A sampling method is a research methodology that is useful in solving daily research problems. It is a science of studying how research is to be carried out and give the work plan of research. Research Methodology therefore is a way of trying to solve research problem. The methodology included publication research, interviews, surveys and other research techniques, and included both present and historical information. When dealing with population that is too large, a sample representative of the whole population is chosen. This sample is chosen to be a sample representation of the population uniformly. This can be achieved by different methods like; simple random sampling, stratified sampling purposive sampling, cluster sampling, and others. The sampling method must be chosen is a way that if used the sampling error is minimized and the estimate of the population computed to be precise.
During research an ex post facto (“Before/After”) research design is usually employed in study survey approach. This is used when the study is ment to deal with an already implemented treatment. According to Kathuri and Pals (1993), an ex post facto is one in which the researcher, rather than creating the treatment, examines the effects of naturalistically occurring treatment that had occurred. The type of ex post facto research that was used as qualitative study and emphasis was placed to ensure that in-depth responses about research problems were well addressed according to the objectives of this study. Survey included extensive literature research / e-recourses, field research observation, interviews and questionnaires used to collect data.
Purposive and Snowball methods of sampling are usually employed in the study. Purposive sampling is applied to get the location in which the units of observation have the required characteristics. Snowball sampling was used to obtain the sample size in each location. To ensure objective and comprehensive data, selection of the informants was done using purposive sampling techniques. This technique is advantageous over probability sampling because it is assumed that not everybody in the target population is knowledgeable about the issues that the researcher intended to investigate into.
n=Z2 PQε2.Equation 1
Where: n is the sample size, Z is confidence level (α=0.05),P is the proportion of the population of interest. Variable Q is the weighting variable and is computed as1-P, and ε is an acceptable error (precision). P will be set to 0.5 since statistically, a proportion of 0.5 results in a sufficient and reliable size one the population proportion is uncertain. This led to Q of 0.5 (1- 0.5). An error of less than 10% is usually acceptable (Kothari, 2004) thus this study will take an error of 0.08 to approximate a sample size of 150 respondents.
According to Kathuri and Pals, (1993), most techniques for measuring social and psychological environment rely on verbal materials in the questionnaires and interviews. Data collection approaches was include, the use of field observations, oral interviews, questionnaires and documentary (primary and secondary) sources. Oral interviews fall under primary data collection. Secondary data in this context of study was obtained from various sources of literature such as: electronic sources, documents, research materials, articles, theses and dissertations. In order to achieve this interview schedule was used. People to be interviewed involved: Key informants, respondents, opinion leaders, council of elders, famers, teachers, and religious leaders as well as government officials etc.
An interview schedule is an instrument set of questions that the interviewer asks when interviewing. This is face to face interaction between the researcher and the respondent. Interviews are of three main type’s namely structured, semi-structured and unstructured interview (Bernard et al, 2005). These interview questions were divided into the following sections:
- The structured interview component included questions with expected responses and the interviewer simply recorded the respondent’s response, for example “Yes” or “No”.
- The semi-structured portion of each interviewee used open ended questions asked by the interviewer. The respondent was given complete freedom of response and expression.
- The unstructured part of the interviews involved more probing questions or comments intended to extract more information from the respondent. This format was used to get more information since it was to provide a high degree of objectivity and clarification.
According to Galvin, K. (2008), we use the following questions to determine which partner to choose for interview: 1) Who has the relevant information? 2) Who is most likely to be capable of providing accurate information? 3) Who is most willing to give information? 4) Who is available for an interview?
Galvin, K. (2008). Responses of Pastoralists to Land Fragmentation: Social Capital,
Kathuri, J. N. and Pals, D. A. (1993). Introduction to educational research. Egerton: University
Educational Media Centre.
Kothari, C. R. (2004). Research Methodology; Methods and Technique. New Delhi, India: Dharmesh Printer.
Kothari, C.R. (2008). Research methodology: Method techniques, Second revised, New York.
Jay L.D and Kenneth N.B Modern mathematical statistics with application. Illinois state
Black, Ken. Business Statistics: For Contemporary Decision Making. John Wiley &
Sons, Inc, 2011. Print