I need help understanding the results of a three way anova. Tests of Between-Sub
ID: 3356003 • Letter: I
Question
I need help understanding the results of a three way anova.
Tests of Between-Subjects Effects
Dependent Variable: CURRENT SALARY
Source
Type III Sum of Squares
df
Mean Square
F
Sig.
Partial Eta Squared
Corrected Model
16393637894.665a
19
862823047.088
69.050
.000
.743
Intercept
20988570641.745
1
20988570641.745
1679.677
.000
.787
jobcat
5150755792.779
6
858459298.797
68.701
.000
.476
sexrace
308992271.787
2
154496135.893
12.364
.000
.052
minority
.000
0
.
.
.
.000
jobcat * sexrace
100500284.863
5
20100056.973
1.609
.156
.017
jobcat * minority
.000
0
.
.
.
.000
sexrace * minority
.000
0
.
.
.
.000
jobcat * sexrace * minority
.000
0
.
.
.
.000
Error
5673001375.149
454
12495597.743
Total
111914789908.000
474
Corrected Total
22066639269.814
473
a. R Squared = .743 (Adjusted R Squared = .732)
Tests of Between-Subjects Effects
Dependent Variable: CURRENT SALARY
Source
Type III Sum of Squares
df
Mean Square
F
Sig.
Partial Eta Squared
Corrected Model
16393637894.665a
19
862823047.088
69.050
.000
.743
Intercept
20988570641.745
1
20988570641.745
1679.677
.000
.787
jobcat
5150755792.779
6
858459298.797
68.701
.000
.476
sexrace
308992271.787
2
154496135.893
12.364
.000
.052
minority
.000
0
.
.
.
.000
jobcat * sexrace
100500284.863
5
20100056.973
1.609
.156
.017
jobcat * minority
.000
0
.
.
.
.000
sexrace * minority
.000
0
.
.
.
.000
jobcat * sexrace * minority
.000
0
.
.
.
.000
Error
5673001375.149
454
12495597.743
Total
111914789908.000
474
Corrected Total
22066639269.814
473
a. R Squared = .743 (Adjusted R Squared = .732)
Explanation / Answer
The result given above is quite simple to understood. There are one dependent variable "Current Salary" and three factors, "jobcat" "sexrace" "minority". Now, to analyse if these factors affects the cuurent salary, we first have to check their interaction effects. If the interactions are significant,we cannot study the main effects.
Now, we see the interaction factors "jobcat*minority" "minority*sexrace" "jobcat*minority*sexrace" has 0 sum of square and there p-value is not defined (look for the "Sig" column"). Now by the thumbrule that p-value is <level of significance indicates the significance of a variable, we can say all the interaction effects are insignificant. So, we can study the main effects,. "jobcat*sexrace" has a pvalue of .156>.05 , usual level of significance, making the interaction insignificant.
Now the p-value ofjobcat and sexrace is 0<0.05. Thus they are significant enough to affect the the current salary.
However, the minority status has again the null p-value, making it insignificant.
Related Questions
drjack9650@gmail.com
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.