Basics of Sampling for Your Thesis

Sampling from a defined set of things (population) is done to get an idea (infer) what the entire population selected is like. One cannot look at the entire group of things as it is not feasible so a small portion (sample) is judged mathematically to make an inference about the entire unit. This is what we do in our experimental section of a science thesis. The sampling method is very important based on it we get data and judge/support the hypothesis in our thesis report. Basically a sampling method helps to derive a small sample which is qualitative or quantitative in nature. Sampling methods used are scientific in nature and have a defined procedure. The methods used are accurate enough and use mathematics to remove uncertainty within a limit for a population.

The population of study based on the problem at hand can be specifically defined as a subset of the entire group that has certain characteristics the researcher is looking at. For example, women who have breast cancer; the population entirely consists of woman, and the subset is that having breast cancer. The population can be huge hence we need a frame of reference.

Sampling frame has all the properties we are looking for and defines the sample. For example: women with breast cancer in Boston Hospital; here we have a sampling framework.

Sampling can be done by probability or random unbiased method or biased based on uncertainty. The biased sampling is also known as non-random sampling. The method is determined by what data is being researched? The sample size is determined to reduce errors in the data usually the larger the better is the result as it is more close to the real picture in the group being studied.

Actual collection of data can be by picking from a group, assigning random values and picking from the group or selecting items with some bias (survey sampling). Once the data is collected it is reviewed for being accurate or not based on a set of criteria placed forward in the research. The data is mathematically interpreted to give some statistical results. And this can be done today with statistics. The inference one makes from the results of statistics manipulation gives the researcher an idea. He/she can then use this knowledge to interpret his or her hypothesis. The accuracy will be dependent on the sampling method, size sample and mathematics chosen in your research design.