An essential component of surveys is the population being profiled and determining the correct proportion of individuals required to represent this target population – sample group. In other words, how much data is required to obtain reliable insights and therefore make accurate decisions upon the survey results.
Calculating the right sample size is crucial to gaining accurate information. If the sample size is too low, it is likely that the data would be less reliable. However, a larger sample size means more time and expense in conducting the survey and furthermore, is not always statistically significant.
Calculating survey sample size
There are basic formulas for sample size calculation, however a guiding principle is to have a statistician determine the sample size given the variables involved:
1. Population Size: How many total people fit the demographic? This could be a specific group: users of a product or employees of an organisation. Not knowing the exact population is a common factor in market research and is not an adverse factor. ‘The mathematics of probability prove that the size of the population is irrelevant unless the size of the sample exceeds a few percent of the total population you are examining’.
2. Confidence Interval: What the research study identified as the acceptable percentage the sample group will differ to the population it represents? The positive or negative deviation is the margin of error.
3. Confidence Level: How often the true percentage of the population who would pick an answer lies within the confidence level, the percentage of the population that lies within the limits of the margin of error.
The most common confidence intervals are 90% confident, 95% confident, and 99% confident. The 95% confidence level means that 19 times out of twenty, results would fall in this – + interval confidence interval.
4. Standard of Deviation: How much variance is expected from responses? .5 is standard and ensures the survey sample with be large enough.
The assistance of a statistician guides the research team through the process in addressing these variables, through applying statistical techniques and formulas bespoke to the research. However, if there is a limited budget or the research is straightforward, statistical calculators available online can prove a useful alternative But it is necessary to check the assumptions that are used in these calculators to confirm they address the variables required for your organisation’s research purposes.
Once the sample size has been determined it is essential that the response rate is considered. Response rate is the ratio of respondents that complete the survey to the end, compared to the total number of surveys issued. Survey response rates vary widely dependent on population, survey design, survey purpose, method used to distribute the survey, etc. Thus, the percentage of those that complete the survey can typically range from 10%-40%.
Given that’s the case, the sample calculation is clarified as the number of surveys that need to be completed to be a statistically accurate study. Hence, the survey must be issued to a greater number of individuals to ensure that participation meets the calculation.
As this post illustrates, there are several factors involved in determining the sample group. It can be a complicated process depending on the complexity of the research project. If your organisation would prefer to focus more on the application of the survey results and leave the survey administration to experienced experts, then do not hesitate to contact Smart Consult; we would be happy to assist.
Should you have a research project or request for proposal that you wish to discuss, then please submit the request here.