Homework answers / question archive / Week 5 Practice with Data For each question provide your Stata code, Stata output, and respond to all questions

Week 5 Practice with Data For each question provide your Stata code, Stata output, and respond to all questions

Health Science

Share With

---

Week 5 Practice with Data

For each question provide your Stata code, Stata output, and respond to all questions.

1.  The file HospitalStay has two variables:  Duration for the number of days a patient is in the hospital and Hospital to designate one of two different hospitals.  It is determined that running a t-test to compare the length of hospital stay is inappropriate due to the data being skewed.  Instead, use a Wilcoxon rank sum test to determine if the length of stay is different in the two hospitals. Make sure to provide your Stata code, output, your decision (reject or fail to reject the null hypothesis), and a conclusion. 

2.  A case-control study is performed with 2388 cases of patients with bladder cancer obtained from cancer registries. The 4660 controls were randomly selected from a similar population in terms of demographics. The purpose of the study is to investigate the possible association between the incidence of bladder cancer and the consumption of alcoholic beverages.  Within cases, 574 individuals are considered drinkers of alcoholic beverages, and 980 of the controls are drinkers. 

Let          p₁ = true proportion of drinkers among the cases

                p₂ = true proportion of drinkers among controls

We wish to test the hypothesis H₀: p₁ = p₂            versus          H₁: p_{1 ≠} p₂

Use Stata to conduct a two-sample test for proportions.  Given the relatively large sample size, assume normal theory can apply.  Is the proportion of drinkers in the control group different from the proportion in the cases?  Use your p-value and a significance level of 0.05 to justify.

3.  Two drugs (A, B) are compared for the medical treatment of ulcers. Although each patient only takes one drug, a patient taking drug A is matched carefully (with regard to age, sex, and clinical condition) with a patient taking drug B. The treatment results based on 200 matched pairs show that for 89 matched pairs, both treatments are effective. For 90 matched pairs, both treatments are ineffective. For 5 matched pairs, drug A is effective (and drug B is ineffective).  For 16 matched pairs, drug B is effective (and drug A is ineffective). 

Use Stata to perform a McNemar’s test (a = 0.05) to assess if there is a significant difference in the effectiveness between the two treatments.  Note that the number of discordant pairs is 21.  Use your p-value and a significance level of 0.05 to assess any differences.