QUESTION 6 (30 Marks) a) Research that investigates an accessible population, whose population parameters are homogenous requires a smaller sample than research into heterogeneous population parameters. An example of a homogenous population parameters would be a group of isiZulu speaking male employees of the same age, same educational background and level, and residing in the same geographical area. Reformulate this description to represent a heterogeneous population. (5 marks) b) Assuming that you want to cross-analyse two variables (e.g. men and women), a larger sample is needed. What this means is that, the size of the sample to be drawn largely depends on both the homogeneity of the accessible population, as well as the objective of the research. Doing a simple random sampling implies that each unit of analysis in the accessible population has got the same chance of being included in the sample. Assuming that the accessible population comprises 8000 employees, each employee has an equal chance of being selected. The extent to which the sample drawn will correspond with the accessible population can be calculated in advance by calculating s (the standard error). i) Based on the data given herein, what would be the standard error, if you are to draw a simple random sample of 100 employees from an accessible population of 8000, of which 4800 are males and 3200 are females? (5 marks) ii) If you double the sample size to 200, what would be the standard error? (2 marks) c) Validity addresses the issue of whether the researcher is actually measuring what he/ she has set out to do. There are four specific types of validity each of which the researcher would ideally want to establish for the research instrument prior to administering it for the actual study. In light of the above, indicate the type of validity that describes the statements given below: (8 marks for i- iv) i) This type includes statistical tests and logical argument ii) This type involves a group of people and content related evidence iii) A potential challenge with this type is that one may fail to formulate the criteria to be used before actually checking the scores iv) One challenge with this type is that people may be biased in terms of their judgement d) The validity and reliability of the interviews and surveys can easily be distorted if the measuring instrument is inappropriate. In view of this, highlight the major differences between interviews and selfadministered questionnaires (5 marks). In doing so, indicate some of the challenges related to the implementation of these data collection tools (5 marks). (10marks) END OF PAPER