A meta-analysis of CSR disclosure in Islamic banks

Purpose This study aims to conduct a comprehensive meta-analysis of Corporate Social Responsibility Disclosure (CSRD) patterns in Islamic banks to provide an aggregated view of the current state of CSRD practices. Design/methodology/approach A systematic meta-analysis approach was used, synthesizing findings from 68 empirical studies selected from the Scopus database. The final data set comprised 433 unique observations with a combined sample size of 9,459. Using a random-effects model based on the Restricted Maximum-Likelihood (REML) method, effect sizes were calculated to provide a robust estimate of overall CSRD levels. Meta-regression and sub-group analyses were conducted, considering country as the moderating factor to explore how variations in disclosure practices were influenced by geographical context. Findings The results indicate that Islamic banks, on average, disclose only 43.64% of potential CSR activities, suggesting a moderate level of engagement. Significant heterogeneity (I² = 80.97%) was observed, driven by differences in country. Meta-regression results confirmed that country significantly moderated the extent of CSR disclosure, underscoring the influence of geographic location on disclosure practices. Originality/value This meta-analysis delivers the first pooled, method-comparable estimate of CSRD in Islamic banks and statistically verifies cross-country heterogeneity by modelling country as a moderator, moving the field beyond narrative reviews.