Perceptions and determinants of households' participation in a randomized evaluation on water quality testing and information in southern Ghana

In this paper, we analyze perceptions and determinants of households' participation in a randomized experiment on water quality testing and information in southern Ghana. Beneficiary households assessed the components of the intervention including its relevance and adequacy in improving understanding of water quality issues. Motivating and constraining factors to participation in the randomized experiment are also assessed. We also estimate the correlates of participation in the intervention. Social and economic benefits derived from the intervention based on perceptions are compared with impacts of the intervention using an instrumental variable approach. We found evidence that subjective analysis estimates of the effects of the intervention are higher than the objective analysis estimates. Households generally perceived the intervention to be relevant in improving their understanding of water quality issues. However, there are differing opinions based on random assignment into either child or adult treatment groups on most- and least-liked attributes of the intervention, and also motivating and constraining factors affecting participation in the intervention. The factors that statistically and significantly influenced participation in the intervention include educational attainment, ethnicity, religious denomination and marital status of the household heads, in addition to the location of residence.