Sampling and sample size
Student’s Name
Institutional Affiliation
Date
Sampling and sample size
Some of the factors to consider when determining an appropriate sample size include the type of research (qualitative and quantitative), the sampling technique, the amount of time and money you have for research, the population parameters we want to estimate, how precise you want the final estimates to be, and the variability (spread) of the population.
To know whether a sample is large enough, a general rule thumb for the large enough sample condition is used. The condition is that n>30 (where n is the sample size). However, this rule depends on what you want to accomplish and what you know about the distribution. The large enough sample condition applies when you have a symmetric or unimodal distribution, and you have moderately skewed distribution (Suresh & Chandrashekara, 2012). The Central Limit Theorem justifies the use of 30 as the minimum sample size in a normal distribution. Empirically, the sample size is said to be enough if it is greater than 30. Therefore, we can conclude that the sample size is not large enough if it is less than 30. This is only applied if the population is normally distributed.
Yes, we can have large sample sizes. Very large sample sizes can lead to bias magnification and amplify confirmation bias. Large sample sizes are not beneficial in any way. First, large samples are close approximations of the population. Since inferential statistics generalize from a sample to the population, it is less of inference if the sample size is too large (Kotrlik et al. 2001).
Furthermore, collecting sample sizes is usually costly, so it would be unwise to collect a larger sample than is needed for your purpose. According to Suresh & Chandrashekara (2012), large samples are a waste of time. As long as you have achieved the confidence level you want, there is no need to have a large sample size.
To determine the appropriate sample size for quantitative research, start by choosing an appropriate significance level or the alpha value. The most commonly used significance level is p=0.05. The next step is to select the power level. The most common power level is 0.8, or 80%. The next step is to estimate the effect size. Typically, an effect size of 0.5 is acceptable for quantitative research (Devane et al. 2004). Lastly, is to organize the existing data.
References
Devane, D., Begley, C. M., & Clarke, M. (2004). How many do I need? Basic principles of sample size estimation. Journal of Advanced Nursing, 47(3), 297-302.
Kotrlik, J. W. K. J. W., & Higgins, C. C. H. C. C. (2001). Organizational research: Determining appropriate sample size in survey research appropriate sample size in survey research. Information technology, learning, and performance journal, 19(1), 43.
Suresh, K. P., & Chandrashekara, S. (2012). Sample size estimation and power analysis for clinical research studies. Journal of human reproductive sciences, 5(1), 7.