Summary
What is it?
Although a sample size calculator looks simple and only takes a moment to calculate the sample size you need, this guide can help you understand what it actually means.
Why use it?
Understanding sample size (how many responses you need) will help you to confidently frame your research and share your results with others. Each step includes an example to show exactly how you pull all of the information together to work out your sample size.
Before you start
Is this tool right for you? To find out more about selecting the best tool, check out the project planner.
What you'll need
Time | 1 hour |
---|---|
Budget | No cost |
Equipment | Computer and internet |
Top tip
Calculate your sample size before you collect your data. This way, you’ll know how many responses you’re targeting for your results to be representative of the group of people (target population) you want to hear from.
Step by step
1. Define your population type
First, think about who you want to reach:
- Members of the public/your local community
- Your attendees
- Your members or subscribers
- Your networks i.e. colleagues, staff, volunteers or other members of your sector, etc.
The group you select becomes your population type. Example: In this example, we want to survey our local community.
Population type |
Our local community |
2. Estimate your population size
The next step is to estimate how many people are in your population. Example: In this example, we looked up the population of our local community on the Australian Bureau of Statistics website (You can use this site if searching for population stats for a city or region). This showed that the estimated population in 2018 (based on the 2016 census data) is 10,000.
Population type | Estimated population size |
Our local community | 10,000 |
3. Decide how sure you want to be about your results
To decide how sure you want to be, there are two technical terms you need to understand. They are:
- Confidence level
- Margin of error
Confidence Level is how confident we want to be that our survey results represent the views of our population (in this example, our local community). A 95% confidence level is most commonly used by researchers; though, you can choose to have 80%, 85% or 90% (less confident) or 99% (more confident) confidence levels.
Margin of error is the range that the population’s responses (if everybody in the whole local community did the survey) may differ to our actual survey results. Margin of error is also a percentage. A generally acceptable margin of error used by researchers is between 4% and 8%.
Now that you have an understanding, you can decide what confidence level and margin of error you want to have. In this example, we want to have a confidence level of 95% and a margin of error of 4%.
Population type | Estimated population size | Confidence Level | Margin of Error |
Our local community | 10,000 | 95% | 4% |
This means that we will be 95% confident that our survey results represent the views of our population (in this example, our local community). It also means that the population’s response to our survey (if everybody in the whole local community did the survey) may be different to our actual survey results by + or – 4%.
If our results showed that 60% of respondents have visited us before, we can say: We are 95% confident that 56% to 64% of our local community have visited us before. We work this out by taking away and adding the margin of error (4%) to our result:
- 60% – 4% = 56%
- 60% + 4% = 64%
4. Determine your sample size
You’re now ready to calculate your sample size! There are several online sample size calculators. A quick web search will give you plenty of options to choose from. If you’re using an online survey tool, it may even have a built-in calculator for you to use.
Online sample size calculators will ask you to enter three things:
- Your population size
- Your confidence level
- Your margin of error
Press calculate, and you’ll have your ideal sample size. In this example, we entered into a sample size calculator the details outlined in the table below.
Estimated population size | Confidence level | Margin of error |
10,000 | 95% | 4% |
This gave us a sample size (the number of responses you need) of 567.
Estimated population size | Confidence level | Margin of error | Sample size |
10,000 | 95% | 4% | 567 |
Test out different confidence levels and/or margin of error to see how this changes your required sample size.
5. Work out your target response rate
Now you know your required sample size, you need to know what response rate you need to target. You’ll need to know the size of the database you have i.e. the number of email addresses you’ll be sending your survey to.
To calculate your target response rate you need to divide the number of responses you need (your sample size) by your database size. In our example, our local council has agreed to send our survey out to their database of 5,000 emails. We calculate our target response rate by making the calculation below:
Sample size (responses needed) | 567 |
Divided by database size | 5,000 |
Target response rate | 11% |
Our target response rate is 11%.
Check out the encourage participation guide to find out what type of incentive is needed to reach your target response rate.