1.

Researchers are interested in whether living in close proximity to a large terrorist attack induces post-traumatic stress symptoms (without formally testing for Post-Traumatic Stress Disorder itself).

Researchers contacted 200 residents of Manhattan Island in the weeks following the September 11, 2001 terrorist attacks. 150 of these residents lived south of Canal street, judged to be "close proximity"; the remaining 50 lived north of Canal Street. The contacted residents gave informed consent and participated in a telephone questionnaire with several questions that assessed the severity of their post-traumatic stress symptoms. All residents who began the questionnaire completed it. Surveyed residents were then determined to have "severe" or "non-severe" post-traumatic stress symptoms using clinical cutoffs validated in other studies.

The results of the survey can be summarized in the following table:

	Severe post-traumatic symptoms	Non-severe post-traumatic symptoms
Lived south of Canal Street	136	14
Lived north of Canal Street	35	15

Question modeled from: Galea et al., "Psychological Sequelae of the September 11 Terrorist Attacks in New York City" - NEJM (2002)

1a.

Perform a chi-square test to assess whether proximity to the event (measured by living north/south of Canal Street) is associated with having severe post-traumatic symptoms.

State the null and alternative hypotheses.

Check any assumptions required for your proposed test. Explain briefly.

Calculate the test statistic and p-value.

State the degrees of freedom, if appropriate.

If you use software, state the command you used and/or include a portion of the output. If you used table, state that.

Interpret the p-value for a non-statistician in the context of the problem. [The p-value is a probability .... What is it the probability in THIS problem? Do NOT use the words "null hypothesis."] Do NOT simply state whether or not the result is statistically significant.

What (if anything) can we conclude about the proposed association?

1b.

Find an odds ratio* and its associated 95% confidence interval for this 2x2 table.

(*The odds ratio for severe Post-Traumatic Symptoms for those south of Canal Street vs. north of Canal Street.)

Interpret the odds ratio and its associated CI with language appropriate for a non-statistical journal.

If you use software, state the command you used and include a portion of the output. If you calculate the odds ratio by hand then show all your work.