II. ESTUDIO DE LA VIDA Y OBRA DE JOAN COSTA
2.4. El dibujo como recurso para sus primeros trabajos
The purpose of sampling is to select a representative set of individuals from the target population in the research study (Gravetter & Forzuno, 2003). To do so requires that the target population be operationalised as a sampling population. A sampling population consists of those final sampling units in the target population that have a positive, non-zero probability of being selected in the sample (Babbie & Mouton, 2001).
A representative sample was required for this study as it is not possible to obtain measurements from each subject in the target population. Sampling refers to taking a portion of a population as representative of that population (Kerlinger & Lee, 2000). The extent to which the observations can be generalised to the target population and the power of the inferential statistics tests depends on the number of subjects in the chosen sample, and the representation of the sample. A chosen sample will only be considered representative in the extent to which it provides an accurate statistical portrayal of the characteristics of the sampling population (Theron, 2012). The ideal is for the sampling and target populations to coincide. As this is seldom the case, the objective therefore should be to try and minimise the gap between the target and sampling populations.
3.5.1 Choice of sampling method
Methods of sampling can be categorised as either probability sampling or non-probability sampling (Kerlinger & Lee, 2000). A brief discussion of the advantages and disadvantages of the two categories follow below to explain the choice and evaluation of the sampling method used in this study.
3.5.1.1 Probability sampling methods
The ultimate purpose of sampling can be seen as to select a set of final sampling units (FSU) from a population in such a way that descriptions of the statistical characteristics of specific attributes of those sampling units (in terms of statistics) accurately portray the parameters of the total population from which the FSUs are selected (Babbie & Mouton, 2001). Probability sampling enhances the likelihood of accomplishing this aim and also provides methods for establishing the degree of possible success. In probability sampling, the entire (sampling) population is known, each individual in the population has a specific non-zero probability of selection, and sampling is done by a random process based on the probabilities (Gravetter & Forzano, 2003).
According to Kerlinger and Lee (2000), random sampling can be utilised as a method to draw a sample from a population so that all possible samples of fixed size n have the same
probability of being selected. In stratified sampling, the population is divided into strata, such as men and women. Multi-stage cluster sampling is the most used method in surveys, and involves successive random sampling of units, or sets and subsets (Kerlinger & Lee, 2000). During systematic sampling, the first sample element is randomly chosen in the first interval of length k and, following on that, every kth FSU is selected from every interval. For example, if the element randomly selected from the elements 1 through 10 is 6, then the subsequent elements are 16, 26, 36 and so on (Babbie & Mouton, 2001).
3.5.1.2 Non-probability sampling methods
According to Gravetter and Forzano (2003), in non-probability sampling procedures, the population is not completely known; individual probabilities cannot be known; and the sampling method is based on factors such as common sense or ease, with an effort to maintain representativeness and avoid bias. In quota sampling, knowledge of strata of the population (e.g. sex, race, religion) is used to select sample members that are considered to be representative, ‘typical’ and suitable for certain research purposes (Babbie & Mouton, 2001). Purposive sampling is characterised by the use of judgment and a deliberate effort to obtain representative samples by including presumably typical areas or groups in the sample (Babbie & Mouton, 2001). Kerlinger (as cited in Van Heerden, 2012) describes accidental sampling as the weakest form of sampling but also states that it is probably the most frequently used. In effect, during accidental sampling the researcher takes available samples at hand.
3.5.2 Data collection procedure
Non-probability convenience/availability sampling (i.e., a non-probability sampling technique) was employed in this study (Babbie & Mouton, 2001). This technique implies that individuals who presented their availability for the study were selected. Various organisations were approached by email to request institutional permission to conduct the research study in the organisation. Due to the non-probability sampling procedure that was used to select the sample it cannot be claimed that the sample is representative of the target population. Although, admittedly, it would have been preferable to conduct the study on all the employees at a particular organisation, the reality is that institutional permission can restrict one to a certain department only. The identities of the organisations who participated in the study are not disclosed in this study to ensure the confidentiality of the information which might affect their company image.
Based on the above-mentioned, the proposed structural model and the proposed procedure for operationalising the latent variables, the target population for this study comprised all
first-line/non-managerial employees in South African organisations. The follower is consequently the only unit of analysis in this study (i.e. the follower can be seen as both the research subject and the research participant).
To ensure the validity of the study, it was decided to include organisations with more than 30 employees in the research, as well as an overall sample of at least 200 employees. The research hypotheses developed in Chapter 2 and listed in Chapter 3 were empirically tested using a sample of 224 respondents. The sample consisted of employees operating in two organisations in South Africa in order to test the hypotheses that a follower’s perception of his/her leader’s ethical leadership ability determines the follower’s perceived organisational justice and his/her perceived ethical climate.
Institutional permission was obtained from the two organisations involved in this study. The two organisations were primarily based in the Western Cape, although the one organisation made use of their Gauteng branch to complete the questionnaires. A questionnaire designed to gather data was distributed through the internet and was sent to the identified participants. Data were also collected by means of paper-and-pencil tests which were distributed to employees who did not have internet access. Participants were required to accept the conditions specified in the instructions for the questionnaire. Participants were assured that confidentially would be maintained by treating their responses as anonymous and that no names would be revealed in the study (See Appendix B for informed consent form). Participants were also guaranteed that the study envisaged no potential risks or discomfort and those responses would not be revealed to managers, but would be stored directly on the Stellenbosch University database.
3.5.3 Demographic profile of the sample
The overall sample consisted of 159 males (71%) and 65 females (29%). The sample presented an average age of 29.58 years, with the majority (68.8%) of respondents aged between 20 and 30. The race distribution of the sample was as follows: African (75%), Coloured (6.7%), Indian (2.7%), White (14.7%) and Other (0.9%). The sample was also compiled from respondents from different companies and industries. The majority of respondents came from non-managerial (64.3%) and lower-level management (21.4%) and mainly from the retail industry (97%). The health and welfare services (2.68) and financial industries (.4%) were also represented in the sample, but in smaller numbers. These descriptive statistics are presented in Table 3.2.
Table 3.2
Demographic variables
DEMOGRAPHIC VARIABLES N % IN SAMPLE
Gender Male 159 71 Female 65 29 Age Below 20 0 0 21 – 30 154 68.8 31 – 40 51 22.7 41 – 50 13 5.8 Above 50 6 2.7 Race distribution African 168 75 Coloured 15 6.7 Indian 6 2.7 White 33 14.7 Other 2 0.9 Employment Full-time 138 61.6 Temporary 86 38.4 Job level Non-managerial 144 64.3
Lower level management (First line manager) 48 21.4
Middle level management 28 12.5
Upper level management (Senior manager) 4 1.8
Industry
Manufacturing 0 0
Retail 217 96.9
Financial Services 1 0.4
Construction 0 0
Health and Welfare Services 6 2.7
Other 0 0