Which of the following results gives the reader the most information concerning statistical significance, sample size, and strength of association?
|A||A relative risk of 2.5 with a 95 percent confidence interval of 2.0 to 3.1|
|B||A p value of 0.0004 and an internal of 0.05|
|C||A relative risk of 5.0 with a 95 percent confidence interval of 0.1 to 9.8|
|D||A p value of <0.05 and a relative risk of 2.5|
Ref: High Yield Biostatistics, Pg: 23
a. A confidence interval is a range of possible values for the attribute that is being assessed and is designed in such a way that this range has a certain probability of including the true value. A 95 percent confidence interval, equivalent to a = 0.05, will include the true value 95 percent of the time.
b. The confidence interval reflects not only the relevant range of possibilities, but reflects sample size and statistical significance as well.
c. A larger range indicates a smaller sample size. A confidence interval for a relative risk that includes 1.0 would indicate a nonsignificant statistical association (there is a 95 percent chance that the relative risks would be the same or 1.0).
d. The p value only indicates the probability that the association occurs by chance. Statistical power only gives the probability of avoiding missing a true difference between exposure groups (type II, or B, error).