How many respondents do you need? Use our sample size calculator to get the right sample size for your next survey. We also answer frequently asked questions on sample size and provide suggestions to improve your surveys in this handy guide.
Online Survey Panels
Ensure high-quality data with guaranteed high-quality target respondents
What is a sample and why sample size matters
Researching the entire population is often not feasible due to many constraints. Instead, sampling is used, which is more practical for analysis and generates results more promptly.
A sample is a subset of individuals selected from the target population to estimate the characteristics of the population. Sample size is important because it determines the accuracy and quality of your survey results. It is especially important in market research as robust data is the key to unlocking meaningful and actionable insights about your target audience.
Want to identify the number of responses required for your survey quickly? Just use our sample size calculator.
If you want to learn more about the theories behind the calculator and sample size, we also compiled everything you need to know in this guide.
Sample Size Calculator
Four important components of sample size calculation
Here are four key components you’ll need to understand in calculating your minimum sample size:
1. Population size
The population size is the total number of people in the population (target audience) you are looking to survey. For example, if you were surveying the general population of a city with 30,000 residents, the population size for your study is 30,000.
When surveying small populations, a finite population correction is used. You can use 100,000 with caution if you do not know the exact size of a large population; the sample size remains quite static over 100,000.
Get quality-assured responses directly from Conjointly
Reliable market research starts with quality sample. With access to millions of consumers from broad communities to deep-profiled niche markets, Conjointly provides numerous options to delivers quality respondents straight into your survey, starting at only $0.90 per response.
2. Confidence level
The confidence level indicates the probability that your confidence interval contains the true population value, if you were to repeat your study with a random sample selection.
For example, with a confidence level of 95%, your study could be repeated several times and your population would fall within the confidence interval on 95% of these occurrences.
The accuracy of your results is determined by how high your confidence interval is. You should also note that as your confidence level increases, so does your required sample size.
Conjointly’s default confidence level is 90% as it is most useful in business settings. However, you can easily adjust the confidence levels in your Conjointly reports.
3. Confidence interval (Margin of error)
The confidence interval is a range of values that are likely to contain your population value and is expressed in percentage points. For example, a confidence interval of ±2% means that your true population will lie either 2 percentage points above or below where it is shown in your results.
Your confidence interval will be higher or lower than your specified value as it is based on your actual sample proportion, rather than your estimated proportion. It also depends on how accurate the results you are seeking are (determined by your confidence level). A higher confidence level results in a wider confidence interval.
Check out the margin of error for different survey sample sizes under the standard 90% level of confidence.
4. Sample proportion
Your sample proportion (expressed as a percentage) indicates your estimated results and is calculated using a Quick Feedback study or previous survey results. Otherwise, 50% is a safe sample proportion to use without hindering the accuracy of your calculated sample size.
What is the formula for determining sample size?
The formula for calculating sample size is:
$$ n = \frac{ N \frac{ z^2 p(1-p) } {e^2} } { \frac{ z^2 p(1-p) } {e^2} + N - 1 } $$where:
nis the sample size,Nis the population size,zis the confidence level (in percent, such as90% = 0.9),pis the sample proportion (in percent, such as50% = 0.5),eis the margin of error (in percent, such as5% = 0.05).
FAQs
What is a good sample size for a market research survey?
The ideal sample size for your study will depend on your population size, confidence level, and confidence interval. The survey design and methodology you use may also require you to adjust your sample size.
A good sample size will satisfy your criteria for accuracy in quantitative research results. It is usually determined by a combination of expected confidence, budget and resource availability for analysis.
Generally, the bigger the sample size, the more accurate your results will be. However, you should consider that large sample sizes are costlier and can complicate your study. You can use our sample size calculator to determine the minimum sample size for your project to ensure your survey is accurate and cost-effective.
What sample size do I need for 95 confidence?
Once you have selected your confidence level as 95%, you will need to determine your confidence interval and population size to calculate the sample size required for your study.
The sample size will be larger for 95% confidence than 90%, if all other factors remain the same.
Even though we recommend using 90% confidence level to achieve the right balance between cost and precision, you can increase it up to 99 percent. However, 99% confidence more than doubles the size of the required sample, compared to 90% confidence. This might produce unnecessary bloat of data for most quantitative research projects.
How do you find the sample size when given the margin of error?
To calculate sample size using your margin of error (confidence interval), you must also know your confidence level and population size.
The margin of error is usually expressed as ± value. This means your survey results can represent a range of values plus or minus the value of the margin of error.
Consider a survey that indicates that 40% of car owners in the USA would not consider the colour red for their next car. With a 5% margin of error, you should express your findings as 35-45% of car owners in the USA would not consider a red car. To gain more accurate insights, you could lower your margin of error to 2%, for example, and expect your findings to lie within a smaller confidence interval of 38-42%.
Why does the margin of error decrease as the sample size increases?
You will see in our sample size calculator, reducing your margin of error results in a large increase in recommended sample size.
It should be noted that this is not a linear relationship, increasing sample size will result in diminishing returns in the margin of error reduction. For example, the following chart illustrates the relationship between the margin of error and sample size, for a population of 1 million, seeking a 90% confidence interval, with a 0.5 sample proportion.
As shown above, moving from a 5% margin of error to a 1% margin of error would require a sample size over 20 times as large.
Why is a small sample size bad? Why is a larger sample size better? Why must sample size be greater than 30?
How do you justify a small sample size in quantitative research?
A small sample size can be justified when:
- The whole population is small. For example, in B2B research, when there are only 100 customers, then it is OK to sample ~30 to get a view of the opinions of the whole customer base.
- The cost of sampling is prohibitive.
Please note, it is not justifiable for most consumer surveys as they require a sample of hundreds of consumers from your product category, including current users and potential buyers. Conjointly provides numerous options and delivers quality respondents straight into your survey, starting at only $0.90 per response.
How does sample size affect validity of a study?
If sample size is not large enough for your study, the internal and external validity will be compromised and it can also result in cases of bias. A higher sample size results in more accurate results. Sample size is only one of many factors that affect the validity of your study.
You should also consider that data can still be useful in some instances even if you do not meet the required sample size for your study. For example, a customer satisfaction survey with open-ended questions can still provide meaningful insights as detailed consumer feedback is indispensable to market research.
