Statistics
STA 544
Homework 7
Work on the following problem set and show your work within the document. Use SPSS as needed.
1. What is the difference between parametric and nonparametric procedures? When might you use a nonparametric test instead of a parametric test?
2. What nonparametric procedure would you use to determine if the number of occurrences across categories is random?
3. What nonparametric procedure would you use to compare two independent samples when you have violated the assumptions associated with a t test?
4. If you wanted to examine the correlation between rank in high school and rank in college, which procedure would you use?
5. Binge drinking by gender. Wechsler and coworkers (1994) found that 19.4% of students at 4-year U.S. colleges engaged in frequent binge drinking. Table 1 reports the data from this study by gender.
TABLE 1. Frequency of binge drinking by gender. Binge drinking is defined as having five or more drinks in a row three or more times during the past 2-week period.
Binge + Bing
1630
5550
1684
8232
Men
Women
a. Calculate the prevalence of binge drinking for males and for females separately.
b. Fill in this blank: In absolute terms, males had a ____% greater prevalence of binge drinking than females.
c. Fill in this blank: In relative terms, males had a ____% greater prevalence of binge drinking than females.
d. Calculate a 95% confidence interval for the prevalence ratio of binge drinking for males.
6. University Group Diabetes Program. A multicentered clinical trial completed in the 1960s assessed the efficacy of various oral hypoglycemics, insulin, and diet in preventing vascular complications of diabetes.cc It found that 26 of 204 patients (13%) treated with the oral hypoglycemic phenformin died unexpectedly from cardiovascular events. In contrast, 2 of 64 control patients (3%) experienced a similar outcome.
a. Is an exact procedure (e.g., Fishers) necessary to test these data or can a chi-square test be used? Justify your response.
b. Test the difference for significance.
