.png%3Fwidth=390%26height=50%26name=Layer_1%20(1).png)
Fifty-seven per cent of Australians switched at least one provider in the past year. Not considered it, not compared plans in a
When it comes to market research surveys, one of the most important decisions you’ll make is determining the right sample size. This number will determine how accurate your results are, and it’s crucial to get it right. So how do you calculate the perfect sample size? It’s not as difficult as you might think!
In this article, we will walk you through the process step-by-step. We’ll also provide a few tips to help you ensure that your survey produces accurate results.
What is Sample Size?
To begin with, let’s define what a sample size is. A sample size is the number of people who participate in your market research survey. This group will be used to represent a bigger population and provide you with insights on the opinions, behaviors, and characteristics of that larger segment. It’s important to get your sample size right because it can directly affect the accuracy of your results. If your sample size is too small, you won’t have enough data points to draw useful conclusions from. On the other hand, if it’s too large, then you could be wasting time and resources collecting more data than necessary.
Variables needed for a sample size calculation
Before you can determine the right sample size for your survey, there are a few variables that you will need to consider. These include, the population size of your target segment, the margin of error you’re willing to accept, and the confidence level you want to achieve. The population size is important because it helps establish how large or small of a sample size is required; while the margin of error and confidence level will help determine how accurate your results must be.
1. Population size:
Population size refers to the total number of people in a given target segment. For example, if you want to conduct market research on Australian adults aged 18-35, your population size would be the total number of Australian adults within that age group.
This is an important factor to consider when calculating the sample size, as it will help determine how large or small your sample needs to be. In general, larger populations require larger samples in order to produce accurate results.
2. Margin of Error (Confidence interval):
Margin of error (also known as a confidence interval) is the range of results that you’re willing to accept when conducting research. It’s expressed as a percentage and it reflects how much variation or uncertainty you are comfortable with in your data.
For example, say 60% of your respondents prefer your brand and your margin of error is 5%. The true figure across the whole population is then likely to sit somewhere between 55% and 65%. The smaller the margin of error you want, the larger the sample you need.
3. Confidence level:
The confidence level is how certain you can be that your result falls within the margin of error of the true population value. It is expressed as a percentage, and in market research it typically sits between 90% and 99%, with 95% the industry standard. The higher the confidence level you want, the larger the sample you need.
For example, at a 95% confidence level, if you ran the same survey 100 times you would expect the result to land within your margin of error in about 95 of them.
4. Standard deviation:
Standard deviation is a statistical measure that describes how far away from the mean (average) of a particular set of numbers the individual numbers are. It’s used to calculate the amount of variability in a data set, and it can help you determine if your sample size is large enough to produce accurate results.
If everyone gives similar answers, the standard deviation is low. If answers vary widely, it is high. The catch when sizing a survey is that you do not know the standard deviation until you have collected the data. The convention is to assume maximum variability, a 50/50 split on a yes or no question, which produces the safest (largest) sample size estimate. That is the assumption behind most sample size calculators, including ours.
Now, when it comes to calculating sample size, you need to decide how large a sample needs to be in order for your results to capture this variability accurately. To do this, you can use a formula that takes into account both population size and standard deviation. The larger your population size and/or higher your standard deviation is, the larger your sample will need to be in order to produce accurate results.
Sample size determines how precisely you can measure; what you measure matters just as much. Our guide to the 10 metrics to include in a brand tracker covers that side of the design.
How to Calculate Sample Size
Before you can calculate your sample size, you will also need to know your z-score. A z-score is a statistical measure that tells you how many standard deviations away from the mean (average) of a data set a particular number lies. The higher the z-score, the more accurately your sample size will represent the target population.
Your z-score is based on the confidence level you want your market research to achieve. A list of typical confidence levels used for market research and their associated z-score can be found in the table below:
| Confidence Level | Z-Score |
| 80% | 1.28 |
| 85% | 1.44 |
| 90% | 1.65 |
| 95% | 1.96 |
| 99% | 2.58 |
Automatic Calculation (Using Calculator):
To help simplify your sample size calculation, we have created an easy to use sample size calculator. Enter your desired inputs below to see what sample size you need for your market research survey.
Population Size
Population Size
The total number of people whose opinion or behavior your sample will represent.
Confidence Level (%)
The probability that your sample accurately reflects the attitudes of your population. The industry standard is 95%.
95
Margin of Error (%)
The range (measured as a percentage) that your population’s responses may deviate from your sample’s.
0
The number above is the easy part.
The harder questions are who should be in the sample, how it is sourced, and whether the quality holds up wave to wave, because a perfectly sized sample of the wrong people produces confident, wrong answers. If you are designing a study and want a second opinion on the sampling approach, talk to us about your study design. It costs nothing to ask.
Manual Calculation:
If you would like to have a go at calculating your sample size yourself, then plug these figures into the formula below.
If you know the total population size of the audience you can use Formula A. Otherwise use Formula B, for an undetermined or unlimited population size.
Formula A
|
Formula B
|
Final thoughts...
Generally speaking, when calculating sample size it’s a good idea to aim for an absolute minimum of 300 respondents. This will ensure that your survey results are reliable and statistically significant. You may also want to consider increasing this number if you need a very precise result or if you are targeting a particularly narrow sub-population.
It is also important to remember that the larger your sample size, the more accurate your data will be – however, there is a point of diminishing returns where adding further people to the survey offers marginal benefit in terms of accuracy. Therefore, it pays to find the right balance between accuracy and cost-efficiency by selecting an appropriate sample size for your market research survey.
By taking all of these variables into consideration, you can determine what an appropriate sample size is for conducting market research on your target segment. It’s important to note, however, that the sample size you choose will depend on the specific objectives of your research – so it’s best to consult with a professional market researcher in order to get an appropriate number for your particular project.
If the survey you are sizing is a brand tracker, sample design decisions compound: the sample you choose in wave one constrains every comparison you make after it. Brand Health designs custom tracking programs where the sampling holds up to commercial scrutiny. Get started with a free consultation or see how our research programs are priced.
Related posts
At some point, most Australian marketing teams that get serious about brand measurement face the same fork: a syndicated tracking
Between now and November, most Australian marketing teams will lock their 2027 plans. Budgets will be argued, channel splits decided,
In June, Ferrari sold 16 new cars in Australia. Lamborghini sold 18. Fiat, a brand every Australian can name and most can picture down