Sampling Techniques for Dissertation Research
When the population representing the phenomenon we want to study is too large, we use the method of sampling or studying a smaller representative group to conduct our dissertation research.
The method we use depends on the subject and the characteristics of the population. There are two main types of sampling for research:
1. Probability sampling: In this method the technique of picking out items for the population to form the sample is such that each population item has an equal chance of being selected. For instance, if one wants to study the construction material used for all the houses in Stratford, one could note down each house’s number on a separate piece of paper, put the chits of paper in a bowl, shake it up and draw as many chits as needed. Also called random sampling.
2. Purposive sampling: Sometimes random sampling overlooks the significant variations in the population, despite using the various measures to control for it. In such a case we use purposive sampling. For instance, if one is writing a dissertation on environmental protection groups then it would be more feasible to deliberately study the more active group rather than picking them out at random. Also called non-probability sampling.
Within these two broad categories of dissertation research sampling, there are many other techniques such as:
– Stratified sampling: The population set is first subdivided based on a certain characteristic and then an equal number of sample items are selected from each stratum. For instance, if one is writing a sociology dissertation on the expenditure patterns of undergraduate students, subdividing them on the basis of ethnicity or major or year might provide better insights.
-Proportional sampling: When the population consist of two or more strata, one can select the sample items for each stratum in proportion to the population size of the stratum. For instance, in the above illustration, if there are twice as many Asian students Latin American ones then this ratio between them should be maintained in sample selection too.
– Systematic sampling: Making a list of population items and selecting every nth item on the list from the population.
– Multi-stage sampling: This dissertation research method of sampling is used when the population is very large. For instance if you are writing a law dissertation on the incidence of divorce cases in a particular state then you break down the state into its constituent district, towns and households such that you select an equal number of households from each town and from each district.
– Multi-phase sampling: In this method of sampling one collects some additional information from certain members of the sample. For instance a finance dissertation on CDOs would benefit from the insight gained from addressing additional questions to former Lehman Brother investment bankers.
You can take a look at our sample dissertation on call centre outsourcing to get an idea of how to arrive at the appropriate sampling technique and the method for administering the same.