Scribd
Upload
2 views23 pages

Sample Size Requirements for Statistical Tests

The document outlines sample size requirements for various statistical tests, including one-sample mean tests, two-sample t-tests, paired t-tests, one-way ANOVA, proportion tests, chi-square tests, correlation, and regression analyses. For example, 35 participants are needed for a one-sample mean test, while 36 per group are required for a two-sample t-test. It also provides guidelines for surveys and pilot studies, suggesting 10-30 participants for unknown populations.

Uploaded by

Eng. George Sarkiss
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
2 views23 pages

Sample Size Requirements for Statistical Tests

The document outlines sample size requirements for various statistical tests, including one-sample mean tests, two-sample t-tests, paired t-tests, one-way ANOVA, proportion tests, chi-square tests, correlation, and regression analyses. For example, 35 participants are needed for a one-sample mean test, while 36 per group are required for a two-sample t-test. It also provides guidelines for surveys and pilot studies, suggesting 10-30 participants for unknown populations.

Uploaded by

Eng. George Sarkiss
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd

ONE-

SAMPLE
MEAN TEST
You need 35 participants to estimate the average blood pressure within ±2 units with 95% confidence,
assuming the population SD is 6.
TWO-SAMPLE
T-TEST
(INDEPENDENT
)
Used when comparing the means
of two independent groups
(e.g., treatment vs. control).
We want to find how many subjects
are needed in each group to detect
a given difference (Δ) with certain
confidence and power.

You need 36 participants in each


group, or 72 total.
PAIRED T-
TEST
 That formula is for the Paired
t-test (also called dependent
samples t-test).
Here’s a step-by-step
explanation of how to
compute the sample size using
it:

You need 25 paired observations


(participants) to detect an 8-unit
mean change in the outcome with
95% confidence and 80% power.
ONE-WAY
ANOVA (K
GROUPS)
To calculate how many participants are required in each group
when comparing ≥ 3 groups using an ANOVA (Analysis of
Variance).

With 3 groups, an effect size f =


0.25, and 80% power, you need
roughly 52 participants per group
(≈156 total) to detect a medium
effect at the 5% significance level.
ONE-
PROPORTIO
N TEST
You need a sample of about 369
people to estimate a population
proportion of 40% with a ±5%
margin of error at 95%
confidence.
TWO-
PROPORTIO
N TEST
CHI-
SQUARE
TEST
You need about 67 total
participants to detect a medium
effect (w = 0.3) at the 5 %
significance level with 2 degrees
of freedom.
CORRELATIO
N (PEARSON
R)
You need approximately 82
participants to detect a correlation
of r = 0.3
with 95% confidence (α = 0.05)
and 80% power.
SIMPLE
REGRESSIO
N
You need at least 26 data points to
detect a relationship that explains
25% of the variance (R² = 0.25)
with 95% confidence and 80%
power in a simple linear
regression.
MULTIPLE
REGRESSIO
N
You need approximately 355 total
observations to detect a medium
effect size (f² = 0.15) with 5
predictors, 95% confidence, and
80% power.
SURVEYS
(UNKNOWN
POPULATION)
step-by-step for the survey
sample size when population
size is unknown (most-
conservative case uses )
 Guideline:
 10–30 participants.
PILOT  Used to estimate σ or p
for planning the main
STUDY 
study.
Exercise: Choose n=20
participants.

You might also like