7. True or False. Circle one.

T F a. If we accept Ho, then we say Ho is true.

T F b. a and are inversely related but not in a quantitative sense.

T F c. Decreasing the sample size of a study will help to reduce

alpha but not beta error.

T F d. The odds-ratio and relative risk are approximately equal

if the outcome in question is relatively common.

T F e. The null hypothesis of the chi-square test for

independence may be written as Ho:p1=p2, wherep

represents the proportion of individuals with the outcome

of interest in each population.

T F f. When we match-pair the subjects in our study, we have

less degrees of freedom than if we had sampled

independently.

T F g. If we set a= 0, we would never make a b-error.

T F h. If the significance level of a study is increased, the

power would decrease.

T F i. If the sample size of a study is decreased, the power

would decrease.

T F j. If the sample size of a study is not changed, but we

choose to increase the power, then the significance level

must increase.

T F k. The margin of error of a confidence interval is a function

of the sample size used in the study.

T F l.The confidence level of an interval estimate is a measure

of its precision.