Module 10 Problem Set

SOC 3391-Fall 2021

Due 10/28/2021

For questions 1 through 10, determine what the appropriate test you would use is.You do not have to test the hypothesis for questions 1-10. For simplicity, I have left off the sizes of each sample.

Possible Answers:

• Independent sample t-test with equal variance
• Independent sample t-test with unequal variance
• Dependent sample t-test
• Z-test for proportions

1. Is the startup load time for computers with Windows 10 (X=5.2, s=3.5) different from that for computers with Windows 8.1 (X=7.1, s=4.2)?
2. Do students who start working with a tutor do better on class assignments after they do (X=72, s=8) than they did before (X=89, s=17)?
3. Are union members more likely to support raising the minimum wage (62%) than non-union members (41%)?
4. Do hourly employees (X=10, s=8.1) take the same number of sick days as salaried employees (X=13, s=10.2)?
5. Does the recovery time following a surgery differ between surgeries performed by robots (X=3.3, s=1.2) and those performed by surgeons (X=4.5, s=3.1)?
6. Are people living in coastal areas more concerned about climate change (P = 0.43) than are those living outside of coastal areas (P = 0.34)?
7. Is the average age at which children first speak different between kids with siblings (X=10.1, s=2.3) and kids without siblings (X=11.2, s=2.5)?
8. For men who got married recently, is there a change in hourly wages for men before (X=18.1, s=20) and after they get married (X=21.5, s=18)?
9. Are speeding tickets more expensive in rural (X=250, s=25) than urban areas (X=285, s=30)?
10. In a matched pair clinical trial, was the immune function of patients different between those on a new drug (X=186, s=12) and those on the placebo (X=126, s=13)?

For questions 11-15, answer the question andshow all intermediate values for full credit.

• For independent sample t-tests: mean values for each sample, the variances for each sample, estimated standard error of the difference in means, the t-ratio, degrees of freedom, the t-critical value, and your decision to reject or retain the null.
• For dependent sample t-tests: mean values for each sample, standard deviation for the difference between groups, standard error of the difference between means, the t-ratio, degrees of freedom, the t-critical value, and your decision to reject or retain the null.
• For proportions: sample proportions, combined proportions, standard error of the difference, z-score, critical z-score, and your decision to reject or retain the null.

Clearlylabel all values and round them to 2 decimal places (e.g., 2.57 instead of 2.5702 or 2). Do not round critical t or z values from your tables.

1. Dating app researchers want to determine if couples that met through an app last longer than do those who met through some other means. In a sample of 25 couples who met through an app, they find that the average length of the relationship is 3 months (s = 2). In a sample of 20 couples who met some other way, the average length of the relationship was 6 months (s=5). Test the null hypothesis that the average length of relationships was the same between these ways of meeting (alpha=0.05).

1. A poll of 645 New Yorkers found that 77% supported a COVID-19 vaccine requirement for enrolling in universities. In a similar sample of 707 Texans, 67% supported a similar mandate. Test the null hypothesis that Texans and New Yorkers have equal support for university mandates of the COVID-19 vaccine (alpha=0.05).

1. A teacher for a toddler classroom has a suspicion that her class prefers crackers to Cheerios. He gathers 10 days of data for each snack (N=10 for each) and finds that the average daily consumption of crackers was 1.5 pounds (s=0.6), and the average daily consumption of Cheerios was 1.2 pounds (s=0.5). Test the null hypothesis that there is no difference in consumption of crackers and Cheerios in this group of voracious toddlers (alpha = 0.05).
2. A Jedi training program is testing whether their training improves the attunement of children to the Force. They do a pre-test when people enter the program and a post-test when they leave. The table below reflects pre and post-test scores for each child tested.

Midichlorians (before)	Midichlorians (after)
2	2.5
2.5	2.5
5	6
4	4.5
3	4
2	2.5
3	3.5
4	4.5
2.5	3
6	10

Test the null hypothesis that the pre- and post-test scores are equal (alpha=0.05).

1. Common to the college experience is the process of cramming the night before a test at the expense of sleep. To determine the effectiveness of this strategy, researchers measured student performance on a test before and after sleeping. The results of their data collection are in the attached dataset. This dataset (Module10_Study) has the following variables:

Before: Score on the test before sleeping

After: Score on the test after sleeping

Using this data and the appropriate statistical test, determine if sleep is associated with a change in test scores. Use alpha = 0.05 and a two-tail test. You must include program output from SPSS/Excel in addition to your answer.