Step 4: Sampling

In an ideal world, we could interview all clients. However, this takes too long and is too expensive, so instead normally we can only interview a sample of clients. If we do the sampling correctly, the data we get will represent the whole population of clients. To learn more about the principles of sampling, click here and to learn more about how precision in sampling works, click here. These topics are also summarised in the video below.


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. The sampling tool takes you through the steps you’ll need to design a sample for your survey. It will take you through the following decision-making process:

Decide whether to collect data at all franchisees or a selection of them: The sample size you will need will partly depend on whether you interview clients at all the franchisees or just some of the franchisees. If there are many franchisees in your programme, and they are spread out, it may take too much cost and time to collect data from clients at all of them. In this case, you can collect data from a sample of the franchisees.

Determine the sample size: If you only interview clients at a random selection of franchisees, because it is easier, you will need to double your sample size.

  • The minimum sample size when interviewing respondents from all franchisees is 100.
  • The minimum sample size when interviewing respondents from a random selection of franchisees (selected using the sample selector tool) will be 200.

If you can interview more than the minimum sample, that will increase your precision. Use the sample size calculator provided to determine what sample size you should use. If you are not sure about sampling and find it confusing, you can stick to the minimum sample sizes given above.

Decide how many clients to interview at each franchisee and how to select them: The sample selector tool will tell you whether to interview every client, or select clients using an ‘interval’. For example, you may need to select every fifth client.

Decide whether to collect data at the franchisee facility or at the clients’ homes

  • Interviewing at the facility:

The most common way to implement a survey of clients is to interview the clients as they leave the facility, just after receiving the service. This means that data collection team members wait at the facility all day. The team will select clients according to the sampling guidance which they have been trained in, and ask selected clients if they could interview them just before they leave the facility.

  • Interviewing at the clients’ homes:

If the franchisees serve very few clients per day, the data collection teams may have to wait at the franchisees for too long, making data collection expensive. An alternative is to have the franchisees collect the names and contact details of all the clients they serve over a one month period. You can then select the required number of respondents from the listed respondents at random. The data collection team will then visit these clients at their homes and interview them.

Whichever approach is used, the interviewers must find a private and comfortable place to conduct the interview, where the respondents know no one can hear their answers.

Step 5: Obtain the relevant approvals and permissions
Step 3: Plan the survey