• Sample size:

It depends from population to population. Generally, a large sample size gives more accurate results. If the sample size is way too small, the test will have less power statistically.

For example,

Sample size = 50 is suitable for a population of size 110.

But sample size = 50 for a population of size 70 is more than the actual requirement.

Approach for small sample size:

In the case of small sample sizes, the non-parametric test approaches are more suitable for the analysis.

Where

Non-Parametric tests

Are the tests that are free of distributions. They don't have any rigid assumptions. They are used when;

• The sample is too small.
• There are outliers or extreme values present in the data.

Some the non-parametric tests are as follows;

• Friedman test
• Mann-Whitney test
• Kruskal-Wallis test
• Spearman Rank correlation test

Therefore if the sample size is less than 100, it is suggestable to continue the study with non-parametric tests.