Hello,
I am conducting a survey using a 5 Likert scale. When I tried to analyze the primary results, (none to all) and (strongly disagree to strongly agree) scales. By the way, I have only one question in the survey that uses none to all. It shows a meaningless (negative) correlation between them. I think the issue is (none to all) scale indicates none at 1. However, another scale(strongly disagree to strongly agree) indicates a negative at 1. Maybe that is the reason why I see a negative correlation. Is there any solution to analyze or interpret it? I use SmartPLS for structural equation modeling. Thank you!
The question isn't quite clear to me. But for a Likert scale or series of items, you are free to reverse the coding on an item for analysis. This is often intentionally when devising the scale. See, for example, https://www.statology.org/reverse-coding/
Thank you for your response. Let me clarify it. I have questions that contain strongly disagree to strongly agree on the 5 Likert scales. Also, I have one question which contains (1)None; (2)Very few; (3)Some; (4)Many; (5)All on 5 Likert scales. When I am trying to see relations between the variables. It creates a negative correlation. I think that It doesn't suppose to be like this because the question which contains none to all is about descriptive norms and the other one is about personal norms. What I expect is a positive correlation between them but in primary data is not like this. Therefore I just thought that maybe there is some dissonance between those two items.
A lot depends on whether you are working with single items or multi-item scales where the component variables have Likert-scoring. In the second case, you should evaluate coefficient alpha to determine whether you can combine the individuals items into a single scale. If so, you can add the variables together or average that sum (it amounts to the same thing).
As for the different scoring labels on your variables, this won't matter to your analysis program, which will treat each of them as simply having 5 categories, regardless of how those categories were labelled.
Sayed Ibrahim Ali
Thank you for your answer. Basically, what I tried to explain from my data comes from 1 coded for none and 5 coded for all options. I think that your second idea will be more reliable for results. That means analyzing them separately. However, then I have got only one question which has none to all. Maybe it could make a weakness circumstance for the results. Thanks again.
Separating the two could make them each less reliable than combining them. One thing you haven't mentioned is the correlation between the two variables -- how strong is it?
David L Morgan The correlation between them is - 0.113. Basically, I think that maybe it makes sense. Descriptive social norms may have a negative correlation with personal norms. Because if you know a lot of people who are keen to act environmentally friendly then the effect of personal norms may decrease when we compare people who know fewer people. I think it is more obvious now for me. Thank you for your contribution by the way.
If they aren't correlated, then really your only choice is to treat them separately.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India