JOMC 201 � Final Exam Practice Questions

Fall 2002

Rules: Open book, open notes, open calculator. No laptops. No passing books, notes, or calculators from person to person.

1.      Two coders are classifying news sources as people or documents. Their table of agreement is:

����������� Report the reliability coefficient as:

����������������������� A: percent agreement

����������������������� B: Scott�s pi

2.      The following output shows the results of a t-test comparing the means of papers in the Knight Foundation sample with those in the Robert Putnam sample on readability. The means are Flesch scores.

����������� Explain, simply and concisely, any difference between the two groups and its statistical significance.

3. Write a short paragraph describing the interesting result, if any, in the following regression output. The dependent variable is the percent who believe the newspaper they read most often. The independent variables are the ratio of minorities on the newspaper to minorities in the community, circulation in 2000, and readability (Flesch formula).

4. A friend of yours is an amateur researcher, and she draws you aside to say, �I understand that you are taking JOMC 201. You therefore should know something about methodology.� She relates her current research proposal: to test the hypothesis that new reporters write with high Flesch scores but that their readability diminishes the longer they are on the job. She plans a content analysis of current newspapers. �If I can show that reporters who are recently hired have high Flesch scores and those who have been there a long while are low, I will have proved my hypothesis.� She solicits your frank comment. Answer her. Since she is your friend, you will, of course, have a constructive solution for every criticism you make.

5. What is a �robust� statistical procedure?

6. Name one innovation by John Tukey.

7. Describe a situation in which you would use a stratified sample and tell how you would obtain it.

8. Describe three ways in which an exit poll has an advantage over a pre-election poll.