Principles of sampling

We have pre-prepared a sample design for you and created an Excel tool that will help you determine the sample size you need, which franchisees you should collect data from, and how many clients to interview at each of those franchisees. There is a short ‘how-to’ guide to using the tool included in the toolkit.

The three main principles of sampling are:

  • Selecting clients at random will help avoid selection bias

Random selection of clients is essential to getting a good dataset. There are no options here, the selection must be random and you must avoid any type of human selection. This means you cannot only choose clients that are in easy and convenient places, you cannot select clients from the best franchisees and so on. Every client of the franchise should have an equal chance of being chosen (except where you are doing some sort of over-sampling for a specific sub-group, which we suggest you avoid – if you do want to do that, please see Annex 1).

 

  • Make the sampling feasible and affordable

To get the perfect dataset, you would interview every single franchise client that comes for services. This is too expensive and time-consuming, so instead we interview a sample of clients. So, the whole point of sampling is to reduce costs and time. The downside of sampling is that it reduces accuracy. Determining the sample that you need to use will be a matter of balancing feasibility and affordability against accuracy.

  • Make the sample large enough that the results have the precision that you need

To make things simpler, we have provided a minimum sample size that will provide you with an acceptable level of precision below. If you can make the sample bigger than the minimum, that is good as it will increase precision. If not, then using the minimum is acceptable, but please don’t go below the minimum.

A random sample that’s too small is like shooting at a target from far away – you are more likely to be far off.
A large sample that’s not random is likely to have selection bias, which means that you will hit the target accurately, but it will be the wrong target!
Precision in sampling
Analysing with EpiInfo (text instructions)