1.
Tests of Between-Subjects Effects
Dependent Variable:Town_trips
Type III Sum of
Source
Squares
df
Mean Square
F
Sig.
Corrected Model
196.548a
3
65.516
.626
.599
Intercept
7437.372
1
7437.372
71.103
.000
Gender
98.455
1
98.455
.941
.334
Upper
15.778
1
15.778
.151
.698
Gender * Upper
13.046
1
13.046
.125
.725
Error
12551.942
120
104.600
Total
24899.750
124
Corrected Total
12748.490
123
a. R Squared = .015 (Adjusted R Squared = -.009)
Sadly, I cannot control the results of the data collection that y’all did at the beginning of
the semester, so none of the effects proved significant. The results of the between
subjects 2-way ANOVA revealed no overall effect (F (3, 120) = 0.63, MSE = 104.60, p >
.05. This indicates that none of our conditions differed from one another. (technically,
you would not have to report the rest, but here goes…) The main effect of class year was
not significant (F (1, 120) = 0.15, MSE = 104.60, p > .05), nor was the main effect of
gender (F (1, 120) = 0.94, MSE = 104.60, p > .05). Even the stinking interaction effect
failed to yield a significant effect, (F (1, 120) = 0.13, MSE = 104.60, p > .05). We did not
have enough evidence to conclude that class year or gender influences the number of trips
students take into town. The only other relevant interpretation that I can make is that it’s
time to write a new homework question.
2)
Males
Females
SST
=
=
=
=
Small
(x) = 15
Mean = 3
(x2) = 47
(x) = 15
Mean = 3
(x2) = 51
Large
(x) = 20
Mean = 4
(x2) = 86
(x) = 5
Mean = 1
(x2) = 7
x2 - G2/N
(47 + 86 + 51 + 7) - (15 + 20 + 15 + 5)2/20
191 - 552/20
191 - 151.25
=
39.75
__________________________________________
SSM
=
SSE
(T2/n) - G2/N
=
(152/5) + (202/5) + (152/5) + (52/5) - 151.25
=
45 + 80 + 45 + 5 - 151.25
=
175 - 151.25
=
__________________________________________
=
SST - SSM
=
39.75 - 23.75
=
23.75
16
______________________________________
SSFam
SSGen
SSFxG
=
=
=
=
=
=
=
=
=
[(TA2/n)] - G2/N
=
(15+15)2/10 + (20+5)2/10 - 151.25
=
(302/10) + (252/10) - 151.25
=
90 + 62.5 - 151.25
=
152.2 - 151.25
=
__________________________________________
[(TR2/n)] - G2/N
(15+20)2/10 + (15+5)2/10 - 151.25
(352/10) + (202/10) - 151.25
122.5 + 40 - 151.25
162.5 - 151.25
=
__________________________________________
SSM - (SSA + SSB)
23.75 - (1.25+ 11.25)
23.75- 12.5
=
1.25
11.25
11.25
__________________________________________
Source
Model
Error
Total
Fam. Size
Gender
G*F
df
3
16
19
1
1
1
Omnibus Test
SS
MS
23.75
7.92
16.00
1.00
39.75
Tests of Individual Factors
1.25
1.25
11.25
11.25
11.25
11.25
Fobs
7.92
7.92
Fcrit
3.24
1.25
11.25
11.25
4.49
4.49
4.49
The omnibus test was significant because Fobs was greater than Fcrit (from the F-table in
the back of the book): F (3, 16) = 7.92, MSE = 1.00, p < .05. This indicates that at least
one of our factors had a significant effect on desired number of children. The main effect
of family size was not significant because Fobs was less than Fcrit: F (1, 16) = 1.25, MSE
= 1.00, p > .05. This indicates that the people from small families would like to have just
as many kids as people from large families. The main effect of gender was significant: F
(1, 16) = 11.25, MSE = 1.00, p < .05. Males want to have more children than females.
Finally, the interaction effect was also significant: F (1, 16) = 11.25, MSE = 1.00, p < .05.
The interaction was observed because whereas males from larger families want to have
more children than males from smaller families, the opposite was observed for females;
that is, females from smaller families reported wanting to have more children than
females from larger families (see figure below).
5
4
3
Males
Females
2
1
0
Small
Large