A public health researcher in Delhi conducted a study comparing the effectiveness of two different health education interventions (A and B) on reducing tobacco use among 120 factory workers. The data showed: Intervention A: 45 workers quit tobacco, 15 continued; Intervention B: 38 workers quit tobacco, 22 continued. The researcher wants to determine if there is a statistically significant difference between the two interventions. What is the most appropriate next step in analysis?

Explanation

## Test Selection in Categorical Data Comparison ### Data Type Identification **Key Point:** The outcome variable here is categorical (quit vs. continued tobacco use), not continuous. The independent variable is also categorical (Intervention A vs. B). ### Why Chi-Square Test is Appropriate | Feature | Chi-Square Test | t-test | |---------|-----------------|--------| | **Data Type** | Categorical (frequencies, proportions) | Continuous (means, SDs) | | **Sample Size** | Works well with n > 5 in each cell | Assumes normality, better for continuous | | **Hypothesis** | Tests association between two categorical variables | Tests difference in means | | **Current Data** | Quit/Not quit × Intervention A/B | Not applicable here | **High-Yield:** Chi-square test (χ²) is the gold standard for comparing proportions or frequencies across categorical groups. It tests whether observed frequencies differ significantly from expected frequencies under the null hypothesis of independence. ### Contingency Table Setup ``` Quit Continued Total Intervention A 45 15 60 Intervention B 38 22 60 Total 83 37 120 ``` **Clinical Pearl:** The chi-square test will determine if the difference in quit rates (75% vs. 63.3%) is statistically significant or due to chance. **Mnemonic:** **CATACAT** — CATegory × CATegory = Chi-square test. ### Assumptions Met - Expected frequency in each cell > 5 (easily satisfied with n=120) - Independent observations (different workers in each group) - Random sampling from the population