Week 5: Questionnaire Designs (surveys) & Sampling Plan - Cluster sampling o The target population is first divided clusters where there are mutually exclusive (similar characteristics) o Then a random sample of cluster is selected. For each selected cluster, either all the elements are included in the sample (one-stage) or a sample of elements is drawn probabilistically (two-stage) o Elements are heterogeneous within a cluster but clusters themselves should be homogeneous with each other opposite to stratified sampling o Example Cluster sampling (Put 20 students into 4 groups but each group would have something in similar, e.g. separating each group by the suburb they live in, so each group is mutually exclusive. Then when selecting students, there are to ways to do it. One way is one-stage, this is where we only choose one group (e.g. sunshine). The other way is two-stage, is when you choose two groups and from each group you would choose 2 students (e.g. sunshine and Caulfield, then randomly choose two students from each of those two groups). (Each group is called a cluster and each student is called an element) The difference between that stratified and cluster is that stratified shows a better representation of the population as all groups as included. When to use non-probability sampling techniques? - Researcher is constrained in time or resources - For exploratory research (want to know more information about the problem) - Target population is small or difficult to identify - Sampling error is a major concern Types of non-probability sampling techniques - Convenience sampling o Attempts to obtain a sample of convenient elements. Often, respondents are selected because they happen to be in the right place at the right time o E.g. Coles workers would ask people who walk into the store do a survey) - Judgemental sampling o Based on the researchers judgement, could be a sample of experts o E.g. Coles workers are instructed to only give out survey to people who look like students.) - Quota sampling o A two-stage restricted judgemental sampling o E.g. have different age groups, such as 20-29, 30-39, etc. Then with the combination of judgemental sampling and convenience, you would ask people who are in those certain age groups to do the survey - Snowball o Initial group of respondents is selected and then ask them to get more respondents. Like word of mouth o E.g. At the beginning you would find a few people such as getting your friend to do it (could be either convenience or judgemental sampling). And lets say that researcher want to have uni students as the sample, so the researcher would get one of his uni friends to complete a
You've reached the end of your free preview.
Want to read all 6 pages?
- Two '16
- researcher, Questionnaire Designs