3.2 Research design

The study adopted the case study research method [47]. Purposive sampling [48] was used to select 12 communities from the two municipalities. Firstly, the study area was zoned into two clusters namely: Berekum municipality and Dormaa municipality. Secondly, purposive sampling was used to select twelve (12) areas from the two municipalities for the survey. Through a mixed methods design [49], both qualitative and quantitative research methods were used for the data collection and analysis. A household survey was conducted with household respondents to understand solid waste management and service delivery across the 12 selected communities. Using Slovin's formula: n = N/1 + N(α)2, where 'n' is the sample size, 'N' is the total number of households, 'α' is the margin of error (0.05), a total sample size of 312 households across the 12 case study communities were randomly selected and involved in the household survey. The sample size of 312 was divided equally among the 12 selected communities. This gave a sample size of 26 for each selected area. Finally, accidental sampling method was used to select the respondents for interview. That is, the first person to be contacted in each selected house was interviewed. If the first person contacted was not ready, the next available person was interviewed. To gather statistical and policy information on solid waste management and service delivery the two municipalities, semi-structured interviews were also undertaken with an official of the Assemblies (Berekum and Dormaa), responsible for the environmental health and waste management of the municipalities. In analysis, the study used a cross-case analysis procedure to analyze the interview data. In this approach, responses to a common question from all interviewees in each category are analyzed together. The findings of the study were validated and verified through focus group discussions with household respondents in each of the 12 case study communities. This approach was appropriate in addressing the inconsistencies that had occurred during the data analysis.
