Sampling is an integral part of the data collection process. Since gathering data on every member of a population is infeasible, if not impossible, it is vital that you identify a sample that is representative of your population of interest. We sample for three main reasons: efficiency, representation, and extrapolation.
- Efficiency – Collecting data on just a fraction of the entire population can provide a sense of the entire population’s attitudes. By not tracking down every person’s opinion, you will save time and money.
- Representation – When sampling from a population, it is important to select a sample that portrays the diversity of possible responses within the entire population as accurately, and proportionately, as possible.
- Extrapolation – With a properly selected sample and well-designed data collection strategy, you can interpret your results and apply them with some degree of certainty to the broader population.
What’s the right number of people to include in a sample? The general rule of thumb is the larger the sample size, the better. This is because a larger sample brings you closer to describing the entire population. However, the greater the sample, the greater the expense, time commitment, and room for complications. Interview-based studies often have smaller sample sizes as compared to other types of research projects, as the data collected from each participant has the potential to be more in-depth. Individual interviews are also more time-consuming and intensive than individual surveys. For this reason, surveys are cheaper per person and may have notably larger sample sizes. However, they remain less generative in terms of detailed information. Depending on your research question you can decide which method will serve you better.
Once your interview responses reach what is called “saturation,” the point at which further interviews only reproduce your pre-existing data, it is an indication that your sample is likely large enough (Saunders et al., 2018). You may find yourself at this stage when your interviews or surveys are no longer resulting in the identification of new concepts, responses are not providing new information, or you’ve completed collecting information from your pre-determined categories of respondents. The video below outlines different kinds of saturation.
