Question

5. The article "Evaluation of a Ventilation Strategy to Prevent Barotrauma in Patients at High Risk for Acute Respiratory Distress Syndrome" (Stewart et al. New Engl.J of Med., 1998: 355-358) reported an experiment in which 120 patients with similar clinical features were randomly divided into a control and a treatment group, each of size 60. The mean ICU stay (days) and sample standard deviation for the treatmert group were 19.9 and 39.1 respectively, and those for the control group were 13.7 and 15.8. Let ur be the population mean of the treatment group and uc be the population mean of the control group. (In parts (a) and (b) you may assume equal variances in the two groups.) (a) Is there sufficient evidence to justify claiming there is a difference in mean ICU stay between the treatment and control groups? Use a 0.05, clearly state your null and alternate hypotheses and give the value of your test statistic. 3) (b) Give a 95% confidence interval for the difference in mean ICU stay between 3) (c) Is there evidence that the population variances of the ICU stay differ betweern the treatment and control groups. the treatment and control groups? Use α-005 ICU stay of the treatment and control groups. (d) Give a 95% confidence interval for the ratio of the population variances for 0,1 0.05 0.025 0.010.005 ta (60) 1.296 1.671 2.000 2.390 2.660 ta(118) 1.289 1.658 1.980 2.358 2.618 ta(120) 1.289 1.658 1.980 2.358 2.617 Fo(60,60) 1.395 1.534 1.667 1.836 1.962 Fa (59,59) 1.399 1.540 1.674 1.846 1.974 F(120,2) 9.483 19.487 39.490 99.491 199.491 Recall that F-o/a(m -1,n-1) 1/Foan-1, m -1)

Answer 1