Significance Difference in Years in Formal Education between Types of Jobs

This study analyzes the significance difference in years in formal education between different types of jobs. The data set is based on a local store's credit issuance decision. Various SPSS tests have been applied to assess the difference and descriptive analysis of factors.

Assessment

Contents
INTRODUCTION...........................................................................................................................................3
Question 1...............................................................................................................................................3
Question 2...............................................................................................................................................7
Question 3...............................................................................................................................................9

INTRODUCTION
The given data set is based on local store who is planning whether they should issue a
credit or not. With an aim of finding a particular outcome, different kinds of SPSS tests have
been applied. The rationale behind applying such testes to assess significance difference between
formal education and job types. As well as descriptive analysis of given factors is also completed
under the project.
Question 1
The expectation is of significant difference in years in formal education (Education) between
types of jobs (Job type). Is this expectation supported by the data?
In the question, dependent data is related to types of job. In the given data set, types of job are
categorized in four segments which are as:
0.00 Manual
1.00 Unskilled
2.00 Skilled
3.00 Managerial
Comparison of variable means:
Test of homogeneity:

Between
-
Subjects
Factors
Value
Label
N
Type of
job
.00 Manual 31
1.00 Unskilled 39
2.00 Skilled 26
3.00 Managerial 10
Descripti
ve
Statistics
Dependent
Variable:
Years in
formal
education
Type of
job
Mean Std.
Deviation
N
Manual 6.9677 2.02458 31
Unskilled 11.3077 3.51803 39
Skilled 10.7692 3.20384 26
Manageria
l 15.0000 2.16025 10
Total 10.2547 3.78232 106

Post-hoc Tamhane Test
Multiple
Compariso
ns
Dependent
Variable:
Years in
formal
education
(I) Type of
job
(J) Type of
job
Mean
Difference (I-
J)
Std. Error Sig. 95%
Confidence
Interval
Lower
Bound
Upper Bound
Tukey
HSD
Manual
Unskilled -4.3400* .71153 .000 -6.1984 -2.4815
Skilled -3.8015* .78637 .000 -5.8554 -1.7476
Managerial -8.0323* 1.07539 .000 -10.8411 -5.2234
Unskilled
Manual 4.3400* .71153 .000 2.4815 6.1984
Skilled .5385 .74868 .889 -1.4170 2.4939
Managerial -3.6923* 1.04815 .004 -6.4300 -.9547
Skilled
Manual 3.8015* .78637 .000 1.7476 5.8554
Unskilled -.5385 .74868 .889 -2.4939 1.4170
Managerial -4.2308* 1.10032 .001 -7.1047 -1.3568
Managerial
Manual 8.0323* 1.07539 .000 5.2234 10.8411
Unskilled 3.6923* 1.04815 .004 .9547 6.4300
Skilled 4.2308* 1.10032 .001 1.3568 7.1047
Tamhane Manual Unskilled -4.3400* .67050 .000 -6.1617 -2.5183
Skilled -3.8015* .72596 .000 -5.8088 -1.7942

Managerial -8.0323* .77388 .000 -10.3865 -5.6780
Unskilled
Manual 4.3400* .67050 .000 2.5183 6.1617
Skilled .5385 .84388 .989 -1.7614 2.8384
Managerial -3.6923* .88545 .002 -6.2409 -1.1437
Skilled
Manual 3.8015* .72596 .000 1.7942 5.8088
Unskilled -.5385 .84388 .989 -2.8384 1.7614
Managerial -4.2308* .92815 .001 -6.8871 -1.5744
Managerial
Manual 8.0323* .77388 .000 5.6780 10.3865
Unskilled 3.6923* .88545 .002 1.1437 6.2409
Skilled 4.2308* .92815 .001 1.5744 6.8871
Based on
observed
means.
The error
term is
Mean
Square(Erro
r) = 8.744.
*. The mean
difference is
significant
at the .05
level.
Years in
formal
education
Type of
job
N Subse
t

1 2 3
Tukey HSDa,b,c
Manual 31 6.9677
Skilled 26 10.7692
Unskilled 39 11.3077
Manageria
l 10 15.0000
Sig. 1.000 .938 1.000
Means for
groups in
homogene
ous
subsets
are
displayed.
Based on
observed
means.
The error
term is
Mean
Square(Er
ror) =
8.744.
a. Uses
Harmonic
Mean
Sample
Size =
20.371.

b. The
group
sizes are
unequal.
The
harmonic
mean of
the group
sizes is
used.
Type I
error
levels are
not
guarantee
d.
c. Alpha =
.05.
On the basis of above tables, this can be stated that value of significance difference under each
types of job is under 0.05. Therefor we can strongly reject to H0 and can go with HA.
As question is asking to compare mean of dependent variable among different independent
variables regards to formal education.
Hypothesis:
H0: There is no significance difference in mean value of types of job and formal education.
HA: There is significance difference in mean value of types of job and formal education.

ANOVA
Type of job
Sum of
Squares
df Mean
Square
F Sig.
Between
Groups 41.548 17 2.444 4.033 .000
Within Groups 53.329 88 .606
Total 94.877 105
The above table states that value of significance difference is of 0.000 that is lower than 0.05.
Therefore, we will go with HA that is related to significance difference in mean value of types of
job and formal education.
Conclusion
In conclusive manner this can be stated that there is relation between people who have different
kinds of job and education. This is so because on the grounds of types of education, people have
job and it has been justified by computed value of significance difference.
Question 2
Using Age, Job years, Education, Additional income and Tot balance construct a predictive
model of Job income.
In order to create a predictive model, a linear regression needs to be applied. Under this question,
dependent variable is job income. While rest of independent variable is age, job years, education
and additional income.
Hypothesis:
H0: There is no evidence that each coefficient is significantly different from zero.

HA: There is evidence that each of the six coefficients is significantly different from zero.
Variables
Entered/
Removed
Model Variables
Entered
Variables
Removed
Method
1
Years in
employment,
Balance of
debt, Years in
formal
education,
Additional
income, Age
in yearsb
. Enter
a.
Dependent
Variable:
Current
job
income
b. All
requested
variables
entered.

Model
Summary
Model R R Square Adjusted R
Square
Std. Error of
the Estimate
1 .681a .464 .436 363.67946
a.
Predictors
:
(Constant)
, Years in
employme
nt,
Balance of
debt,
Years in
formal
education,
Additional
income,
Age in
years
ANOVAa
Model Sum of
Squares
df Mean
Square
F S
i
g
.
1 Regression 11199614.19
3
5 2239922.839 16.935 .000b

Residual 12961749.15
3 98 132262.746
Total 24161363.34
6 103
a.
Dependen
t
Variable:
Current
job
income
b.
Predictors
:
(Constant
), Years in
employm
ent,
Balance
of debt,
Years in
formal
education,
Additiona
l income,
Age in
years
Coefficient
s

Model Unstandardized
Coefficients
Standardized
Coefficients
t S
i
g
.
B Std. Error Beta
1
(Constant) -99.460 138.602 -.718 .475
Age in years 8.186 4.114 .196 1.990 .049
Years in formal
education 71.545 10.391 .530 6.886 .000
Additional income .014 .134 .009 .108 .914
Balance of debt .033 .020 .126 1.671 .098
Years in employment 7.626 7.780 .098 .980 .329
a.
Dependent
Variable:
Current job
income
On the basis of above table, this can be stated that H0 is accepted as there is no evidence that
each coefficient is significantly different from zero. This is so because almost each independent
variable’s relation with dependent variable is more than 0.05.
Conclusion
In terms of conclusion, it can be stated that current job income has no relation with different
kinds of factors like age, education, balance of debt etc. This is so because in ANOVA test, each
value of significant difference is greater than 0.05.
Question 3
Do the data support the claim that years in employment (Job years) is significantly lower for
female compared to male respondents (Gender)?

In this question, years in employment is considered as dependent variable. While another
variable is related to gender. In this aspect independent T test needs to be applied but before
doing so normality test should be applied that is as follows:
Descriptive
Gender Statistic S
t
d
.
E
r
r
o
r
Years in
employment
Female
Mean 1.5833
.
4
1
6
6
7
95% Confidence
Interval for Mean
Lower
Bound .7214
Upper
Bound 2.4453
5% Trimmed Mean 1.2963
Median 1.0000
Variance 4.167
Std. Deviation 2.04124
Minimum .00
Maximum 9.00

Range 9.00
Interquartile Range 2.00
Skewness 2.362
.
4
7
2
Kurtosis 7.055
.
9
1
8
Male
Mean 5.2439
.
7
4
5
4
7
95% Confidence
Interval for Mean
Lower
Bound 3.7607
Upper
Bound 6.7272
5% Trimmed Mean 4.3089
Median 3.0000
Variance 45.569
Std. Deviation 6.75051
Minimum .00
Maximum 30.00
Range 30.00
Interquartile Range 5.00

Skewness 2.129
.
2
6
6
Kurtosis 4.339
.
5
2
6
On the basis of above table, this can be assessed that value of mean and median are close
together. As well as above data meeting ideal criteria of skewness and kurtosis which is under
+/-1 and +/-7.
Box plot:

Histogram:

On the basis of above chart, this can be find out that there is difference in values of years in
employment of male and female. Though, this is just a graphical presentation, further some tests
have been applied in order to find a particular outcome.
Hypothesis:
H0: There is no significant difference in variances in years in employment between female and
male.
HA: There is significant difference in variances in years in employment between female and
male.

α = level of significance 5% (0.05)
Group
Statistics
Gender N Mean Std.
Deviation
Std. Error
Mean
Years in
employment
Female 24 1.5833 2.04124 .41667
Male 82 5.2439 6.75051 .74547
Indepe
ndent
Sampl
es Test

Levene's
Test for
Equality of
Variances
t
-
t
e
s
t
f
o
r
E
q
u
a
l
i
t
y
o
f
M
e
a
n
s
F Sig. t df Sig.
(2-
tailed
)
Mean
Differenc
e
Std.
Error
Differenc
e
9
5
%
C
o

n
f
i
d
e
n
c
e
I
n
t
e
r
v
a
l
o
f
t
h
e
D
i
f
f
e
r
e
n

c
e
Lower U
p
p
e
r
Years in
employme
nt
Equal
variance
s
assumed
10.78
2
.00
1
-
2.61
4
104 .010 -3.66057 1.40045
-
6.4377
2
-.8834
2
Equal
variance
s not
assumed
-
4.28
6
103.82
8 .000 -3.66057 .85401
-
5.3541
4
-
1.9670
0
On the basis of above table, this can be inferred that value of significance difference in Levene's
Test for Equality of Variances is of 0.001 that is lower than 0.05. Therefore, this can be
interpreted that H0 should be rejected as there is no significant difference in variances in years in
employment between female and male.
While in the t-test for Equality of Means the value of significance difference is more than 0.05.
Conclusion
As per the above table this can be concluded that level of significance difference is 0.001<0.05.
Thus we will retain to HA that is about existing significance difference in variances in years in
employment between female and male. And there is sufficient evidence that years in
employment (Job years) is significantly lower for female compared to male respondents
(Gender).