Study design and statistics · BEGINNER LESSON
Sample size
Sample size is the number of subjects, participants, patients, or experimental units included in a study. It is chosen to make the study large enough to answer the research question while avoiding unnecessary use of resources or participant exposure.
What sample size is
Sample size is simply the count of study units: people, patients, or other experimental units. In quantitative research, it is chosen so the sample can represent the target population well enough to support valid inferences about the population. In many studies, deciding this count is called sample size determination.
[2] A Practical Methodological Primer on Sample Size Determination ...
Why it matters in quantitative studies
An appropriate sample size helps a study detect effects that are scientifically or clinically meaningful while avoiding waste and avoidable exposure. If the sample is too small, the study may be underpowered, meaning it can miss real effects and produce false-negative results; if it is too large, it may use more resources than needed. Sample size planning is linked to effect size, significance level (alpha), and statistical power (1 minus beta).
[2] A Practical Methodological Primer on Sample Size Determination ...
How it works in prediction-model studies
For predictive algorithms, sample size is not only about hypothesis testing. It also affects model fit, calibration, discrimination, clinical utility, and stability of individual predictions. Smaller datasets usually increase model-based error and can reduce fairness and generalisability. In this setting, researchers aim for an acceptable level of model degradation, which means how much performance is lost compared with a perfect model.
[1] A general sample size framework for developing or updating a ...
Mechanisms and practical factors
Sample size planning often balances four linked quantities: effect size, alpha, power, and N. A smaller effect usually requires a larger sample, or else lower alpha or lower power. Practical adjustments may be needed for expected loss to follow-up, unequal group sizes, cluster sampling, or multicenter designs. For prediction models, larger samples are also needed when there are many predictor parameters relative to the number of outcome events.
[2] A Practical Methodological Primer on Sample Size Determination ... [1] A general sample size framework for developing or updating a ...
Limits and special cases
There is no single correct sample size for every study. The right number depends on the research aim, the method, the outcome, and the acceptable level of error. Qualitative research follows different principles: adequacy is not set by power calculations, and data collection often continues until new interviews or observations no longer add important themes. In that setting, concepts such as thematic saturation and information power are used instead of the usual quantitative rules.
[3] Rigorous Qualitative Research Is Defined by Different Principles ...
SEE THE IDEA
How sample size affects research
What the researcher wants to learn
How many units are included
Enough observations for the method
Precision, power, or model performance
The study question guides the sample size choice, and the chosen sample size shapes the quality and reliability of the results.
A larger or better-matched sample often improves quantitative inference, but the needed size depends on the research method and goal.
TRY IT YOURSELF
Test your understanding
Choose an answer, then check the explanation.
PUT IT TOGETHER
Match each concept to its meaning
Drag a label onto its description, or select a label and then select a description. This also works with a keyboard.
4 pairs to match
Sources and further reading
This general lesson was generated from public source material. It explains a concept; it does not interpret a private paper or provide medical advice.