A public health researcher in Delhi conducted a study comparing the effectiveness of two different health education interventions (video-based vs. pamphlet-based) on reducing tobacco use among 120 factory workers. The outcome was binary (quit smoking: yes/no). After data collection, the researcher needs to determine whether there is a statistically significant difference in quit rates between the two groups. What is the most appropriate next step in statistical analysis?

Explanation

## Test Selection for Categorical Data **Key Point:** When comparing categorical outcomes (binary: yes/no) across two independent groups, the chi-square test of independence is the gold standard. ### Why Chi-Square is Correct The data structure is: - **Independent variable:** Intervention type (categorical: video vs. pamphlet) - **Dependent variable:** Smoking cessation outcome (categorical: quit vs. not quit) - **Sample size:** n = 120 (adequate for chi-square) This creates a 2×2 contingency table: | Intervention | Quit | Not Quit | Total | | --- | --- | --- | --- | | Video | a | b | a+b | | Pamphlet | c | d | c+d | | Total | a+c | b+d | 120 | The chi-square test statistic is: $$\chi^2 = \frac{N(ad - bc)^2}{(a+b)(c+d)(a+c)(b+d)}$$ **High-Yield:** Chi-square tests categorical association; t-tests compare means of continuous variables. ### Assumptions Met - Both variables are categorical (nominal) - Groups are independent (different workers, different interventions) - Expected frequency in each cell ≥ 5 (likely, given n = 120) **Clinical Pearl:** In public health intervention studies, binary outcomes (disease present/absent, behavior adopted/not adopted) are extremely common—chi-square is your go-to test.