MetaAnalysis Activity
NAME_____________________________
A researcher wished to complete a metaanalysis reviewing studies that explored the relationship between the amount of selfdisclosure and the degree of liking reported between recent acquaintances. In addition to effect sizes, researchers had a group of raters assess study quality on a scale ranging from 0 (lowest possible quality) to 10 (highest possible quality). Other moderator variables include the date of publication of the study, the number of independent variables in the study, whether the relational partners were samesex or oppositesex partners, and whether the samples were college students or adult populations. The researcher has decided to use r as the measure of effect size.
Study and Date
Sample size (n)
Direction of Relationship
R
Number of Variables in Studies
Quality Ratings
Sex composition of the relationship pairs
1 = same sex
2=opposite sex
Type of Sample:
1 = college
2 = adult
Fisher’s Zr
1975
100
+
.33
3
8
1
2
1980
120

.15
2
6
2
1
1990
290
+
.28
5
8
2
1
1990
80
+
.40
3
9
2
2
1992
150
+
.22
4
4
1
1
1997
140

.20
4
5
2
1
1999
170
+
.30
5
7
1
2
2002
300
+
.33
2
6
1
2
2002
175
+
.25
2
8
1
1
· The researcher wanted to identify relationships that are at least a “slight” relationship. State the hypothesis for this type of relationship:
· In the column on the far right of the table, insert the Fisher’s Z transformed correlation for each of the studies.
· What would be the mean effect size for these data?___________________________
what would be the mean effect size if transformed back into raw scores?
HINT: To transform a Z_{Fisher} correlation back to an untransformed correlation, the following formula is used: .
· What is the result of the diffuse comparison test?_____________________________
Focused Comparison Tests:
a. For the number of independent or predictor variables moderator variable:
1. What is the Z value for the focused comparison test of the “number of variables” moderator variable?_______________________________________________________
HINT: The Linear Contrast coefficients for nine studies are:
4 3 2 1 0 1 2 3 4
2. Is this effect statistically significant with an alpha risk of .05?______________
b. For the sex composition of the relationship pairs moderator variable:
1. What are the orthogonal polynomials for each of the studies in the following arrangement?____________________________________________________________
Study Date 
1975 
1980 
1990 
1990 
1992 
1997 
1999 
2002 
2002 
Sex Composition 
1 
2 
2 
2 
1 
2 
1 
1 
1 
Polynomial with Ties 









2. Is this effect statistically significant with an alpha risk of .05?______________
c. For the type of sample moderator variable:
1. What are the orthogonal polynomials for each of the studies in the following arrangement?___________________________________________________________
Study Date 
1975 
1980 
1990 
1990 
1992 
1997 
1999 
2002 
2002 
Type of Sample 
2 
1 
1 
2 
1 
1 
2 
2 
1 
Polynomial with Ties 









2. Is this effect statistically significant with an alpha risk of .05?_____________
d. There are three continuous variables in the study: the study date for publication, the number of variables, and experts’ quality ratings.
·
What are the intercorrelations for the continuous
moderator variables in the study? ____
_______________________________________________________________________
_______________________________________________________________________
· What is the correlation between each of these moderator variables and the effect size observed?
_______________________________________________________________________
· What interpretations must be made about the influences of moderator variables?
________________________________________________________________________