Description of stratified random sampling
For the patient satisfaction survey involving a population of 600 patients, the most suitable sampling method is Stratified Random Sampling. It involves the separation of the target population into subgroups, based on specific differences identified by the researcher to increase the “randomness” of the samples. Consequently, there will be several subgroups from which the random samples can be picked. The method has seven stages, namely, defining the population, selecting the stratification, listing the population and listing it according to the stratification, working out the proportionate stratification and finally using the random sample (Fink, 2002).
Application of Stratified Random Sampling
The target group consists of 600 patients of different age groups, racial backgrounds and suffering from various conditions among other heterogeneous factors. In this case study, age will be used to categorize patients into desired subgroups. When patients are grouped, there will be five age clusters, starting from 65 and above, between 64 and 41 between 40 and 25, between 24, 16, and 15 and below. Using their ages to stratify patients, the researcher can minimize bias arising from the different age related interpretations of satisfaction. The method will ensure they get a balanced picture of the satisfaction levels, depending on how each age group views the services they are receiving (Fink, 2002). Using a different method such as non-stratified random sampling may result in having too many people from one particular group sampled. As a result, stratified random sampling can be justified in that it provides a highly representative sample, owing to the probabilistic method used.
Labor-intensive and expensive
There are two main challenges that one will have to contend with while using this stratified random sampling method. The first is that, it is very labor intensive and requires several people to work at gathering all the requisite stratification data. In addition, it is also very time consuming, which makes it very expensive. A proposed solution to counter this challenge could be for some of the patients to act as volunteers in administering the study to their fellows.
Divergent definitions of satisfaction
A second challenge the researcher will encounter is that the varied age groups in the study may have very different definitions of what they consider satisfactory. Given the age difference between the elderly and the very young, it may be difficult to get questions that the two extremes can relate to on equal terms. For example, teens may want more attention from the nurses and other caregivers, while the elderly patients may require much less. As a result, the study will need to be designed in such a way that the questionnaire questions are understood in the same way by all patients irrespective of their age. Admittedly, getting perfection is difficult, if not impossible, however, the situation can be improved by ensuring the administration of the survey involves open, instead of closed questions. In addition, professionals, who are familiar with their psychology and specific clinical needs, should interview young children. Accordingly, they can be guided in understating and interpreting the questions to ensure the survey is as objective and accommodating as possible. In the end, while this sampling method does not guarantee perfection, it provides a relatively accurate way of preparing a balanced sample for the study.
References
Fink, A. (2002). How to sample in surveys. London: Sage Publications.