Skip to main content

Table 2 Results of the weighted ANOVAs for the influence of categorical variables on the effect sizes

From: Perfectionism and binge eating association: a systematic review and meta-analysis

Moderator variable

k

r+

95% CI

ANOVA results

rl

ru

Age

    

F(2,28) = 4.48, p = .02

R2 = .20

QE(28) = 160.40, p < .001

Adolescents (11–17 years)

4

.14

.004

.28

Young adults (18–25 years)

20

.22

.15

.27

Adults (> 25 years)

7

.03

− .08

.14

Sample type

    

F(2,28) = 14.90, p = .001

R2 = .56

QE(28) = 103.84, p < .001

Clinical

10

.04

− .05

.12

Non-clinical undergraduates

16

.22

.16

.28

Non-clinical others

7

.24

.15

.33

Design

    

F(1, 29) = 9.94, p = .004

R2 = .30

QE(29) = 132.19, p < .001

Cross-sectional

26

.21

.17

.26

Longitudinal

7

.03

− .07

.12

Perfectionism scale

    

F(5, 25) = 3.12, p = .03

R2 = .32

QE(25) = 109.25, p < .001

EDI versionsa

8

.03

− .06

.12

FMPS versionsb

11

.26

.19

.33

HMPS versionsc

5

.21

.10

.32

M-CUP

1

.28

− .02

.53

PSPS

1

.22

− .02

.42

Measures combinedd

7

.15

.06

.24

Sample size

    

F(2, 28) = 1.56, p = .23

R2 = .04

QE(28) = 171.72, p < .001

80–200

10

.09

− .004

.19

201–400

13

.19

.11

.26

 > 400

8

.20

.11

.30

Geographic location

    

F(5, 25) = 1.56, p = .21

R2 = .10

QE(25) = 136.35, p < .001

USA

15

.16

.09

.23

Canada

10

.17

.08

.25

Belgium

3

.28

.12

.42

Australia

1

− .14

− .41

.16

Canada and UK

1

.29

.03

.51

Binge eating scale

    

F(11, 19) = 8.99, p < .001

R2 = .90

QE(29) = 28.84, p = .07

EDE versionse

11

.02

− .03

.07

EDI versionsf

7

.30

.24

.35

EDDS

3

.23

.14

.32

BES

2

.26

.15

.37

MEBS

2

.30

.19

.40

Measures combinedg

2

.28

.17

.38

Binge Scale

1

.15

.008

.29

Binge = 1/No Binge = 0

1

.13

− .04

.29

BMS

1

− .12

− .32

.09

EPSI

1

.28

.05

.48

ESS

1

.13

− .02

.27

Frequency of binge eating

1

.30

.21

.47

  1. K, number of studies; r+, mean effect size; rl and ru, 95% lower and upper confidence limits around r+; R2, proportion of variance accounted for by the moderator variable; F, statistic to test the statistical significance of the moderator proposed by Knapp and Hartung; QE, Statistic for testing the model misspecification; M-CUP, Measure of Constructs Underlying Perfectionism; PSPS, Perfectionistic Self-Presentation Scale; EDDS, Eating Disorder Diagnostic Scale; BES, Binge Eating Scale; MEBS, Minnesota Eating Behavior Survey; BMS, Bulimic Modeling Scale; EPSI, Eating Pathology Symptom Inventory; ESS, Eating Self-Efficacy Scale
  2. aStudies that use the Eating Disorder Inventory or the Eating Disorder Inventory-2 to assess perfectionism
  3. bStudies that use the Frost Multidimensional Perfectionism Scale or its brief version
  4. cStudies that use the Hewitt Multidimensional Perfectionism Scale or its brief version
  5. dStudies where perfectionism score is calculated by combining different measures
  6. eStudies that use any of the following tools: the Eating Disorder Examination Questionnaire, the Eating Disorders Examination, the Child Eating Disorder Examination-Questionnaire, or the Eating Disorder Examination Questionnaire 6.0
  7. fStudies that use the Eating Disorder Inventory or the Eating Disorder Inventory-2 to assess binge eating
  8. gStudies where binge eating score is calculated by combining different measures
Back to article page

Journal of Eating Disorders

ISSN: 2050-2974