BAHIR DAR UNIVERSITY
COLLEGE OF BUSINESS AND ECONOMICS
SCHOOL OF GRADUATE STUDIES
PHD IN MANAGEMENT STUDIES
ARTICLE SUMMARY REPORT
ON
SAMPLE SIZE DETERMINATION: A PRACTICAL GUIDE FOR
HEALTH RESEARCHERS
SUMMARIZED BY:- ASFAW SHIMELS (PhD Candidate)
SUBMITTED TO:- GETANEH MIHRET (Ph.D.)
FEB. 17/2026 G.C
Bahirdar University, Bahirdar, Ethiopia
Summarized by Asfaw Shimels (PhD. candidate)
Title Sample size determination: A practical guide for health researchers
Author Alaa Althubaiti
Co-Author
Journal Journal of General and Family Medicine
Year of Publication 2023
DOI: [Link]
Publisher John Wiley & Sons Australia, Ltd (on behalf of the Japan Primary Care
Association)
Authors credibility and Alaa Althubaiti holds a PhD and is affiliated with:
institutional Affiliation
King Saud bin Abdulaziz University for Health Sciences (Saudi Arabia)
King Abdullah International Medical Research Centre
i
SAMPLE SIZE DETERMINATION: A PRACTICAL GUIDE FOR HEALTH RESEARCHERS
1. INTRODUCTION
In the initial stage of planning a research study, sample size calculation or power calculation answers
the question, “How many participants or observations need to be included in this study?” If the sample
size is low, the research outcome might not be reproducible.1 Informal guidelines for sample size based
on the experience of researchers are used in most research studies and may be sufficient, as is the case
in pilot studies.
2. Why Sample Size Matters
Sample size calculation (sometimes called power calculation) is not just a technical requirement. It is
about balance. If the sample is too small, researchers may fail to detect real effects. This leads to false
negatives and unreliable findings that cannot be reproduced. On the other hand, if the sample is too
large, the study may detect statistically significant differences that are so small they are not clinically
meaningful.
3. Key Concepts Researchers Must Understand
To calculate sample size properly, researchers must think carefully about:
The type of statistical analysis they plan to use
Confidence level (often 95%)
Power of the study (commonly 80%)
Margin of error (precision)
Effect size (the smallest meaningful difference worth detecting)
Among these, effect size is often the hardest to decide. It reflects what difference would truly matter in
practice not just statistically. A small effect requires a large sample to detect. A large effect can be
detected with fewer participants.
4. Sample Size in Different Study Designs
1. Descriptive Studies
In descriptive studies (such as prevalence surveys), researchers aim to estimate a proportion or average.
The required sample size depends mainly on:
Desired confidence level
Margin of error
Estimated standard deviation (for means) or estimated proportion (for prevalence)
1
Interestingly, total population size is often not necessary unless the population is very small and finite.
2. Comparing Two Groups
When comparing two groups (e.g., treatment vs. control), equal group sizes usually provide maximum
statistical power. However, real-world conditions such as cost or data availability may prevent equal
allocation.
The sample size depends on:
The statistical test used (e.g., t-test)
Expected difference between groups
Variability in measurements
Desired power and significance level
3. Regression Analysis
For regression studies (examining relationships between variables), sample size becomes more
complex. Too few observations can cause over fitting, meaning the model works well for the current
data but fails in other populations. More predictors require more observations. Some traditional rules
suggest 10 observations per variable, but the author notes that this may not always be sufficient or
appropriate.
Probability vs. Nonprobability Sampling
The article also distinguishes between:
Probability sampling (random selection, better generalizability)
Nonprobability sampling (e.g., convenience sampling)
When nonprobability sampling is used, formal sample size calculation may be less meaningful because
generalization to a broader population is limited. In such cases, researchers should clearly report how
participants were selected and how many were invited.
In a Nutshell
The article makes an important point: Sample size calculation is not just a formula it is a thoughtful
process. It forces researchers to clarify their research question, define meaningful differences, and
consider feasibility.
The author advocates for transparency, careful planning, and realistic expectations. Sample size
determination should strengthen research not become a mechanical checklist item.