstly, targeted financial literacy programs should be designed and scaled up to specifically enhance debt management, financial planning, and budgeting skills, incorporating modules on Sharia-compliant finance principles to align with their religious values. Secondly, and concurrently, financial inclusion policies must be strengthened to create more accessible, affordable, and tailored credit products for women-owned MSEs, potentially leveraging Islamic banking windows and microfinance institutions to improve credit flow. By integrating skill-building with proactive access facilitation, these stakeholders can effectively bridge the capability-resource gap, thereby maximizing the translation of financial competencies into tangible gains in business performance, sustainability, and economic empowerment.

This study acknowledges several limitations that should be considered when interpreting its findings. The cross-sectional nature of the research design captures relationships at a single point in time, preventing definitive conclusions about causality between the variables. The geographical focus on Muslim women entrepreneurs in Mbarara City, while providing valuable contextual insights, limits the generalizability of the results to other regions or demographic groups. Furthermore, the reliance on self-reported data for measuring performance and financial practices may introduce the potential for social desirability bias, where respondents might overstate their competencies or business success. Future research could address these limitations by employing longitudinal designs, expanding the geographical scope for comparative analysis, and incorporating objective financial metrics to complement subjective performance measures.