Deck 15: Statistical Methods of Data Evaluation

Question 1

What is the goal of significance testing?

A) To be able to reject the null hypothesis
B) To be able to accept the null hypothesis
C) To provide an objective, agreed upon criterion to decide whether differences are due to chance
D) To provide an objective, agreed upon criterion to decide whether differences are due to the intervention

Answer 1

To provide an objective, agreed upon criterion to decide whether differences are due to chance

Question 2

Use of a pretest as well as a posttest, rather than a posttest by itself, is valuable because

A) One can evaluate and statistically adjust for any pretest sensitization effect
B) One can use a directional test to evaluate change, i.e., did patients improve
C) Reduces bias due to multiple statistical tests
D) The error term is reduced in the statistical test and power is increased

Answer 2

The error term is reduced in the statistical test and power is increased

Question 3

How can power be augmented in an investigation?

A) Through the use of posttests
B) Through maintaining conventional levels of significance
C) Through the use of directional tests
D) Through controlling the side effects of treatment

Answer 3

Through the use of directional tests

Question 4

Statistical significance, magnitude of effect, and clinical significance:

A) Overlap greatly and any two allows one to infer the other one
B) Reflect different facets of the data and bear no necessary relation to each other
C) A large effect size is likely to mean both that the effect was statistically and clinical significant
D) a and b

Question 5

What is statistical significance?

A) A criterion used to determine the level of confidence of two groups differing at levels greater than chance
B) A criterion used to determine the level of confidence of the null hypothesis being false
C) The practical value or importance of the effect of an intervention
D) A method for expressing differences between groups in a common metric

Question 6

Intent-to-treat data analyses is designed

A) To preserve the random composition of groups
B) To keep the investigator from changing the hypotheses that will be studied
C) To evaluate the results with those who completed the pre, treatment, and post
D) Estimates data that the subject would have had from others in the same treatment condition but who completed treatment

Question 7

Which of the following is a valid method of determining the necessary sample size of an investigation?

A) Through the effect size, alpha, and power
B) Through the clinical significance of an investigation
C) Through conventions in his or her field of research
D) Through the use of significance tests

Question 8

When are multivariate analyses appropriate?

A) When utilizing a factorial design
B) When measures are conceptually related
C) When measures are uncorrelated
D) When interest is in individual measures

Question 9

What is alpha?

A) A method of expressing differences between groups in a common metric
B) The probability of rejecting the null hypothesis when that hypothesis is true
C) The probability of rejecting the null hypothesis when it is false
D) The probability of accepting the null hypothesis when it is false

Question 10

What is an advantage of meta-analysis?

A) Can examine variables that were not a focus of the original research
B) Can examine the relations between measures
C) Can compare the validity of different investigations
D) Can examine the clinical significance of investigations

Question 11

Which of the following is an important consideration specific to making multiple comparisons?

A) Controlling variability in the data
B) Controlling the reliability of the measures
C) Controlling the experimental-wise error rate
D) Controlling the per-comparison error rate

Question 12

A statistically significant effect at the p < .00001 level, means

A) One can reject the null hypothesis
B) One can reject the null hypothesis and the investigator's hypothesis is strongly supported
C) The result is statistically significant and the effect size is large
D) The finding is very significant

Question 13

What is power?

A) An objection to the use of significance testing in psychological research
B) The expression of differences between groups in a common metric
C) The probability of accepting the null hypothesis when it is false
D) The extent to which a difference can be detected when a difference exists

Question 14

A Bonferroni adjustment is designed

A) To allow one to detect effects that might be just at the borderline level e.g., p < .0 which might be important for theoretical reasons
B) To control for experiment-wise error rate
C) To increase power by using a more sensitive statistical test
D) a and b

Question 15

Which of the following is the most dependent on sample size?

A) Validity
B) Correlation coefficient
C) Alpha
D) Effect size