Hi Friends,
I have a question regarding the interpretation of kendall tau b correlation coefficient.
I have two questions (ordinal by ordinal ) and they have the following available responses (Strongly Agree (1), Somewhat Agree(2), Neutral (3), Somewhat Disagree (4) and Strongly Disagree (5). Sample size 127
How should I interpret the Kendall Tau b results while comparing these two likert type statements?
For example _ Kendall Tau b = ( -.495)
What it suggests:
-Does it suggest that the respondents who agree or disagree on Q1, disagree or agree respectively on the second question in their individual capacities?
-Does it suggest that the respondent who agree or disagree on Q1, disagree or agree respectively on the second question when compared with the other respondents?
- Does it mean the respondents (whole group) when agree to Q1, do the opposite (disagree) with Q2 thereby both the questions are negatively correlated?
Please share your knowledge
SPSS Statistics generates one main table for the Kendall's tau-b procedure that you ran in the previous section. If your data passed assumption #2 (i.e., there is a monotonic relationship between your two variables), which we explained earlier in the Assumptions section, you will only need to interpret this one table. However, since you should have tested your data for this assumption, you will also need to interpret the SPSS Statistics output that was produced when you tested for it (i.e., your scatterplot results). Remember that if your data failed this assumption, the output that you get from the Kendall's tau-b procedure (i.e., the table we discuss below) will no longer be correct.
Ehsan Namaziandost Thank you for providing me the answer. But it has led to my second question. Is the output received from the kendall tau (degree of monotonic relationship) not the confirmation of whether there is a monotonic relationship or not. How to check the monotonic assumption when we have likert type statement (SA- A-N-D-SD). The scatter plot shows dots all over the place as many respondents have provided different-different responses. So how to generate a scatter plot for the likert type statement. The Spearman rho is also used for finding the association between the two ordinal variables and some scholars have said that its output is a confirmation (in terms of degree of monotonic relationship) whether there is a monotonic relationship between the variables or not.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size