Session 11: Random Error

Question 1

Alpha

Answer 1

The probability of rejecting the null when the null is true (Type I error)

Question 2

Beta

Answer 2

The probability of failing to reject the null when the alternative is true (Type II error)

Question 3

Confidence Limit Ratio

Answer 3

The ratio of the upper to the lower limit of a confidence interval. A measure of the precision represented by a confidence interval.

Question 4

Confidence Interval

Answer 4

A range of plausible values for a population parameter that depends on the point estimate, its variability, and the sample size. The definition of “plausible” depends on the confidence interval the researcher chooses - a 99%, 95%, or 90% confidence interval, for example.

Question 5

Point Estimate

Answer 5

The association between the exposure and disease estimated in the study

Question 6

Power

Answer 6

The probability of rejecting the null when the alternative is true (1-beta)

Question 7

P-value

Answer 7

The probability of obtaining a test statistic computed from the data (t-test, chi-square) as or more extreme than the observed value, if the null were true.

Question 8

Significance Level

Answer 8

The criterion chosen for rejected the null hypotheses, corresponding to the maximum type I error rate.