Statistical studies involving surveys, experiments and observation require an organizational plan or design prior to research implementation. One aspect of research design determines sample size. Sample size defines the number of people or animals participating in the study, or how many repetitions of the experiment should take place in order to obtain substantial findings to help answer research questions. Problems arise when research designers choose a sample size that is too large or too small. Small sample sizes generally use less resources than large sample sizes but present several potential issues influencing the quality of the research findings.
Other People Are Reading
A small sample size may result in the lack of statistical representation of a phenomenon. For example, a researcher performs a study of transportation behaviours to find out what percentage of college students in a particular city take the bus to school. If her study includes a small sample size of 10 people and those 10 people all walk, bicycle or drive to school, then her sample size has failed to represent those who take the bus. Lack of representation as a result of small sample size becomes particularly evident when attempting to study phenomena that occur infrequently. On the other hand, according to Russell V. Lenth of the University of Iowa, a very large sample size may include phenomena of little importance. For example, if one person cartwheels to school in a sample size of 1,000, cartwheeling will show up in the resultant data as 0.1% of students cartwheel to school even though cartwheeling does not really represent a significant method of school transportation.
When a study involves a small sample size that inhibits its ability to result in findings of substance, the resources used in that study were wasted. Studies performed without much time, supplies and energy do not present as much of a waste problem as those requiring a long duration, such as agricultural studies involving one or more growing seasons. On the other hand, a study with an excessive sample size may result in findings, but also wastes resources if the same findings could have been found with a smaller sample size.
Studies with small sample sizes may pose unnecessary risks to human and animal subjects. If the study requires subjects to participate in potentially harmful activities, but uses a sample size too small to result in findings, all of the research subjects engage in risk with no benefits to the furtherance of knowledge. For example, a study of an experimental pharmaceutical with harmful side effects that uses a small sample size may not result in conclusive data. Therefore, all subjects within that small sample size experienced harmful side effects with no benefit of conclusive data and gained knowledge. On the other hand, studies may also pose unnecessary risk to subjects because of excessively large sample sizes. The same study with an experimental pharmaceutical using too large of a sample size does reach conclusive data, but exposes more subjects than necessary to the pharmaceutical's harmful side effects.
- 20 of the funniest online reviews ever
- 14 Biggest lies people tell in online dating sites
- Hilarious things Google thinks you're trying to search for