Question

An instructor measured her students’ attitudes toward statistics in general at various points throughout the semester. Their data are presented in the table below. Conduct the Friedman test using SPSS to test whether attitudes changed over the course of the semester. Be sure to use the method that provides more informative output (test statistic, etc.). Show your output file.

Student	Prior to 1st Class	At the Midterm	After the Final
A	3	4	7
B	2	3	5
C	4	3	6
D	5	6	7
E	4	2	7
F	1	3	5
G	3	5	9
H	6	5	8
I	5	6	8

Based on your output, what was the value of the test statistic (chi-square)?(Please round your answer to 3 decimal places)

Based on your SPSS output, what was the p-value of the Friedman test? (Please round your answer to 3 decimal places)

Based on your p-value from #39 and assuming a .05 alpha level, was the result of the Friedman test significant? (i.e., should you reject the null hypothesis?)

Yes or No?

Answer 1

Firstly define the null and alternate hypothesis as follows:

The three variable are not different from each other, that is, the attitude towards Statistics remains same throughout the semester. .

Atleast one of the three variables is different from the remaining variables.

Now, proceed as follows in SPSS in order to carry out the Friedman test and receive the test results.

Step 1: In the 'Variable View' tab, define the three variables "Prior_to_Class 1", "At_Midterm" and "After_Final" in accordance with the given data and choose the measure as "ORDINAL" for all the three variables. Now, input the given data into the "Data View" tab as shown below:

Step 2: Go to Analyze Nonparametric Tests Legacy Dialogs K Related Samples...

The screenshot is shown below:

Step 3: A dialogue box showing "Tests for Several Related Samples" appear.

Step 4: Now drag all the three variables defined above into the box provided below "Test Variables" , in the following manner:

Step 5: Check the box corresponding to 'Friedman' in the "Test Type" area. This option is automatically chosen by SPSS in case of no test selected.

Step 6: Click on the Statistics button and select the option "Quartiles".

Step 7: Click Continue and the dialogue box showing "Tests for Several Related Samples" will re-appear.

Step 8: Click "OK" to get the results of the Friedman test. The outputs obtained are:

Based on the output received by SPSS, the following results can be stated:

The value of the Chi-Square test statistic is 14.000 at 2 degrees of freedom and the corresponding p-value is 0.001, as shown below in the following picture:

Here, .

Since , hence, there is sufficient evidence to reject the null hypothesis and it can be concluded that atleast one of the three variable is different from the remaining variable. This implies that the students attitude towards the subject Statistics changed during the observed semester.