Question

Which one of the following is a definition of a sampling distribution? A. The frequency distribution of a statistic based on repeated sampling. B. The frequency distribution of a parameter based on the population. C. The distribution of a random sample based on a statistic. D. The distribution of a sample based on the characteristics of the population.

A drugmaker is testing the effectiveness of a new cancer drug. Similar to many new drugs, this chemotherapy drug has significant side effects, making it essential that the drug maker ensures that the drug is significantly more effective than the current treatment. The clinical trial is using the existing treatment as the control and will ultimately statistically test the outcomes using a dependent samples t-test. In order to be as certain as possible not to approve this drug if it isn’t highly likely, it is better, the drug maker could:

A. Lower alpha from .05 to .01 to lower Type I error potential

B. Increase alpha from .05 to .1 to lower Type II error potential

C. Lower beta from .05 to .01 to lower Type I error potential

D. Increase beta from .05 to .1 to lower Type II error potential

Which of the following is a definition of criterion-related validity? A. The degree to which the items represent the construct of interest fully. B. The degree to which a measure of a construct predicts scores on another measure of a theoretically relevant construct. C. The degree to which items on a measure appear to measure what is expected by the test taker. D. The degree to which your measure does not correlate with measures of theoretically unrelated constructs.

Answer 1

i) A. The frequency distribution of a statistic based on repeated sampling.

ii) Here we want to lower the type I error, thus the correct option is: A. Lower alpha from .05 to .01 to lower Type I error potential

iii) B. The degree to which a measure of a construct predicts scores on another measure of a theoretically relevant construct.