Population
Population in research means the whole composition of individual or items that could possibly be included in the study. This is the whole inclusion on which inferences will be drawn based on random sampling (Wysocki, 2008). Subpopulation is a subset of a population where it has different properties therefore said to be heterogeneous. For example, in our study, victims of disasters are heterogeneous since they come from different social, economic, and political backgrounds.
Sampling
A sample is a subgroup of individuals selected from a population. Sampling in simple terms means selecting a portion of the population in the study area which will be a representative of the whole population in the research (Wysocki, 2008). In our case for disaster management, sampling would mean selecting a portion of the victims (population) to use in our research on how their involvement will help the management team.
Sampling strategies are plans the researcher choose in selecting their sampling group to participate in the study. It involves careful thinking and consideration of relevancy between the phenomena under study and the group chosen to be observed (Ardilly & Tillé, 2006). To achieve this, the researcher sets limits on who to be included and who to exclude in the study in our case, it will involve thinking in the lines of who to include based on gender, age, religion, social status and so on since these social measure have different effects during and after the disaster.
A good sampling strategy in quantitative studies showed ensures representativeness. In our study, we should be able to answer the question; does this sample represent the key characters of the population we are studying? This calls for a strategy that is not biased or erroneous. Since errors are likely to occur, as researchers, we should control or minimize them.
Types of sampling
Quantitative research has a complex process of selecting the sample size and ensuring its similarity with the population. This makes it concerned about sampling issues and techniques. In the case of disaster management, it will be difficult to select a sample that has similar characteristics to the whole population of the victims (Ardilly & Tillé, 2006). This is due to the consideration that victims of these disasters are all over the world from different social, economic and political set ups.
There are two major sampling methods appropriate for this kind of study; random and non-random sampling. These two methods are associated with fewer errors because they use statistical procedures to estimate the sample group therefore less margin of errors obtained from the sample (Ardilly & Tillé, 2006).
Random sampling: this is a sampling technique based on the theory of probability and normally produces samples that are representative of the whole population. This gives each element in the study population an equal probability of being selected. There are three methods of random sampling techniques, these are, simple random, stratified sampling, and cluster sampling techniques (Gregoire & Valentine, 2008).
Stratified technique involves preselecting which portion of the population to be studied. In our case, since disasters are all over, we may choose to study a portion of say hurricane Katrina in the USA. Then using the table of random sampling or interval, we select our sample from these strata of population (Gregoire & Valentine, 2008). Caution is to be taken when using this strategy since inferences made back to the population need to be weighted in order to correlate back to the actual proportions.
Cluster sampling involves the selection of groups rather than individuals. Both one stage and two stage cluster sampling can be used. One-stage involves the use of either simple random or stratified methods while two-stage takes the selection further to select a sample from the first group. Probability proportional to size is used to further refine the process (Gregoire & Valentine, 2008). In our study this would mean selecting samples from different disaster situation, say from earthquake, hurricane, or terrorist victims.
Non-random sampling: it is the most commonly used method of sampling. However, it is not credited to offer representative sample. Convenience sampling is the most employed. It involves having two groups, an experimental and control group. The study is carried out then comparison done between the two groups. In our case it would involve selecting an experimental group from the victims and a control group from the non-victims (Gregoire & Valentine, 2008). We would then compare the effects of the disaster by analyzing each group’s social, economic, and political status.
How the sample will be drawn
Lottery method of sampling: each member of the victim population is given a unique number, then they are thoroughly mixed and without looking, the researcher picks a number which makes up the sample group. Mostly this process is carried out using a computer because it can be cumbersome if done by hand (Dawson & Dawson, 2009).
Sampling with replacement: is a method of random sampling in which members or items of the population can be chosen more than once for inclusion in the sample. In this method, the person picked in the sample has the same chance of being picked again.
Sampling without replacement: it’s different from the above method in that once an item is picked, it cannot be picked again. This makes the sample have unique items.
The most appropriate method in our study is the lottery or sampling without replacement. These two methods will ensure that all the items in our population have an equal probability of being selected.
Sample size
The size of the sample in research is a very important criterion as it reflects the accuracy, reliability and validity of the research findings. The larger the sample sizes the better. A better level accuracy in results and drawing implications from the results can be achieved by use of large samples (Dawson & Dawson, 2009). However, the inclusion of the whole research population is the ideal way of ensuring zero errors in sampling though it is not feasible given the financial and time limits. This has therefore influenced the development of sampling methods in the research field (Dawson & Dawson, 2009).
If the study population is small we can take it as a whole but with large populations like in our case, we can only increase validity, accuracy, and reliability by taking large samples.
References
Ardilly, P., & Tillé, Y. (2006). Sampling Methods. New York: Springer Science+Business Media, Inc.
Babbie, E. R. (2013). The practice of social research. Belmont, Calif: Wadsworth Cengage Learning.
Dawson, C., & Dawson, C. (2009). Introduction to research methods: A practical guide for anyone undertaking a research project. Oxford: How To Books.
Gregoire, T. G., & Valentine, H. T. (2008). Sampling strategies for natural resources and the environment. Boca Raton: Chapman & Hall/CRC.
MAT6315F08 - Chapter Three - Sampling Methods. (n.d.). Retrieved from http://mat6315f08.wikispaces.com/Chapter+Three+-+Sampling+Methods
Random Sample. (n.d.). Retrieved from http://sociology.about.com/od/Types-of-Samples/a/Random-Sample.htm
Wysocki, D. K. (2008). Readings in social research methods. Belmont, CA: Thomson/Wadsworth.
