How do I calculate sample size for my MD/MS thesis?

For prevalence studies, use Cochran's formula n = Z²p(1−p)/d², where p is the expected prevalence from a previous study, d is the absolute precision (usually 5%), and Z is 1.96 for 95% confidence. For comparative studies, use the two-means or two-proportions formula with 80% power. This free calculator does all three and writes the protocol paragraph for you.

What prevalence value should I use in the formula?

Take the prevalence or incidence of your condition from a previously published study similar to yours — ideally from the same region or population. Cite that study as your reference in the protocol. If a range is reported, you may use the mean value.

Should I add dropouts to my calculated sample size?

Yes. It is standard practice to inflate the calculated sample size by 10% to account for non-response, incomplete records or loss to follow-up. This calculator includes an optional dropout adjustment.

Is this sample size calculator free?

Yes — completely free, unlimited use, no sign-up. It is provided by Glomerulus, a doctor-led thesis support service for medical postgraduates in India.

Sample Size Calculator for MD, MS & DNB Thesis (Free)

How is sample size calculated for a medical thesis?

Sample size calculation is one of the first things your ethics committee and thesis evaluators check — and the most common reason protocols get sent back. The method depends entirely on your study design and primary objective.

1. Prevalence / descriptive studies (single proportion)

If your study estimates how common a condition is — for example, "prevalence of anaemia among antenatal women" — use Cochran's formula: n = Z²p(1−p)/d². You need three inputs: the expected prevalence p from a previously published study (cite it!), the absolute precision d (usually 5%), and the Z value (1.96 for 95% confidence).

2. Comparing two means

If your primary outcome is continuous — comparing mean blood loss, mean HbA1c, mean VAS score between two groups — use n = 2(Z_α/2+Z_β)²σ²/d² per group, where σ is the pooled standard deviation from a reference study and d is the clinically meaningful difference between group means.

3. Comparing two proportions

If your outcome is categorical — comparing complication rates, success rates or cure rates between two groups — use n = (Z_α/2+Z_β)²[p₁(1−p₁)+p₂(1−p₂)]/(p₁−p₂)² per group.

Examiner tip: Always state the reference study from which you took your prevalence or expected difference, and always add ~10% for dropouts. A sample size without a cited reference is the single most common protocol query.

Frequently asked questions

What prevalence value should I use?

Take it from a previously published study similar to yours — ideally the same region and population — and cite that study in your protocol. If a range is reported, use the mean.

Should I add dropouts to the calculated number?

Yes — inflate by about 10% for non-response, incomplete records or loss to follow-up. The calculator above does this automatically.

My calculated sample size is too large for my study period. What do I do?

Options include widening precision (d) from 5% to 7–10%, choosing a different reference prevalence closer to 50%... carefully — or redefining the study question. This needs judgement; message us and we'll sort it out.

Is this calculator free?

Yes — unlimited use, no sign-up, no payment. Bookmark it and share it with your co-residents.

Tool and guide reviewed by Dr. Simon Jude, MD (Community Medicine) — Assistant Professor, SPSS & Research Methodology certified (PGIMER). Last updated June 2026.

Related free tools

Study Design Finder — find the right design for your research question
Statistical Test Chooser — pick the correct test for your analysis
How to Write a Thesis Protocol — full guide

Sample Size Calculator for MD, MS & DNB Thesis

Want this done for you — with the exact formula your university expects?