




Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Material Type: Exam; Class: BASIC APPLIED STATISTICS; Subject: Statistics; University: University of Pittsburgh; Term: Spring 2008;
Typology: Exams
1 / 8
This page cannot be seen from the preview
Don't miss anything!





Statistics 200 Spring 2008 (Pfenning)
This is a closed book exam worth 150 points. You are allowed to use a calculator and a two-sided sheet of notes. There are 8 problems, with point values as shown. If you want to receive partial credit for wrong answers, show your work. Don’t spend too much time on any one problem.
(a) explanatory variable (b) explanatory variable type: (i) quantitative (ii) categorical (iii) not clear (c) response variable (d) What comparison is being made? i. patients who do or do not take CoQ ii. patients before and after taking CoQ (e) Which one of the following additional pieces of information would be most helpful in deciding whether CoQ10 is really beneficial for migraine sufferers? i. Was there a control group taking a placebo? ii. Were the patients randomly chosen to participate in the study? iii. How did researchers define a migraine? iv. Were different dosages of CoQ10 tested? v. How realistic was the setting?
(a) Which design is an observational study? (i) A (ii) B (iii) both (iv) neither (b) Which design is more vulnerable to confounding variables? (i) A (ii) B (iii) both the same (iv) neither
(c) What is the most worrisome flaw in Design A?
(d) Which of the above in this problem are closed questions? (a), (b), or (c) (Circle any that are closed.)
4000BC 150AD
120
130
140
Boxplots of 4000BC and 150AD (means are indicated by solid circles)
Fatal No Fatal Heart Attack Heart Attack Total Combination Drugs 29 1171 1200 Other Drugs 21 1779 1800 Total 50 2950 3000
(a) Find the probability of a fatal heart attack for those taking the combination drugs. (b) Find the overall probability of a fatal heart attack. (c) For those 3000 women, fatal heart attacks were (i) more likely if combination drugs were taken (ii) less likely if combination drugs were taken (iii) neither (combina- tion drugs made no difference) (d) If fatal heart attacks and type of drug treatment were independent of each other, how many of the 1800 women who took other drugs would have suffered a fatal heart attack? (e) Before concluding that the combination drugs are responsible for higher rates of heart attacks, which one of the following is most important? i. Make sure the researchers were blind. ii. Make sure there wasn’t a tendency for women who were more at risk of fatal heart attacks to be prescribed the combination drugs. iii. Make sure the women studied were all the same ages. iv. Make sure the women studied were a variety of ages. (f) Question (e) concerns itself with (i) data production (ii) displaying and summarizing (iii) probability (iv) statistical inference
(a) What would be an appropriate display? (i) bar graph (ii) histogram (iii)side-by-side boxplots (iv) scatterplot (b) Which of these would provide the best summary? (i) compare percentages (ii) compare means and standard deviations (iii) compare Five Number Summaries (iv) report the correlation
(a) What would be an appropriate display? (i) pie chart (ii) histogram (iii)side-by-side boxplots (iv) scatterplot
(a) What is the explanatory variable? (b) The regression equation predicts that for each additional hundred square feet, price increases by thousand dollars. (c) Which of the following is the best guess for r? (i) -.83 (ii) -.43 (iii) -.13 (iv) .13 (v) .43 (vi). (d) Predict the price of a home with size 40 hundred square feet. (e) If a home with size 40 hundred square feet actually cost 523 thousand dollars, then the residual is (be sure to include sign + or -). (f) What is the typical size of prediction errors made by the regression line? (g) If the home with size 40 and cost 523 were omitted from the regression, the value of r would (i) increase (ii) decrease (iii) stay the same (h) According to the output, there was an outlier home which had size 30 and cost
The regression equation is price = - 141 + 15.9 size Predictor Coef SE Coef T P Constant -140.9 123.3 -1.14 0. size 15.894 4.057 3.92 0. S = 59.62 R-Sq = 68.7% R-Sq(adj) = 64.2% Unusual Observations Obs size price Fit SE Fit Residual St Resid 7 30.0 457.0 335.9 19.9 121.1 2.15R R denotes an observation with a large standardized residual
20 30 40
500
400
300
200 size
price