1. You design a clinical trial with the null hypothesis that surfactant does not improve 28 day mortality in ARDS. You set your p value at 0.05. Your study finds that surfactant has no effect on 28 day mortality in ARDS, even though in reality, there is an effect. What type of error have you committed?a) Type 1 error b) Type 2 error c) Alpha error d) Type 3 error 2. You calculate a 95% confidence interval of 5 to 10. True or false: There is a 95% probability the population mean is between 5 and 10.a) True b) False 3. Which of the following depend on the prevalence of the disease in the population?a) Sensitivity b) Specificity c) Negative predictive value d) Odds ratio 4. What is the odds ratio for smoking and lung cancer?a) 19 b) 0.33 c) 10 d) Cannot calculate with information given 1. B. You accepted the null hypothesis when you should have rejected it or in other words, your study failed to find a difference when one exists. This is an example of a type 2 or beta, error. An Alpha, or type 1 error would have occurred if your study found surfactant made a difference when in reality, it does not. It corresponds to your predetermined p value for significance (ie typically 0.05 meaning a 5% chance of committing a type 1 error). 2. B. False. The 95% confidence interval states that if repeated samples are taken and confidence intervals generated, 95% of those intervals will contain the population mean. This is a subtle but important difference. 3. C The negative predictive value (and the positive predictive value) depend on the prevalence of the disease in the population whereas sensitivity, specificity, and the odds ratio do not. 4. A The odds ratio is the ratio of the odds of developing the outcome in the exposed vs. nonexposed. This is (10/10)/(4/76) or AD/BC =760/40=19. |