I suspect that people who complete higher degrees wait longer to get married. To
ID: 3133152 • Letter: I
Question
I suspect that people who complete higher degrees wait longer to get married. To test this hypothesis, I use General Social Survey data from 1976-2006 to calculate the average age o marriage for people who have a range of levels of education from less than high school to graduate degree. I calculate an ANOVA using SPSS. I obtain the following output:
AGE WHEN FIRST MARRIED
N
Mean
Std. Deviation
Std. Error
95% Confidence Interval for Mean
Min
Max
Lower Bound
Upper Bound
0 LT HIGH SCHOOL
7866
21.41
5.696
.064
21.28
21.54
12
90
1 HIGH SCHOOL
13351
21.86
4.347
.038
21.79
21.93
13
70
2 JUNIOR COLLEGE
974
22.55
4.204
.135
22.29
22.81
14
50
3 BACHELOR
2887
24.05
4.200
.078
23.90
24.20
14
73
4 GRADUATE
1383
24.89
4.550
.122
24.65
25.13
14
55
Total
26461
22.15
4.882
.030
22.09
22.21
12
90
ANOVA
AGE WHEN FIRST MARRIED
Sum of Squares
df
Mean Square
F
Sig.
Between Groups
26398.575
4
6599.644
289.008
.000
Within Groups
604135.361
26456
22.835
Total
630533.936
26460
State the null and research hypotheses and in plain english describe what you can conclude about the association between age of marriage and education in the US population? Describe the association AND state the conclusion from the hypothesis test. Be specific and use numbers from the SPSS output to support your conclusions.
AGE WHEN FIRST MARRIED
N
Mean
Std. Deviation
Std. Error
95% Confidence Interval for Mean
Min
Max
Lower Bound
Upper Bound
0 LT HIGH SCHOOL
7866
21.41
5.696
.064
21.28
21.54
12
90
1 HIGH SCHOOL
13351
21.86
4.347
.038
21.79
21.93
13
70
2 JUNIOR COLLEGE
974
22.55
4.204
.135
22.29
22.81
14
50
3 BACHELOR
2887
24.05
4.200
.078
23.90
24.20
14
73
4 GRADUATE
1383
24.89
4.550
.122
24.65
25.13
14
55
Total
26461
22.15
4.882
.030
22.09
22.21
12
90
Explanation / Answer
H0:There is no difference between all the means
H1: Atleast one of the mean is not same
alpha:5% level
Test statistic:
F cal=289.008
Conclusion:
P value=0.0000 and alpha=0.05
P< alpha we reject the null hypothesis. The results are significant
Hence Atleast one of the mean is not same