# SPSS ANOVA Output

ANOVA
SAT Verbal score
Sum of Squares
Between
Groups
Within
Groups
Total

df

Mean Square

F

Sig.

702488.363

6

117081.394

17.541

.000

15425280.619

2311

6674.721

16127768.982

2317

Post-hoc Tests
Multiple Comparisons
Dependent Variable: SAT Verbal score
Tukey HSD
Mean
Std. Error
Difference
(I-J)
(I)
(J) ETHNICIT
ETHNICIT
African
Asian
-12.47
6.544
American
Hispanic
.66
6.415
White
-36.13
5.455
Asian
African
12.47
6.544
American
Hispanic
13.13
6.139
White
-23.66
5.128
Hispanic
African
-.66
6.415
American
Asian
-13.13
6.139
White
-36.79
4.962
White
African
36.13
5.455
American
Asian
23.66
5.128
Hispanic
36.79
4.962
* The mean difference is significant at the .05 level.

Sig.

95%
Confidence
Interval
Lower Bound

.226
1.000
.000
.226

-29.29
-15.84
-50.16
-4.36

Upper
Bound
4.36
17.15
-22.11
29.29

.141
.000
1.000

-2.66
-36.85
-17.15

28.91
-10.48
15.84

.141
.000
.000

-28.91
-49.55
22.11

2.66
-24.03
50.16

.000
.000

10.48
24.03

36.85
49.55

1. Interpret the results
The ANOVA summary table reveals that the between-group mean square (the variation
explained by the model) is 117,081.394 (702,488.363 / 6), and the within-group mean
square (the variation unexplained) is 6,674.721 (15,425,280.619 / 2,311). The F-ratio is
17.541 (117,081.394 / 6,674.721), and the p-value < .05 indicates that the difference in
the mean scores of the four ethnic groups is statistically significant. The Post-hoc tests

tell that the mean SAT score of White students is significantly higher than that of Black,
Asian, or Hispanic students.
2. Calculate the effect size
R2 = MSM / MST
= 702,488.363 / 16,127,768.982
= .04

R = .2 (The effect size is fairly small.)

Reference:
Field, A. (2005). Discovering Statistics Using SPSS for Windows. Sage Publications
Ltd., London.