Business Context
NovaBio, a mid-sized research organization, surveyed project leads to understand which statistical methodology they use most often in applied research. Leadership wants to know whether regression-based methods are used more frequently than classical hypothesis testing in current projects.
Problem Statement
Use the survey results to test whether the proportion of projects using regression as the primary methodology is significantly higher than the proportion using hypothesis testing.
Given Data
| Methodology Group | Projects Surveyed | Projects Using Method as Primary Approach | Observed Proportion |
|---|
| Regression-based methods | 180 | 99 | 0.550 |
| Hypothesis-testing methods | 160 | 72 | 0.450 |
Additional parameter:
| Parameter | Value |
|---|
| Significance level | 0.05 |
Requirements
- State the null and alternative hypotheses for a one-tailed test.
- Compute the sample proportions for both groups.
- Calculate the pooled proportion and the standard error under the null.
- Compute the z-statistic and p-value.
- Construct a 95% confidence interval for the difference in proportions.
- Decide whether the evidence supports the claim that regression is used more often.
- Briefly explain what the result means for research-methodology planning.
Assumptions
- Each project is counted once and classified into one primary methodology.
- The surveyed projects are independent across teams.
- Sample sizes are large enough for a normal approximation to the difference in proportions.
- The survey process did not systematically favor one methodology group.
