1. The researcher is interested in whether teacher self-efficacy at time one is influenced by gender, education, and the interaction between gender and education.
Tests of Between-Subjects Effects
Dependent Variable: efficacyt1
Source Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared
Corrected Model 4.983a 5 .997 .745 .590 .013
Intercept 7946.400 1 7946.400 5940.201 .000 .955
Gender .070 1 A .070 .052 .819 .000
education 4.861 2 2.431 C. 1.817 .164 E. .013
gender * education 1.123 2 .561 .420 D. .658 .003
Error 375.903 281 B. 1.338
Total 14940.528 287
Corrected Total 380.885 286
a. R Squared = .013 (Adjusted R Squared = -.004)
• What are the null hypotheses?
• Interpret each of the letters in the table:
• A
• B.
• C.
• D.
• E. Interpret the partial eta-squared for education (even though the effect is not statistically significant)
2. We measured teacher self-efficacy across 3 time periods. Time was the within-subjects factor and education level (q0006recode) was the between-subjects factor.
The sphericity assumption was met. The null hypothesis for education level as a main effect was rejected at an alpha of .01. (THAT TABLE IS NOT INCLUDED)
Tests of Within-Subjects Effects
Measure: MEASURE_1
Source Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared
time Sphericity Assumed 32.201 2 16.100 51.366 .000 .165
Greenhouse-Geisser 32.201 1.978 16.279 51.366 .000 .165
Huynh-Feldt 32.201 2.000 16.100 51.366 .000 .165
Lower-bound 32.201 1.000 32.201 51.366 .000 .165
time * q0006recode Sphericity Assumed 1.200 4 .300 .957 .431 .007
Greenhouse-Geisser 1.200 3.956 .303 .957 .430 .007
Huynh-Feldt 1.200 4.000 .300 .957 .431 .007
Lower-bound 1.200 2.000 .600 .957 .385 .007
Error(time) Sphericity Assumed 162.991 520 .313
Greenhouse-Geisser 162.991 514.305 .317
Huynh-Feldt 162.991 520.000 .313
Lower-bound 162.991 260.000 .627
• What is the null hypothesis for time that is tested above? Using an alpha of .01, what do you conclude about the null? Interpret the effect size if you reject the null.
• What is the null hypothesis for the interaction? Using an alpha of .01, what do you conclude about the null. Interpret the effect size if you reject the null hypothesis.
