4.5 Sampling
4.5.4 Sampling Techniques
Theampling techniques in research can be classified as probability and non-probability sampling techniques (Creswell and Poth, 2017). The various sampling techniques are briefly explained in the following section for the benefit of the reader and future research students.
Probability sampling technique
The probability sampling technique entails a sampling method in which the subjects of the population get an equal chance to be selected as a representative sample. Examples of probability sampling methods are given below:
Simple Random Sampling
Some scholars assert that a simple random sample is a subset of a statistical population in which each member of the subset has an equal probability of being chosen (Brannen, 2017; Moser and Kalton, 2017; Montgomery, 2017, Schilling and Neubauer, 2017).
Stratified Sampling
Stratified sampling is a sampling technique that facilitates the researcher to stratify the population into distinct groups (Vermeulen, 2017). Izquierdo, Rodrigues and Fueyo (2008) concur with Lundström and Wikberg (2017), who postulate that stratified random sampling is put into use to highlight a definite subgroup within the research population. This technique is valuable in such studies because it guarantees the presence of the key sub-group within the sample. Researchers similarly make use of stratified random sampling when they want to observe existing relationships between two or more sub-groups.
Cluster Sampling
Cutting, Karger, Pedersen, and Tukey (2017) describe cluster sampling as a type of sampling method which allows the researcher to divide the population into separate groups, called clusters.
A simple random sample of groups is then designated from the population (Hayes and Moulton, 2017; Scott, 2017 and Foss, 2017).
Systematic Sampling
Systematic sampling refers to a type of probability sampling method wherein sample members
from a bigger group of the population are selected indiscriminately starting from a fixed periodic interval (Senaratna, et al., 2017). The sampling interval is considered by dividing the population size by the desired sample size (Foss, 2017; Gundersen, Jensen, Kieu and Nielsen, 1999; Fortin, Stewart, Poitras, Almirall and Maddocks, 2012).
Multi-stage Sampling
Multistage sampling combines some of the sampling procedures discussed above in stages (Creswell et al., 2017). Multistage sampling is defined as the probability sampling technique wherein the sampling is carried out in several stages such that the sample size gets reduced at each stage (Moser and Kalton, 2017). Pavan, Schreier and Temes (2017) assert that multistage sampling presents a multifaceted form of group sampling.
Non-probability
Non-probability sampling is a sampling technique where the samples are gathered in a process that does not give all the individuals in the population equal chances of being selected (Yeager, et al., 2011). In this sampling method, it is not known for sure which individual from the population will be selected as a sample (Sadler, Lee, Lim and Fullerton, 2010). There are five categories of non- probability sampling technique that a researcher can use: quota sampling, self-selection sampling, snowball sampling, convenience sampling and purposive sampling (Moser and Kalton, 2017).
Quota sampling
Quota sampling is an approach for collecting representative data from a cluster. Compared to random sampling, the requirement for quota sampling involves representative individuals be selected out of an explicit subcategory (Moser et al., 2017). As an example, a researcher might require a sample of hundred (100) women, or hundred (100) individuals between the age ranging from twenty (20) to thirty (30) (Hungin, Whorwell, Tack and Mearin, 2003).
Self-selection sampling
Moser et al. (2017) highlight that a sample is self-selected when the inclusion or exclusion of sampling units is determined by whether the units themselves agree or decline to participate in the sample, either explicitly or implicitly. Self-selection sampling is a type of convenience sample involving study participants or subjects volunteering to take part, usually in response to request of
public importance (Moser Modesto and Tichapondwa et al., 2017) Hungin, Whorwell, Tack and Mearin, (2003) postulate that self-selection bias rises in any condition in which individuals form themselves into a group, causing a biased sample with non-probability sampling.
Snowball sampling
Wig, Laumann, Cohen, Power, Nelson, Glasser and Petersen (2013) regard snowball sampling as a situation where research subjects recruit other participants for a study. It is useful where potential research participants are hard to find. It is referred to as snowball sampling because, in theory, once the ball starts rolling, it picks up more snow on the way and becomes bigger and bigger (Sheu, Wei, Chen, Yu and Tang, 2009).
Purposive or judgemental sampling
Scalzo, (2017) concurs with Brooks, and Manias (2017) and Saunders et al. (2009) who argue that purposive sampling facilitates the researcher to use their judgement to make a choice that enables them to appropriately answer the research questions and address the study objectives. Saunders et al., (2009) further note that purposive sampling is also referred to as judgemental sampling. This technique is often helpful in cases regarded highly explanatory concerning the research problem (Brooks and Manias, 2017). Saunders et al., (2009) identify the technique’s applicability for case study research. On the other hand, scholars indicate the qualities of the technique as cost- effectiveness (Malhotra, 2010) and as a fast and appropriate way of collecting primary data (Mayer, 2017). For this study, convenience sampling was used to sample women-owned or managed SMES because of the strong link that it had to the research problem as discussed in the following section.
Convenience sampling
Convenience sampling is regarded as one of the key types of non-probability sampling methods (Ali, et al., 2017; Crookes and Davis, 1998; Saunders et al., 2009). It is comprised of research participants who are easily reachable (Skowronek and Duerr, 2009). For example, a researcher can consider collecting data from students from his or her own class (Hedt and Pagano, 2011).
Convenience sampling is also known as haphazard sampling (Welman et al., 2005; Saunders et al., 2009). This sampling technique was employed in sampling women-owned or managed SMESs in Gweru, to adequately suit the key focus of the research which is women-owned SMESs. Elements of the population are easily selected by the researcher since they are easily accessible, and the
process enhances the attainment of rich information that broadens the study findings (Case, Burwick, Volpp and Patel, 2015; Bryman and Bell, 2007). Literature indicates that convenience sampling costs much less and consumes less time than other sampling methods because of its easy- accessibility and measurability of elements (Malhotra, 2010). Nonetheless, convenience sampling has its own limitations such as being prone to bias (Welman et al., 2005; Saunders et al., 2009).