After getting score of pre-test and post-test, the next thing to do is analyze data. However, before analyzing the data by using t-test formulation, the
researcher did a test of normality and a test of homogeneity.
1. Test of Normality and Test of Homogeneity
Test of normality and homogeneity in this research were conducted through SPSS Special Package for the Social Sciences 20 version. Test of
normality was conducted in order to know whether the distribution from the two classes were normal or not. The test of normality was using Kolmogorov-
Smirnov and Shapiro-Wilk table. Sig. score in Kolmogorov-Smirnov and Shapiro-Wilk table should be above 0.05 in order to have normal distribution
data. In similar to the test of normality, the test of homogeneity in this research
was also conducted through SPSS 20. Test of homogeneity was conducted to know whether the data from two classes had the same or different variant. The
test of homogeneity was using Levene table. Sig. score in Levene table should be above 0.05 in order to have homogeny distribution data. These two kind of
tests are conducted to pre-test score and post-test score. To compute data, the steps are needed as follows.
a. Open SPSS program b. Go to variable view and fill in the columns as follows.
a Name: write Class in first row. This is to indicate and differentiate between experimental class and controlled
class. Write score in second row. b Typecolumn was numeric
c Width column is filled 8 d Decimal change this row from 2 to 0
e Label column was left blank f Value column none
g Missing column is none h Columns is 8
i Align column is right
j Measure column was unknown k Role column is filled with input
In calculating homogeneity and normality, column name and column decimal are needed to pay attention to.
c. Click data view, in score column, compute 1 as the symbol of experimental class and 2 as the symbol of controlled class.
d. In score column, compute score of each class e. Click analyze data descriptive explore
f. Put score variable in dependent list, while class variable in factor list
g. Click plots tick histogram, normality plots with test and power estimation
h. Click continue i. Click Ok
j. Data of normality and homogeneity would appear in a box. Here are the examples of data served in table.
Tests of Normality
class Kolmogorov-Smirnov
a
Shapiro-Wilk Statistic
df Sig.
Statistic df
Sig. Score
1 .145
34 .067
.946 34
.094
a. Lilliefors Significance Correction
Test of Homogeneity of Variance
Levene Statistic df1
df2 Sig.
score Based on Mean
.057 1
66 .811
Based on Median .058
1 66
.810 Based on Median and with
adjusted df .058
1 65.840
.810 Based on trimmed mean
.056 1
66 .814
In test of normality, sig. score is needed to be above 0.05 to be considered that data is normal distributed. Meanwhile in test of homogeneity, sig. score in