Statistics

Question 1

Correltation

Answer 1

Measures the relative strength of the linear relationship between two variables -1 to +1

Question 2

Cohen’s r

Answer 2

Effect size

.1 = small
.3 = medium
.5 = large

Average in psych = .21

Question 3

Linear regression

Answer 3

Need an independent and dependent variable

Question 4

Slope

Answer 4

Rise over run, degree of change between X and Y (direction and steepness)

Question 5

Y Intercept

Answer 5

The value of Y when X = 0

Question 6

Regression equation

Answer 6

E (yi / xi ) = Beta 0 (Intercept) + Beta I (Slope at a given point)

SAME AS Y=MX+B

Question 7

Statistical significance

Answer 7

The probability of getting the observed value of the test statistic, or a value with even greater significance against the null hypothesis, if H0 is true

p < 0.05 to be statistically significant

Question 8

Criticisms of Statistical Significance

Answer 8

1. p < 0.05 is and arbitrary cutoff
2. With a large enough sample size basically anything could be statistically significant
3. Only statements in relation to the null hypothesis can be made

Question 9

Clinical significance

Answer 9

Are the results clinically relevant?

1. The client’s post-intervention score is 2 SD away from the mean of the DYSFUNCTIONAL population
2. The clients post-intervention score is 2 SD within the mean of the functional population
3. The clients post-intervention score is closer to the mean of the functional population than the mean of the dysfunctional population

Question 10

4 categories of clinically significant change

Answer 10

Recovered (meet criteria for clinically significant change)

Improved (have significantly changed but not enough to be considered a full recovery)

Unchanged (No change over time)

Deteriorated (significant worsening of symptoms)

Question 11

Number Needed to Treat (NNT)

Answer 11

The average number of patients who need to be treated to prevent one additional bad outcome (# of patients that need to be treated for one to benefit compared with a control)

Question 12

A researcher correlates two variables: Dietary restraint
and binge eating. The correlation = .30, and the p-value
= .10. Which of the following is true?

A. The two variables are significantly related with a small
effect size
B. The two variables are significantly related with a
medium effect size
C. The two variables are significantly related with a large
effect size
D. For every 1 unit increase in dietary restraint, there is a
corresponding 0.3 unit increase in binge eating
E. The two variables are not significantly related

Answer 12

E

Question 13

A researcher conducts a linear regression
analysis. The slope was 2.0, and the intercept was
1.0. If X = 5, what is the expected value of Y?
A. 6
B. 7
C. 8
D. 9
E. 10
F. 11

Answer 13

F

Question 14

A researcher tests the effectiveness of interpersonal
therapy when treating anxiety disorders using a
randomized control design. How can she determine if
her results are clinically significant?

A. If the p-value is < .05
B. If the NNT is 100 or greater
C. If most participants no longer meet criteria for an
anxiety disorder at post-test
D. All of the above support clinical significance
E. None of the above support clinical significance

Answer 14

C

Question 15

Mediation

Answer 15

X —> M —> Y

Question 16

Moderation

Answer 16

Relationship between X any Y depends on a third variable (the moderator)