Fasting blood sugar was measured in 50 diabetic patients before starting a new exercise programme and again after 3 months in the same patients. The investigator wants to know whether the mean change is statistically significant. The most appropriate test is:
- A Unpaired t-test
- B Paired t-test ✓
- C Chi-square test
- D One-way ANOVA
Explanation
When the same subjects are measured twice, the observations are paired, and the correct analysis is the paired t-test applied to the differences within each subject. The unpaired (Student) t-test requires two independent groups, so it would ignore the within-subject correlation and lose statistical power. ANOVA compares three or more independent group means, and chi-square tests proportions, not continuous means.
Reference: Mahajan's Methods in Biostatistics for Medical Students and Research Workers, 9th ed.
High-yield for: NEET PGINI-CETNExTFMGEUSMLEPLABMRCP
Written and medically reviewed by the StethoPrep medical team.