Determining your sample size

Once you have decided to conduct a market research study, the next step is to determine the number of interviews to be completed. There are two important issues to take into consideration when determining this number: the desired reliability of the results and the budget.

Recommended sample size

The sample size which is normally recommended for most research studies is based on the industry standard 95% confidence level, with an accuracy of ± 5%.

The confidence level refers to the likelihood that the final results will not deviate by more than a certain percentage from the actual population statistics (i.e., the results obtained if everyone in the population were surveyed).

Using the 95% confidence table

The level of reliability of the final data can be determined. Keeping in mind the desired goal of ± 5%, when conducting a study on a population size of 100,000 it would be necessary to conduct 400 surveys. In this case, the reliability of the results would be ± 4.9% at the 95% confidence level. This means that in 95 out of 100 repetitions of the survey, the results will not vary more than ± 4.9%.

Below is the chart used to calculate the reliability of the information gathered at the 95% confidence level.

With the ideal sample size determined, a decision must be made as to the financial feasibility of surveying the given number of people. When estimating the costs for a market research study, it is important to keep in mind that data collection (surveying) normally makes up the highest percentage the study’s budget. With each study, it is necessary to strike a balance between cost and reliability.