Abstract/Overview
Monitoring and evaluation are important management functions in ensuring that program/project objectives remain on course and the expected outcomes are realized. However, there is limited information on how different monitoring and evaluation (M&E) design components affect program outcome(s). Specifically, the extant literature is not clear on how frequency of monitoring regimes affects ultimate program outcomes. The study‟s main objective was to assess the effect of monitoring frequency regimes on program outcomes in the Dispensers for Safe Water program in Kakamega County. The study focussed on three program outcomes namely; adoption of chlorine dispensers by target communities, dispensers‟ empty rates and promoters‟ capacity to undertake promotional activities. The specific objectives were: to evaluate the effect of monitoring frequency on community‟s adoption of chlorine dispensers; to assess the effect of monitoring frequency on the level of dispenser‟s empty rates; and to examine the effect of monitoring frequency on community promoter‟s capacity to undertake promotional activities. The dynamic capability theory was used to conceptualize the connection between effects of changing M&E system on program outcome(s).The study adopted a comparative cross-sectional research design and used both quantitative and qualitative data. The study relied on secondary data from 15,336 households collected by Dispenser for Safe Water (DSW) program between January 2013 and December 2019, while qualitative data was collected from 17 key informants consisting of 12 community promoters and 5 DSW management team members. Quantitative data was analysed through descriptive and inferential statistics which included frequency, mean, standard deviation and percentages. In addition, regression analysis was done to test for significance of effect of frequency of monitoring on each of the three program outcomes. Analysis of variance (ANOVA) was used to compare means and proportions in different monitoring regimes. Thematic analysis was used to analyse qualitative data from management team and community promoters. In terms of chlorine dispenser adoption, shifting from one-month regime to two-month regime had no significant effect on the level of adoption. However, changing frequency of monitoring from one-month to three-month monitoring regime could reduce the proportion of households using the dispenser by between 4.6% and 20.2%. Therefore, changing frequency of monitoring by more than one month had significant negative effect on chlorine adoption. On average, increasing the frequency of monitoring by one month increased the rate of dispenser empty rates by 6.9% while increasing by two months increased the proportion of empty dispensers by 14.6%. The promoters‟ capacity to undertake promotional activities did not significantly change when the program increased their frequency of monitoring by one month but improved by 12.4% if it was increased by two months. Program management attributed increased chlorine adoption and decreased dispenser empty rates during one-month or two-month monitoring regimes to timely and informed decisions derived from frequent monitoring reports. Reduced community-promoter engagement during one- or two-month regimes was attributed to the inability of promoters to offer more time for engagement beyond the monitoring process. DSW was justified to implement the two-month regime on the basis of cost of monitoring since there is no significance difference between one-month and two-month regimes. The findings of this study should help program implementers, monitoring and evaluation personnel, and policymakers to understand how frequency of monitoring affects each program‟s key indicators when developing monitoring and evaluation designs.