102
Table 4.39 I am Willing to Recommend to Others to Shop at
Carrefour Answer
Frequency Percent
Strongly Disagree Disagree
Neutral 1
1 Agree
51 51
Strongly Agree 48
48 Total
100 100
Source: Primary data Processed 2016
From the table above 4.39 explain that 48 or 48 of respondents stated strongly agree, 51 or 51 of respondents
agreed, 1 or 1 of respondents expressed a neutral, 0 or 0 of respondent disagreed and 0 or 0 of respondents stated strongly
disagree with the statement that I am willing to recommend to others to shop at Carrefour.
D. Classical Assumption Test Results
1. Data Normality Test Results
Normality test aims to test whether the regression model or residual confounding variables have a normal distribution. The
regression model was good and decent distribution of the data used is normal or near normal Ghozali, 2011:160.
103
Figure 4.2 Normality Test Results in Graph
Source: SPSS output the results of the primary data that have been processed, 2016
Based on the analysis of data in figure 4.2 above the normal curve p-plot, it can be concluded that the normal curve p-plots seen
point spread around the diagonal line and distribution is too far or wide. Means of this curve indicates that the corresponding
regression model assumptions of normality and fit for use. In addition to researchers test chart also complete normality
test with statistical tests. One of the statistical tests that can be used to test the residual normality is a non-parametric statistical test
Kolmogrov-Smirnov K-S Ghozali, 2011:164.
104
By hypothesis:
a.
Make a hypothesis describing sentence :
H0 = Samples derived from data or population v normally distributed.
Ha = Samples derived from data or populations that are not normally distributed.
b.
Rule testing If the probability sig 0,05 then Ho is accepted
If the probability sig 0,05 then Ho is rejected
c.
Decision Normal distribution of data
Table 4.40 Normality Test Results in Statistics
One-Sample Kolmogorov-Smirnov Test
Unstandardiz ed Residual
N 100
Normal Parameters
a,b
Mean 0E-7
Std. Deviation
1.18011892
Most Extreme Differences
Absolute .096
Positive .096
Negative -.062
Kolmogorov-Smirnov Z .956
Asymp. Sig. 2-tailed .320
a. Test distribution is Normal. b. Calculated from data.
Source: SPSS output the results of the primary data that have been processed, 2016
105
According to the table 4.40 the value of the Kolmogorov - Smirnov test was 0,956 so it can be seen that the value
unstandardized residual value Asymp. Sig 0,05 and this means that data is distributed normally.
2. Test Results Multicollinearity