Question

A study was undertaken to assess the effect of a drug in lowering serum cholesterol levels. 15 obese women and 10 non-obese women formed the 2 limbs of the study. Which test would be useful to correlate the results obtained?

• A. Unpaired t test
• B. Paired t test
• C. Chi square test
• D. Fischer test

Hint:

Before-and-after readings taken from the same person need a paired t test, not an unpaired one.

Answer 1

The Correct Option is B

This question is about picking the right statistical test for a before-and-after drug study on two groups of women, obese and non-obese.

1. Unpaired t test: This test compares the means of two separate, independent groups of subjects, such as the obese group compared against the non-obese group as two different samples. It does not fit a before-after comparison done on the same women, so it is not the test asked for here.
2. Paired t test: Each woman in a limb gives two readings, cholesterol before the drug and cholesterol after the drug. Since both readings come from the same person, they are linked, and the paired t test is built exactly for comparing such related pairs of measurements. This fits the study design.
3. Chi square test: This test is meant for categorical data, such as counts sorted into groups, not for continuous numeric readings like cholesterol level in mg/dL. It does not apply here.
4. Fischer test: Fischer's exact test is used for small sample categorical data arranged in a 2x2 table, again not suited to continuous before-after cholesterol values.

Since the cholesterol level in each woman is measured twice, once before and once after the drug, the two values are paired within the same subject, and the paired t test is the correct choice to correlate these results.

Let's summarize:

• Same subject measured twice (before-after) needs a paired t test.
• Two separate independent groups measured once each would need an unpaired t test instead.

So the test useful here to correlate the results obtained is the paired t test.