36-201 INTRODUCTION TO STATISTICAL REASONING

Assignment #7  

                Due:  In Lab, Thurs Mar 4 and Mar 5.

Be sure to write your name and SECTION on your paper.  

STAPLE, DO NOT PAPERCLIP OR FOLD/TEAR, YOUR PAGES TOGETHER.

1. Use a WWW Browser (e.g. Netscape) to look at the following two
science reports from the Associated Press.

	http://www.stat.cmu.edu/~brian/201/week07/ap/bone-study.html
	http://www.stat.cmu.edu/~brian/201/week07/ap/growth-study.html

(I found these stories on the WWW by clicking on the "Health and 
Science" link in the "AP Wire" section near the bottom of the web page 
http://www.postgazette.com/news/default.asp, but the postings change 
regularly.)

Answer the following questions.  A sentence or two at most is needed for
each answer.

  (a) One of these studies is clearly an observational study; and the
  other might have been an experiment or an observational study.
  
      (i) Which is which?  Why?

      (ii) Explain why the one that is clearly an observational study 
      probably could never have been done as an experiment.

  (b) Neither study employed a control group, or a second treatment
  group for comparison.  

      (i) Explain why this makes it harder to decide that the treatment
      used really caused the changes that the experimenters saw, than if
      a randomized comparative study had been used. [ a correct but 
      rather general answer is alright here ]

      (ii) Explain why it might have been unethical or impractical to
      have a control group in these two studies (your answer may be 
      different for each study). [ please give reasons that are specific
      to each study; part of your answer might be related to (a)(ii) above ]

      (iii) By carefully reading the articles, explain why, in these
      cases at least, the conclusions made by the researchers may be
      alright anyway. [ a general answer will *not* be enough here.
      Please refer to specific information in the article in your
      answer ]

2. Moore, p. 31, problems 1.17-1.20 In each case, simply identify which
boldface number is the value of a *parameter* and which is the value
of a *statistic*.

3. Moore, pp. 31-32 problem 1.22 (compare with Fig 1-4, p. 29) AND ALSO
problem 1.23

4. Moore, p. 52, problem 1.44

5. Moore, p. 55-57, problems 1.52, 1.53.