Scribd
Upload
44 views

Ancova Example: Ancova Output From GLM/Univariate in Spss

An ANCOVA was conducted to compare post-test scores between two groups (group 0 and 1) while controlling for pre-test scores. The ANCOVA found a significant effect of group on post-test scores after controlling for pre-test scores. Specifically, the estimated marginal mean post-test score was 23.474 for group 0 and 64.526 for group 1, with a significant difference of 41.053 between the two groups.

Uploaded by

fatemeh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
44 views

Ancova Example: Ancova Output From GLM/Univariate in Spss

An ANCOVA was conducted to compare post-test scores between two groups (group 0 and 1) while controlling for pre-test scores. The ANCOVA found a significant effect of group on post-test scores after controlling for pre-test scores. Specifically, the estimated marginal mean post-test score was 23.474 for group 0 and 64.526 for group 1, with a significant difference of 41.053 between the two groups.

Uploaded by

fatemeh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 2

Ancova Example

Tests of Between-Subjects Effects


 Ancova output from Dependent Variable: post

GLM/Univariate in
Type III Sum
Source of Squares df Mean Square F Sig.
Corrected Model 4842.105a 2 2421.053 12.123 .005

SPSS Intercept
pre
3112.762
842.105
1
1
3112.762
842.105
15.587
4.217
.006
.079
group 4199.482 1 4199.482 21.029 .003
Error 1397.895 7 199.699
Total 25600.000 10
Corrected Total 6240.000 9
a. R Squared = .776 (Adjusted R Squared = .712)

 Regression output ANOVAc

Sum of
 4842.105-642.623 Model
1 Regression
Squares
642.623
df
1
Mean Square
642.623
F
.918
Sig.
.366a

=4199.482 Residual
Total
5597.377
6240.000
8
9
699.672

2 Regression 4842.105 2 2421.053 12.123 .005b


Residual 1397.895 7 199.699
Total 6240.000 9
a. Predictors: (Constant), pre
b. Predictors: (Constant), pre, group
c. Dependent Variable: post
Ancova Example
Tests of Between-Subjects Effects
 Compared to Dependent Variable: post

standard regression Source


Corrected Model
Type III Sum
of Squares
4842.105a
df Mean Square F Sig.
Partial Eta
Squared

output
2 2421.053 12.123 .005 .776
Intercept 3112.762 1 3112.762 15.587 .006 .690
pre 842.105 1 842.105 4.217 .079 .376

 Note how partial eta- group


Error
4199.482
1397.895
1
7
4199.482
199.699
21.029 .003 .750

squared is just the Total


Corrected Total
25600.000
6240.000
10
9

squared partial a. R Squared = .776 (Adjusted R Squared = .712)

Coefficientsa

correlations from Unstandardized Standardized

regression
Coefficients Coefficients Correlations
Model B Std. Error Beta t Sig. Zero-order Partial
1 (Constant) 10.316 9.184 1.123 .298

 The coefficient for


pre .526 .256 .368 2.054 .079 .321 .613
group 41.053 8.952 .822 4.586 .003 .801 .866
a. Dependent Variable: post

the dummy-coded
variable is the Dependent Variable: post
group

difference between group Mean Std. Error


95% Confidence Interval
Lower Bound Upper Bound
marginal means in .00
1.00
23.474a
64.526a
6.325
6.325
8.517
49.570
38.430
79.483
the Ancova a. Covariates appearing in the model are evaluated at the
following values: pre = 25.0000.

You might also like