Tests of Significance — t, chi-square MCQ — NEET PG Practice Question | NEETPGAI
Tests of Significance — t, chi-square
medium
users PSM
A public health researcher in Delhi conducted a study comparing the effectiveness of two different health education interventions on reducing tobacco use among 200 factory workers. Group A (n=100) received intensive counseling, while Group B (n=100) received standard written material. After 6 months, 65 workers in Group A and 45 workers in Group B had quit tobacco use. The researcher wants to determine if the difference in quit rates between the two groups is statistically significant. Which test of significance should be used, and what is the most appropriate null hypothesis?
A. Chi-square test; H₀: There is no association between type of intervention and tobacco cessation
B. Unpaired t-test; H₀: The mean quit rates in both groups are equal
C. Paired t-test; H₀: The difference in quit rates before and after intervention is zero
D. Fisher's exact test; H₀: The odds ratio of quitting is equal in both groups
Explanation
Test Selection for Categorical Data Comparison
Key Point
When comparing categorical outcomes (quit/not quit) between two independent groups, the chi-square test is the appropriate test of significance.
Why Chi-Square Test?
1.
Data Type: Both variables are categorical (binary outcome: quit vs. did not quit; group: intervention A vs. intervention B)
2.
Sample Size: With n=100 in each group and expected frequencies >5 in all cells, chi-square assumptions are met
3.
Independence: The two groups are independent (different workers in each group)
Null Hypothesis Formulation
High-YieldNEET PG
The null hypothesis for chi-square test states that there is no association between the categorical variables (intervention type and cessation outcome).
Contingency Table Structure
Table
Intervention
Quit
Did Not Quit
Total
Group A (Counseling)
65
35
100
Group B (Written)
45
55
100
Total
110
90
200
Chi-Square Test Formula
χ2=∑E(O−E)2
where O = observed frequency, E = expected frequency
Clinical Pearl
In public health intervention studies, chi-square is the standard test for comparing binary outcomes (success/failure, disease/no disease, intervention effect/no effect) across independent groups.
When to Use Alternative Tests
Fisher's exact test: Used when expected frequencies are <5 in any cell (small sample sizes)
t-test (unpaired): Used for continuous data (e.g., comparing mean blood pressure, mean weight loss)
t-test (paired): Used for before-after measurements in the same subjects
Mnemonic
CATS — Categorical data → chi-square; Alternative test (Fisher's) when frequencies are small; T-test for continuous data; Same subjects (paired) vs. different subjects (unpaired)
Practice similar questions
Sign up free to access AI-powered MCQ practice with detailed explanations and adaptive learning.
