Messenger students' engagement scale: Community perspectives on school-based malaria education in Ethiopia

A two-year school-based malaria education intervention was developed to engage students as implementers in malaria prevention and control in the Oromia region, Ethiopia. The current study aimed to validate messenger students' engagement scale (MSES) in malaria education. The scale development process was done stepwise. Multiple behavioural theories were examined to derive possible domains of engagement. Next, a pool of items was developed by linking the domains with malaria target behaviours. The items critically reviewed, pretested and refined for clarity and appropriateness. A cross-sectional survey of 451 sample households with school-going children was conducted in five districts of the Jimma-Zone in March-2019. Exploratory factor analysis (EFA)/principal component analysis (PCA) was executed to evaluate the construct validity of the scale. Rotated factor loading coefficients of ≥0.4 were retained. Items loaded on multiple factors were retained on the factor with a higher loading score. Cronbach's alpha of 0.7 was used as the cutoff point for reliability. Discriminative validity was declared based on Pearson correlation (between the extracted factors) coefficients that were moderate (r < 0.7), and less than the respective variance explained (VE) by each factor. The validity of convergence of domains with the overall MSES assessed (0.4 < r < 0.9). Logistic regression for key malaria preventive practices was conducted to assess predictive validity. The study explored a 24-item MSES in six domains of malaria education: reminding, supporting, monitoring, messenger credibility, role modelling and norm setting. The domains explained 67.82% variance of MSES, with a reliability of 93.3%. The factors were convergent with the overall scale (r = 0.764-0.834). Most factors were discriminative, with moderate correlation to each other. Adjusted odds ratios showed engaging credible students in reminding malaria preventive messages and modelling practices predicted exposure to message, insecticide-treated net (ITN) utilisation and cleaning surrounding. The current MSES is reliable, valid and predictive of malaria preventive practices.