Optimal Sample Size: The optimal or suitable size of sample in a survey or poll is the function of four discrete factors:
1. Size of the population: The size of the source population matters a lot. Usually, as the needed sample gets bigger the source population rises, apart from it will raise at a declining rate. Likewise, as the source population gets smaller, special adjustments have to be made.
2. Segmentations desired: Usually, we analyze the source population as an entire, however sometimes one might want to make sure it is representative of the demographic or other distributions in the source population. The more you wish for to segment the outcomes, the larger the sample might require to be.
3. Degree of variance in responses from the population: If the respondents' responses tend to be tightly clustered, then we do not require to sample as many people to acquire the same confidence as we would when the responses range broadly. However until we do some surveying and analyze the data, we won’t know the variance. In such cases, we should set a conservative assumption about the variance.
4. Tolerance for error: The more confident you want to be about the results, the larger the sample.