Performing A Paired Data of Nonparametric Test: Presenter: Muhammad Azri Aminudin
Performing A Paired Data of Nonparametric Test: Presenter: Muhammad Azri Aminudin
Performing A Paired Data of Nonparametric Test: Presenter: Muhammad Azri Aminudin
PAIRED DATA OF
4.0 NONPARAMETRIC
TEST
PRESENTER : MUHAMMAD AZRI AMINUDIN
Non parametric test propose the cases where samples are paired: Wilcoxon signed rank test.
The Wilcoxon Signed Rank test assumes that each measurement pair is distinct from the other
measurement pairs. With ordinary and continuous variables, this test can be used.
For this test, we use another data that suitable.
Look for the Student’s t row in the Test for Location table.
The Statistic column identifies the t-statistic of 82.87955.
The p Value column identifies the p-value as Pr > |t| of
<.0001, which is less than the significance level of 0.05 so
we reject the null hypothesis. It concludes the mean value
of the variable is significantly different from zero.
USING PROC TTEST TO TEST PAIRED DIFFERENCES
OUTPUT :
RESULT :
The value of Pr > |t| is <0.0001 which is less than the
significance level of 0.05. So we reject the null
hypothesis. This test has the same result as the
analysis of the difference variable. We conclude that
the mean difference from the two methods is
significantly different from 0. We interpret the p-
value for this test the exact same way as described
for PROC UNIVARIATE.
NORMAL PROBABILITY PLOT
OUTPUT : RESULT :
Figure above displays the box plot of Wilcoxon scores for Height. This graph corresponds to the Wilcoxon
scores analysis shown previously.
Box plot are available for all PROC NPAR1WAY score types except median scores, which are displayed in a
stacked bar chart. If ODS Graphics is enable but do not specify the PLOTS = option, PROC NPAR1WAY
produces all plots that are associated with the analysis that we request.