Answers to the Review questions from Homework 1

Review Exercise Answers (Chapter 2)

2. (a) False. There are many more Berettas out there to steal. So you need to calculate the rate (or percentage in this case) stolen. 300/47,598 is about 6 stolen per 1,000 cars.

Corvettes are stolen at a rate of 134/18,938 or about 7 per 1000. If anything, thieves prefer Corvettes.

You might think about the problem this way: if there were as many Corvettes on the street as Berettas there would be many more than 134 Corvettes stolen.

(b) False. Again, you need to think about this as rates (or percentages) that eliminate the differences in production numbers. The rates of 7 and 9 given suggest that Z cars are more likely to be stolen. If it helps, figure how many Z cars would have been stolen if there were 133,000 produced in a year.

4. (a) They controlled for age and sex as possible confounding effects (see page 13).

(b) Wrong. Ex-smokers may be unusual people -- that is many people give up smoking because they got sick (e.g. a guy has a heart attack and his doctor says QUIT SMOKING). So it isn't necessarily the case that people who quit get sicker.

9. (a) False. The experiments found no difference in the death rate from color cancer and found increases in the death rate from lung cancer.

(b) True. There is self-selection going on here -- that is, people who chose to eat fresh fruits and veggies are probably different from people who don't eat such things regularly and these differences, not diet, may be generating the outcomes.

(c) False. Experimentation should eliminate confounding (as this is why people experiment) but in this example, the controls could be doing something different in their diets.

11. (a) The treatment group consists of those who finished boot camp. The control group consists of all the other prisoners -- those who did not volunteer and those who volunteered but didn't finish the program.

(b) This is an observational study. Who is in charge here? The prisoners...they decided (voluntarily) who would be in boot camp and who would not.

(c) The spokesperson said it works...in that sense...it is true...but from what we know, we might not place too much faith in the program. Best answer is false.

From Chapter 3 Review

4 (a) looks like about 25%

(c) 140-150mm, the rectangle looks bigger

(d) 135-140

7. (i) natural causes -- old people tend to die from these things

(ii) trauma -- younger people tend to be somewhat more reckless and have a better chance of dying in an accident, murder, suicide...

8. (a) True, if you adjust for the interval size, you'll see

(c) False. Examine the class intervals. While the percentages just about total 100%, the heights are draw incorrectly and are misleading.

Chapter 4 Review

1 (a) Average = 50 (sum of 41,48,50,50,54,57 divided by 6)

SD = square root of ( sum of the squared deviations / 6)

4. Income -- average is bigger than median. (see p.64-65)

Education -- average is smaller than median (p.39)

6 (a) 60, 50 , 40

(b) (i) median is bigger than average

(ii) median is about equal to the average

(iii) median is less than the average

7. (a) Average weight of men 66x2.2 = 145lbs

SD about 20 lbs

Average weight of women 55x2.2= 121lbs. SD about 20 lbs

(b) 68% 57kg-75kg is just about -1 and +1 SD see table A-105