SRI, or socially responsible investment, is a relatively new concept used to describe an investment that considers social, ethical, and environmental concerns. The purpose of this study is to investigate if collectivism, concern for the environment, financial performance, and awareness of SRI influence an individual’s propensity to invest in socially responsible investments (SRI). Secondly, the study evaluates the influence of the TPB (Theory of Planned Behavior) model constructs, attitude, subjective norms, and perceived behavioral control on the SRI investment intention of individual investors. A structured questionnaire was used to collect data on 449 individual investors for this cross-sectional investigation. The data were then analyzed further with a two-step structural equation modeling technique performed in Smart PLS 3.2.9. The PLS-SEM analysis found that collectivism, environmental concerns, financial performance, and awareness of SRI all had significant positive effects on attitudes toward SRI, which, in turn, resulted in SRI investment intention. Further, subjective norms and perceived behavioral control had a significant impact on individuals’ intentions regarding SRI.

BACKGROUND Hospitals use Facebook for education, communication, relationship building and business promotion. Hospitals use their dedicated Facebook pages to achieve communication objectives with promotional, emotional and informational orientation. Despite increasing use of Facebook among hospitals, there are limited research evidences available regarding Facebook post characteristics, content-theme and audience engagement with respect to Indian private hospitals’ Facebook pages. OBJECTIVE This study explored relationship between Facebook post characteristics and engagement outcomes for selected private multispecialty hospitals in India. METHODS Content analysis was conducted using posts published during 2024 on official Facebook pages of 10 selected private multispecialty hospitals. Final analysis included 3,137 posts. Post characteristics included in this study were post format, call-to-action (CTA) presence, content orientation and weekday or weekend posting. Content orientation was coded as informational, relational and promotional post content. Total engagement was calculated by combining likes, comments and shares. Engagement was standardized per 10,000 page followers and it was transformed using natural logarithm. Main effects general linear model examined adjusted associations while controlling for stable differences across hospital pages. Type III tests, partial eta-squared, estimated marginal means, Bonferroni-adjusted comparisons, HC3 heteroscedasticity-consistent inference and residual-based sensitivity analysis were used. RESULTS Model was significant at F(14, 3122) = 152.58, p < .001. It explained 40.6% of the variance in log engagement. Hospital identity showed the largest association (partial η² = .386). Content orientation was the strongest post-level factor at F(2, 3122) = 53.51, p < .001, partial η² = .033. Promotional posts had the highest adjusted engagement. It was followed by relational and informational posts and all pairwise comparisons were found significant. Video posts showed somewhat higher adjusted engagement than non-video posts whereas CTA presence was not significant. Weekday posting activity was significant in primary model but not in the residual-based sensitivity analysis. CONCLUSIONS Engagement outcomes were associated strongly with hospital specific context and content orientation rather than with simple structural or timing characteristics of Facebook posts. Findings are limited to selected hospital Facebook pages and indicate associations rather than causal effects. Engagement metrics should be interpreted in relation to hospital context, content purpose and audience size. Lower engagement with informational posts should not be interpreted as evidence of lower value of educational or public health.

This study presents a comparative analysis of various machine learning algorithms for credit risk assessment. The algorithms were tested on two credit datasets: German Credit Dataset and Australian Credit Dataset. The performance of the algorithms was evaluated based on several metrics, including sensitivity, specificity, accuracy, F-score, and Kappa. The results showed that the FCPFS-QDNN algorithm outperformed other algorithms in both datasets, achieving high accuracy, sensitivity, specificity, and F-score. On the other hand, the ACO Algorithm and Multilayer Perceptron algorithms were found to perform poorly in both datasets. The findings of this study have significant implications for credit risk assessment in banking and financial institutions. The study recommends the use of the FCPFS-QDNN algorithm for credit risk assessment due to its superior performance compared to other algorithms. Request access from your librarian to read this chapter's full text.