There are controversial literature on whether to use the mean of likert scale items (by likert scale i'm referring to a scale that ranges from strongly disagree (1) to strongly agree with value of 5). How ever it is common to see such practice in research. Therefore i want some reliable information and source on whether this practice is right and the process of computing the mean value and setting the cut of point for interpretation.
First compute the mean value by adding all the weights and then divide them by five:
1+2+3+4+5=15
15: 5= 3
Next, depending on the direction of the targeted hypothesis, compare each response with the mean. The responses above the mean value (i.,e 3) are significant.
Regarding the questionnaire items of a Likert-scale , first Friedman Test is used to see whether there is a significant difference among items or not. then, the means are compared. if no significant difference, the story is different. BEST REGARDS
o
I would recommend emphasizing the median in such cases - it's fine to report Likert means and give the standard disclaimer that it isn't exactly a valid number, but as representative summary statistics, quantile based numbers are usually more informative for ordinal data in my view.
- Badges
- Science topic
- Similar topics
- Philosophy
- Esthetics
- Interpretation