5 Table of variable inclination The next description is the determination of score
categorization, obtained by each variable. From score is then divided into 3 categories. This was carried out based on the Mean
and SD of the acquired. Variable data research categorized by the following rules:
1 very Good group All students whose score were X
≥ M + 1.5 SD. 2 Good Groups
All of students whose score were M ≤ X . M + 1.5 SD
3 Less groups All students who have score M
– 1.5 SD ≤ X M 4 group is very less
All students who have score X ≤ M – 1.5 SD.
Djemari Mardapi, 2008: 123 6 Pie Chart
Pie chart is created based on the data of the trend that has been shown in the table variable inclination.
2. Analysis Prerequisite Test
To get a proper conclusion required analysis of data. Before data is analysed then first performed the test requirement analysis, namely
linearity and test multicolinearity test.
a. Linearity Test
Linarites test is intended to determine whether the variables are free and bound variables have a linear relationship or not. Between the
free variables and bound variables is said to be linear if the increase in
score free variables followed by increase in variable. To know it, the two variables to be tested with the F on the significance level of 5.
The formula used is:
Description: F
reg
: price for regression line RK
reg
: the average quadratic regression line RK
res
: the average quadratic residues Sutrisno Hadi, 2004: 13
Results F
count
be consulted with F
table
with 5 significance level. If the smaller F
count
the same as F
table
, meaning the relationship between free variables and bound variables is liniear. Conversely, if
F
count
is greater than F
table
means variable relationship free and bound variables are non-linear.
b. Multicolinearity Test
Multicolinearity test is used to find out whether multicolinearity occurs between the free variables of the one with the other free
variables. The technique used is statistics by Product the Moment. The formula is as follows:
∑ ∑ ∑ √{ ∑
∑ }{ ∑
∑ }
Description:
r
xy
: coefficient of correlation between variables X and Y N
: number of respondents ∑XY
: multiplication of the number of variables X and Y ∑X
: the amount of the value of the variable X ∑Y
: the amount of the value of the variable Y ∑X
2
: the sum of the square of the value of the variable X ∑Y
2
: the number of the square of the value of the variable Y Suharsimi, 2010: 213
If the coefficient of correlation between the free variables is smaller or equal to 0.600, then do not occur Multicolinearity between
free variables, regression test can be continued Danang Sunyoto, 2007: 89. Whether or not, to know the existence of Multi co-linearity
might be used other ways, namely by values of tolerance ɑ and the value of the variance inflation factor VIF. Free variables are
experiencing Multicolinearity if ɑ
count
ɑ and VIF
count
VIF. Instead, free variables are not subjected to Multicolinearity if ɑ ɑ
count
and VIF
count
VIF.
3. Hypothesis Test