Statistical analysis plays a vital role in thesis or dissertation presentations. In any report, numbers speak more than words. A pictorial representation of number rather than explaining in words attracts the eye of the panelists and helps in gaining better grades. While doing this statistical analysis, the most important point is to test the sample or collected data. Sample size is more important for sample testing because it helps in identifying the actual significant differences. A larger sample is required as it is more reliable than a smaller sample. The final result will be affected based on the size of the sample we have collected. A good sample size can help in detecting a statistically significant result whereas smaller samples have lesser probability of detecting a statistically significant result which means waste of time and resources. A smaller sample might lead to Type II error. Many null studies cannot detect the desired difference as the sample size would be significantly low.
The Calculation of Sample Size
The sample size must be calculated before collecting the data and failing to do so will affect the value of the study. Some of the common errors include unrealistic assumptions, not performing any calculations, failing to investigate etc. The calculation must be done during the planning stage of the study and should be based on the type of research and design. A professional statistician can help on these aspects and it is very important to take their guidance rather than picking random assumptions. There are many approaches to determine the size before conducting the survey. Different sampling techniques include Purposive Sampling, Quota sampling, snowball sampling etc. each sampling method should be chosen based on the kind of study and expected results. Often there is a temptation to go for lower sample size or to make short cuts which leads to failure of the study.