Data Analysis Using Spss t Test 1224391361027694 8

7/28/2019 Data Analysis Using Spss t Test 1224391361027694 8

1/24

Data Analysis Using

SPSSt-test


2/24

t-test

Used to test whether there issignificant difference between themeans of two groups, e.g.:

Male v female

Full-time v part-time


3/24

t-test

Typical hypotheses for t-test:a) There is no difference in affective

commitment (affcomm) between male

and female employeesb) There is no difference in continuance

commitment (concomm) between maleand female employees

c) There is no difference in normativecommitment (norcomm) between maleand female employees


4/24

Performing T-test

Analyze

Compare Means

Independent-Samples T-test


5/24


6/24


7/24

Performing T-test

Select the variables to test (TestVariables), in this case:

affcomm

concomm

norcomm

And bring the variables to the TestVariablesbox


8/24


9/24


10/24

Performing T-test

Select the grouping variable, i.e.gender; bring it to the groupingvariablebox

Click Define Groups


11/24


12/24

Performing T-test

Choose Use specified values

Key in the codes for the variablegender as used in the ValueLabels. In this case:

1 - Male

2 - Female

Click Continue, then OK


13/24


14/24

T-Test: SPSS Output

Group Statistics

357 3.49720 .731988 .038741

315 3.38016 .696273 .039231

357 3.18838 .756794 .040054

315 3.15159 .666338 .037544

357 3.24090 .665938 .035245

315 3.27540 .647409 .036477

GENDER OF

RESPONDENT

MALE

FEMALE

MALE

FEMALE

MALE

FEMALE

affcomm

concomm

norcomm

N Mean Std. Deviation

Std. Error

Mean


15/24

Independent Samples Test

1.048 .306 2.116 670 .035 .117040 .055308 .008442 .225638

2.123 666.213 .034 .117040 .055135 .008780 .225300

5.353 .021 .665 670 .506 .036788 .055335 -.071863 .145440

.670 669.997 .503 .036788 .054899 -.071006 .144582

.656 .418 -.679 670 .497 -.034500 .050813 -.134272 .065271

-.680 663.726 .497 -.034500 .050723 -.134097 .065096

Equal variances

assumedEqual variances

not assumed

Equal variances

assumed

Equal variances

not assumed

Equal variances

assumed

Equal variances

not assumed

affcomm

concomm

norcomm

F Sig.

Levene's Test for

Equality of Variances

t df Sig. (2-tailed)

Mean

Difference

Std. Error

Di ffe rence Lower Upper

95% Confidence

Interval of the

Difference

t-test for Equal ity of M eans

T-test: SPSS Output


16/24

From the SPSS output, we areable to see that the means of therespective variables for the twogroups are:

Affective commitment (affcomm) Male 3.49720 Female 3.38016

Continuance commitment (concomm) Male 3.18838 Female 3.15159

Normative commitment (norcomm) Male 3.24090 Female 3.27540


17/24

T-test: Interpretation

For the variable affcomm

Levenes Test for Equality of Variancesshows that F (1.048) is not significant

(0.306)* therefore the Equal variancesassumed row will be used for the t-test.

* This score (sig.) has to be 0.05 or less to beconsidered significant.


18/24


Under the t-test for Equality ofMeans look at Sig. (2-tailed)for Equal variances assumed.

The score is 0.035 (which is lessthan 0.05), therefore there is asignificant difference betweenthe means of the two groups.


19/24


1.048 .306 2.116 670 .035 .117040 .055308 .008442 .225638

2.123 666.213 .034 .117040 .055135 .008780 .225300

5.353 .021 .665 670 .506 .036788 .055335 -.071863 .145440

.670 669.997 .503 .036788 .054899 -.071006 .144582

.656 .418 -.679 670 .497 -.034500 .050813 -.134272 .065271

-.680 663.726 .497 -.034500 .050723 -.134097 .065096

Equal variances


not assumed

Equal variances

assumed

Equal variances

not assumed

Equal variances

assumed

Equal variances

not assumed

affcomm

concomm

norcomm

F Sig.

Levene's Test for



Mean

Difference

Std. Error


95% Confidence

Interval of the

Difference




20/24


For the variable concomm

Levenes Test for Equality of Variancesshows that F (5.353) is significant

(0.021)* therefore the Equal variancesnot assumedrow will be used for the t-test.

* This score (sig.) is less than 0.05, so thereis significant different in the variances of thetwo groups.


21/24


Under the t-test for Equality ofMeans look at Sig. (2-tailed) forEqual variances not assumed.

The score is 0.503 (which is morethan 0.05), therefore there is nosignificant difference between themeans of the two groups.


22/24


1.048 .306 2.116 670 .035 .117040 .055308 .008442 .225638

2.123 666.213 .034 .117040 .055135 .008780 .225300

5.353 .021 .665 670 .506 .036788 .055335 -.071863 .145440

.670 669.997 .503 .036788 .054899 -.071006 .144582

.656 .418 -.679 670 .497 -.034500 .050813 -.134272 .065271

-.680 663.726 .497 -.034500 .050723 -.134097 .065096

Equal variances


not assumed

Equal variances

assumed

Equal variances

not assumed

Equal variances

assumed

Equal variances

not assumed

affcomm

concomm

norcomm

F Sig.

Levene's Test for



Mean

Difference

Std. Error


95% Confidence

Interval of the

Difference




23/24


For the variable norcomm

Levenes Test for Equality of Variancesshows that F (0.656) is not significant

(0.418)* therefore the Equal variancesare assumedrow will be used for the t-test.

* This score (sig.) is more than 0.05, so thereis no significant different in the variances ofthe two groups.


24/24


Under the t-test for Equality ofMeans look at Sig. (2-tailed) forEqual variances assumed.

The score is 0.497 (which is morethan 0.05), therefore there is nosignificant difference between themeans of the two groups.

Data Analysis Using Spss t Test 1224391361027694 8

Documents