Question

An Internet report from January 2005 is titled, "Study: Anti-Seizure Drug Reduces Drinking in Bipolar Alcoholics." A chi-square test was carried out based on results mentioned in the report, which explains that takers of the drug (Valproate) and placebo were questioned after 6 months to see if they had engaged in heavy drinking (five or more drinks daily for men, four or more daily for women).

	Heavy Drinking	No Heavy Drinking	Total
Valproate	14	18	32
Placebo	15	7	22
Total	29	25	54

1. The P-value was found to be 0.077. Which one of these characterizes the size of the chi-square statistic?

much less than 3.84

somewhat less than 3.84    

somewhat greater than 3.84

much greater than 3.84

2. In order to produce more convincing evidence of the drug's benefits, what would be the most obvious change to make if a second study could be carried out?

use a smaller sample size

analyze the results separately for women and men    

use a larger sample size

use a more representative sample

3. What would the P-value have been if a one-sided z test had been carried out to see if population proportion engaging in heavy drinking is less for Valproate-takers than for placebo-takers?

4. The chi-square statistic was 3.13. What would the z statistic have been if a z test had been carried out instead of a chi-square test?

Answer 1

1)somewhat less than 3.84   

2) use a larger sample size

3) p value =0.077/2 =0.0385

4) z statistic =sqrt(3.13)=1.77