Inferential Statistics
(k) Carry out a test of significance to decide whether Cal Poly males and females differ with respect to how likely they are to report back problems.
1. Define the parameter(s) of interest.
2. State the hypotheses both in symbols and in words.
3. Identify an appropriate test procedure and state and if a theory-based approach comment on the validity conditions.
4. Use technology to carry out the test of significance and 95% confidence interval and copy the output into your report.
Press one of the links below to view the appropriate technology instructions.
- From your previous results, use the Contingency Analysis hot spot to select Two Sample Test for Proportions.
- Use the radio button to toggle to "Yes" as the response outcome of interest.
- Which is the appropriate p-value?
|
- Choose Stat > Basic Statistics > 2 Proportions
- Specify the Backproblems variable in the Samples box and the Gender variable in the Sample IDs box.
- Under Options change the direction of the Alternative hypothesis.
- Press OK twice.
- Note: If using a one-sided alternative, you will need to retrace your steps and use a two-sided alternative to find a 95% confidence interval.
|
- Tell R the success counts and the sample sizes: prop.test(x = c(24, 8), n = c(55, 45), alternative="XXX")
- Enter less, greater, or two.sided for XXX. Note: You may have to run the command once to see in which direction R is subtracting.
- For the confidence interval: Retrace your steps use "two.sided" for the alternative.
|
- You can use the Theory Based Inference applet, typing in the sample sizes and counts, or pasting in the data from the Excel file.
- Check the Test of significance box and specify the direction of the alternative. Press Calculate.
- For the confidence interval: check the Confidence interval box.
|
5. State your decision and summarize your conclusions in context, including interpreting the confidence interval in context.