SAMPLING METHODS
This refers to a statistical procedure which is concerned with selection of individual observation in a study. It enables the researcher to make statistical inferences in the study population. A sample refers to part of the population that represents the characteristics of the entire population. There are various types of sampling methods which are used in studies. The most common methods of sampling include random sampling simple random sampling, systematic random sampling and stratified random sampling
Simple random sampling technique
Simple random sampling technique is a pure straight forward sampling strategy which is used to obtain samples in a study. It is one of the most commonly used methods in statistics as it offers equal chance to the individuals in a population for observation. Simple random sampling can be done by assigning all the individuals in a population numbers and then generating random numbers to select the individuals to be used as a sample. The advantage of this method is that it provides no room for biasness as compared to other sampling methods. It is easy to generalize the results from a study conducted by random sampling due to the distribution. It is necessary to assume that the population that you are working with is homogeneous and assign them numbers that can be used for randomization. There are many programs which are currently used in generating random numbers. The random numbers will assign positions to the selected samples without bias making them more representative of the population in question.
Disadvantages of simple random sampling technique may include the costs involved while listing or numbering the potential respondents in a population (Heckathorn., 1997). This can be really time consuming especially in populations where very large numbers are involved. Although large samples are preferred, having such may be a challenge when large numbers are involved. The other disadvantage is that not all populations which are being studied are homogeneous. Taking an assumption that the population is homogeneous may ignore very important differences that could have serious impacts on the study. For example, among human population, there are adults and children, male and female or even different races or people living in different environments. The differences would contribute to variation. Additionally, it is not good where face to face interactions are desired in studies that cover large geographic areas.
Systematic random sampling
Systematic random sampling is slightly different from simple random sampling. In the systematic sampling, every Nth individual is selected to be included in the sample (Kothari, 2004). The method requires that the individuals in a population are assigned positions and it is decided a priori the positions to be selected and these forms the samples in the population. It is done by determining the sampling fraction which is obtained by dividing the actual sample size by the total population. The sampling fraction becomes the guide for conducting systematic sampling. In this method, the first sample has to be chosen in a random manner while the succeeding samples are chosen following the order which is decided like say the 3rd, 4th or even 5th item (Cooper et al., 2006). The advantages of this method is that if it is done correctly, it will give results which is an approximation of the simple random sampling. It is both an efficient and a cost-effective method of sampling. It is suitable for collecting information from geographically diverse population where face to face contact is not necessary. The main disadvantage is that it can only be applied when a list of the entire population is available. Populations with periodic patterns may not be suitable for this approach as it will lead to biasness.
Stratified random sampling
Stratified random sampling is one of the probability sampling methods which involves dividing the population into groups also referred to as strata (Johnson and Onwuegbuzie, 2004). The groups can either be proportionate or disproportionate. In order to use this method, the researcher or the researchers have to identify relevant stratums and ensure their actual representation in the sample. Each of the subjects in the stratum have to be identified with unique numbers. It is important to select each of the unique numbers in each of the strata and ensure that they form part of the sample.
Purposive sampling technique
The purposive sampling technique utilized in the research was snowballing sampling technique. The technique (snowball) is where the participants of the study recruits others for a study or test (p. 62). Snowball sampling, which is a non-probability sampling approach, is used where finding potential respondents is difficult. With no randomized sampling involved, the research uses the researcher’s judgment to select participants. In snowball sampling, the following two steps are critical, first, identifies potential subjects from a population and in majority of cases, only one or two are initially found. Two, asks the identified potential subjects to recruit others. However, participants are made aware they do not have to provide any other names. Two critical advantages sets apart snowball sampling making it appropriate to use. One, it makes possible for research to be conducted where it would be otherwise impossible to take place due to lack of respondents. Two, it assists in helping the researcher discover characteristics of a population the researcher hardly knew it existed.
