Key takeaways
- Sample size refers to the number of people or observations included in a study
- It directly impacts the accuracy and reliability of research findings
- The right sample size balances precision with time and cost efficiency
- Representativeness matters as much as size in producing meaningful insights
What is sample size?
Sample size is the number of individuals, responses, or observations included in a research study to represent a larger population.
For example, if you survey 1,000 people out of a population of 1 million, your sample size is 1,000.
Understanding what sample size is plays a critical role in research quality. A well-chosen sample size helps ensure that findings accurately reflect the wider population. This is especially important in market research, where decisions about products, campaigns, and strategy rely on dependable insights.
At Kantar, understanding what determines sample size helps organisations generate insights they can trust. By applying robust sampling principles, researchers can make confident, data-backed decisions that stand up to scrutiny.
Why sample size matters in research
Sample size is one of the most important factors influencing the credibility of research results.
A sample that is too small can lead to misleading conclusions, while one that is unnecessarily large can waste time and budget. The key is finding the right balance.
Three core concepts explain why sample size matters:
- Confidence level: This indicates how likely it is that your results reflect the true population. A 95% confidence level means that if you repeated the study multiple times, results would fall within the expected range in most cases.
- Margin of error: This shows how much your results may vary from the true population value. For example, a ±3% margin of error means the actual value could reasonably fall within three percentage points above or below your result.
- Statistical power: This refers to the ability of a study to detect real differences or patterns within the data.
To illustrate, a survey of 50 people may produce a margin of error of around ±14%, while a study with 1,000 respondents could reduce this to around ±3%. The larger the sample, the more precise your results typically become.
However, bigger is not always better. The goal is to achieve reliable results efficiently. Sample design should be carefully managed to ensure accuracy without unnecessary cost or complexity.
Factors that influence sample size
There is no single ‘correct’ sample size for every study. The right number depends on several factors.
Survey objectives and research design
Your research goals play a central role in determining sample size.
For example:
- Measuring brand awareness typically requires a broad, representative sample
- Testing reactions to a concept may require smaller, more targeted groups
Clearly defined survey objectives help ensure your sample size aligns with what you are trying to learn.
Population size
It is important to distinguish between population and sample.
- The population is the full group you want to study. It includes all individuals that fit your research criteria. For example, if you’re conducting a study on Canadian adults, the population would be all adults living in Canada.
- The sample is a subset of the population that’s selected for data collection that ideally represents the larger group.
In many cases, increasing the population size does not dramatically increase the required sample size. For example, a study of hundreds of thousands of consumers may still achieve reliable results with a few hundred carefully selected respondents.
Margin of error and confidence level
These two concepts are closely linked to sample size.
- A smaller margin of error requires a larger sample
- A higher confidence level also increases the required sample size
Most market research studies aim for:
- 95% confidence level
- ±3–5% margin of error
Variability in the data
If your audience has diverse opinions or behaviours, a larger sample size is often needed to capture that variation accurately.
For example:
- Attitudes toward a new product launch can vary widely
- Feedback from a small, homogeneous group may vary less
Higher variability increases the need for larger samples.
Available resources
Practical constraints also influence sample size, including:
- Budget
- Timeline
- Access to respondents
Working with a research partner helps ensure these constraints are balanced with statistical requirements. Kantar’s global reach makes it easier to access high-quality respondents efficiently across markets.
How to determine sample size
Understanding how to determine sample size allows you to plan research more effectively.
Step-by-step process
Define your research goals
Start by clarifying your survey objectives.
Identify your population size
This includes all individuals that fit your research criteria. For example, all potential buyers of a new product.
Choose your confidence level and margin of error
Decide how much error you are willing to accept in your results and the level of certainty you want that the study population would provide answers within the margin of error. Commonly 95% confidence and ±5% margin of error.
Estimate population variability
If unknown, use 0.5 as a standard estimate
Estimate the sample size
Use a sample size formula or calculator, which considers the population size, margin of error, and confidence level. You can often find online calculators that can help you plug in these values to get a sample size.
Sample size formula: n = (Z^2 × p × (1-p)) / E^2]
Here’s what each part means:
- n = the sample size you’re trying to find.
- Z = the z-score that corresponds to your chosen confidence level (1.96 for 95%, 2.58 for 99%, 1.65 for 90%).
- p = the estimated proportion of your population with a certain characteristic (0.5 if unknown).
- 1 - p = the remaining proportion of the population that doesn’t have that characteristic.
- E = the margin of error you’re willing to accept, written as a decimal (for ±5%, use 0.05).
Adjust for subgroup analysis
If you plan to analyse specific subgroups within your population (e.g., age, income level), consider increasing your sample size to ensure that those segments are adequately represented.
Example calculation
Imagine you want to estimate smartphone usage in a city of 500,000 people.
Using a 95% confidence level and ±5% margin of error, the calculation results in a required sample of approximately 385 respondents.
This demonstrates how a relatively small, well-designed sample can represent a much larger population effectively.
Tools and professional support
While online calculators can provide quick estimates, more complex studies benefit from expert input.
Kantar’s research services team applies proven methodologies and data science expertise to determine optimal sample sizes across global markets, ensuring results are both reliable and efficient.
Sample size and representativeness
Sample size alone does not guarantee accuracy. Representativeness is just as important.
A smaller, well-balanced sample can often produce more meaningful results than a larger, biased one.
A representative sample reflects key characteristics of the target population, such as:
- Age
- Gender
- Geography
Learn more about building a representative sample here.
Avoiding common pitfalls
Even well-designed research can be undermined by common sampling issues.
- Sample size too small: Results may be unreliable and unrepresentative
- Sample size too large: Resources may be wasted without meaningful gains in accuracy
- Sampling bias: Certain groups are over- or under-represented
- Non-response bias: Those who respond may differ significantly from those who do not
At Kantar, advanced weighting techniques, rigorous quality checks, and customised screening processes help minimise these risks and ensure robust, credible insights.
How Kantar ensures reliable results
At Kantar, rigorous sampling methodologies are combined with advanced technology and global expertise.
Key strengths include:
- Access to diverse, high-quality respondent panels worldwide
- Robust quality control and data validation processes
- Advanced analytics and modelling techniques
- Compliance with recognised research standards
This ensures every study achieves the right balance of accuracy, efficiency, and representativeness.
FAQs about sample size
What is a good sample size for a survey?
A typical range is between 300 and 1,000 respondents for general population studies. Smaller samples may be suitable for niche audiences, depending on research goals and required precision.
What is the difference between sample size and population size?
The population is the full group you want to study, while the sample size refers to the smaller group you actually survey.
Why does margin of error matter?
It indicates how close your survey findings are likely to be to the true population value. A smaller margin of error means more precise results but requires a larger sample.
How do I know if my sample is representative?
A representative sample reflects important characteristics of your population. This can be achieved through techniques such as quota sampling and weighting.
For more detail, see our guide to creating a representative sample.
Conclusion
Understanding how to determine the right size is essential for producing reliable, actionable research.
The right sample size:
- Improves accuracy and confidence in findings
- Balances precision with efficiency
- Supports better business decisions
Working with experienced research partners ensures your study is designed for success from the outset.
Want to learn more?
For a deeper dive, explore our online training module on determining sample size and build your confidence in research design. Or contact our research experts today.
