Is a small sample (about 150) from a population of 250 problematic in a thesis?
Yes, the size of the population is an important consideration in designing a research project. If the population is only 250, then a sample of 150 equals 60% of the population. Why not study the whole population. Generally, the law of large numbers works best with populations of at least a thousand. A 10% random sample from a population of 1,000 (N = 100) is probably representative.
Hi Maria,
Of course, we cannot collect data from the entire population due to time and financial constraints. The choice of your sampling technique will make the difference. when designing a sample, usually, the researcher should consider several decisions and take into account the nature of the research problem and the specific questions that evolve from the question, objectives, time and budget.
There are two types of sampling technique (Probability and non-probability). while i was doing my PhD, I have used non-probability sampling technique where you can collect data from the participants based on who are willing to participate and easily accessible to be recruited and included in the research. this technique limited my findings in terms of generalising the results to the whole population. Again, the number of participates may also depend on the data analysis technique, for example if you are using Structural Equation Modeling as a data analysis technique, then you should have at least 350 participants.
Good luck
Thank you Ali. Just like you, I have focused on convenience sampling but some researchers like Saunders et al. (2009) and Cohen et al. (2007) believe that non-probability samplings cannot be used for inferential statistics. However, your consideration in replying my question is highly appreciated.
Regards
- Badges
- Science topic
- Similar topics
- Methodology
- Research Methodology