Monitoring And Evaluation Tools For Sanitation: A Comparative Study Of Mobile Phone And Traditional Paper Based Survey Of Defecation Practices In The Ningo-Prampram District Of Ghana

ABSTRACT

Introduction

The Millennium Development Goal targets for sanitation seems unachievable by the 2015 deadline. Therefore, the 2010-2015 Joint Monitoring Programme strategy requires sustainable and continued efforts at all levels to improve this. However, there are data challenges and limitations from existing Monitoring and Evaluation (M & E) survey tools. In most developing countries such as Ghana, data collection is traditional paper based. However, mobile phone use and access is high and therefore the need for new innovative M & E tools. This study compared the validity, reliability, feasibility and cost effectiveness of using mobile phone SMS M & E survey tool and a traditional paper based survey tool in a rapidly expanding peri-urban setting in Prampram, in the Ningo-Prampram District of the Greater-Accra Region of Ghana, for monitoring sanitation practices.

Methods

The study was a comparative prospective cohort study and used both qualitative and quantitative methods for data collection. Four hundred and fifty-eight respondents were randomly selected based on data from the Dodowa Health Research Centre. Validity was measured using Cronbach’s alpha and Kappa for varying levels of agreement for both tools. Reliability was measured using the test-retest method. Various response rates were compared for feasibility on levels of acceptability and ease of use. The total costs of the tools were estimated and compared.

Results and conclusions

There were no significant differences in the reliability and validity of the paper based tools and the mobile phone SMS tools. However, although the mobile phone SMS M & E survey tool is feasible and potentially acceptable in terms of reliability, validity, cost and timely collection of

iii

sanitation data for effective policy formulation, there are challenges, which could be further addressed.