Page 79 - IJET_July2021_final

Table-9: Tests of Normality for Readiness Scores of teacher educators,
research scholars and pupil teachers towards blended learning approach

Criterion Variable: Readiness towards blended learning approach
Category Designation Kolmogorov-Smirnov a Shapiro-Wilk

Readiness Statistic df Sig. Statistic df Sig.
Scores towards
blended Teacher .126 106 .000 .919 106 .000
learning Educators
approach Research .135 105 .000 .923 105 .000
Scholars
Trainee .117 279 .000 .929 279 .000
a. Lilliefors Significance Correction

From table-9, it is evident that the non parametric statistics technique).
statistical value of Kolmogorov-Smirnov Further before applying Kruskal Wallis
test of the Readiness Scores of Teacher H Test, all the assumptions related to
Educators, Research Scholars and Kruskal Wallis H Test were also tested,
Pupil Teachers is 0.126, 0.135 and which are given below:
0.117 respectively, whose probability
of significance at df (106), df (105) and Assumption#1- Independence of the
df (279) is 0.000 which is less than Groups
0.01, hence significant at 0.01 level of It is clear from the observation of
significance. In this perspective the null independent variable that there is
hypothesis “The given distribution of the independence of the observation in
readiness scores of teacher educators, collecting rural, urban and semi urban
research scholars and pupil teachers groups. Therefore, the first assumption
towards a blended learning approach of Kruskal Wallis H Test of Independence
differ significantly from the normal of the Groups is fulfilled.
distribution” is rejected. Therefore, it
can be concluded that the assumption Assumption#2 The scores of
of normality of readiness scores of dependent variables must be at least
Teacher Educators, Research Scholars on ordinal scale
and Pupil Teachers is not satisfied or it
is violated. It is clear that these scores are on an
interval scale. Therefore, the second
The above description indicates assumption of Kruskal Wallis H Test that
that the readiness scores are not the scores of dependent variables must
normally distributed. Because the first be at least on ordinal scale is fulfilled.
assumption of the normality of data of
one-way ANOVA is not satisfied at all the Assumption#3 Uniformity of
levels of dependent variables, hence we distribution of dependent variable at
use Kruskal Wallis H Test (alternative
both levels of independent variable

Indian Journal of Educational Technology 69
Volume 3, Issue 2, July 2021