Propensity score methods to analyse attitudinal change during the early stages of the Covid-19 pandemic within the African Country of Malawi

Abstract

AbstractThe objective of this study is to determine attitudinal changes within the African country of Malawi during the early stages of the COVID-19 pandemic between the months of May and June 2020. This study is relevant and will serve as a baseline that will allow for tracking attitudinal change throughout the COVID-19 pandemic. This will give insight to areas where extra resources, information, healthcare workers, and information can be applied to the Malawi population should a similar event occur. Propensity score methods are used for this study as it is a methodology that allows for making causal statements about treatment effects, known here as the participants attitudes to COVID-19. Both single level and multilevel propensity score modelling methods are undertaken as there is an identified need for multilevel modelling which was undertaken using a fixed effects approach. This study highlights that the multilevel modeling methodology used for any study be carefully assessed and not just be granted in using a random effects methodology. This study identifies the utility of propensity score methods in highlighting methodological issues of a survey study. Propensity score methods highlighted a large, unbalanced dataset that shows possible bias in the undertaking of the survey. Further, Featured Indicator Variables are created in aiding contextual effect modelling and the interpretation of the results. This study although limited to the early stages of the COVID-19 pandemic within Malawi, serves as a baseline for longitudinal studies and interrupted time–series modelling. This study highlights determining attitudinal change, not only for Malawi but the methodology can be used for other countries in prediction, directing resources, information, health care workers, and services should a similar event occur.