Greetings researschers
Hope all is well at your end.
Can someone pls confirm and share references with regard to my following query-
I am using PLS-SEM, my model has reflective scales. The items are on 5 point likert scale and comprise standardised scales. The constructs have negative and positive questions.
Before analysis, are the scores of negative questions to be reversed or the data is to be used without any alteration?
You need to reverse the responses on the likert scale for negative items. Otherwise, the indicator loadings will appear as a negative value and it will affect the other convergent validity measures (e.g., AVE).
Hello Sakshi,
As AVE is based on the mean of squared loadings, there isn't a strict requirement to reverse-score the negatively-worded items. However, some folks find it comforting to do so, so that, ideally, all loadings would have comparable signs.
That said, do be aware that it is not uncommon for mixtures of positively and negatively worded stimuli associated with the same target to appear to yield two, separate factors rather than one unified, bipolar factor.
Finally, if individual Likert-type item responses are used as your manifest/observed variables, do recognize that the scaling (e.g., 1-5) may not correspond to the actual scale values that one would obtain in a more rigorous approach.
Good luck with your work.
Factor measurement could become less reliable when there are negative correlations. Therefore, to maintain scale direction, a researcher could employ reverse scoring or reverse coding, transforming responses of the negatively stated items so that a low score converts to the corresponding high score, thus reflecting the results well on the measure. The following might come in handy.
Hair, J., & Alamer, A. (2022). Partial least squares structural equation modeling (PLS-SEM) in second language and education research: Guidelines using an applied example. Research Methods in Applied Linguistics, 1(3), 100027. https://doi.org/10.1016/j.rmal.2022.100027
Höhne, J. K., & Krebs, D. (2018). Scale direction effects in agree/disagree and item-specific questions: A comparison of question formats. International Journal of Social Research Methodology: Theory & Practice, 21, 91–103. https://doi.org/10.1080/13645579.2017.1325566
Leon, C. M., Aizpurua, E., & van der Valk, S. (2022). Agree or disagree: Does it matter which comes first? An examination of scale direction effects in a multi-device online survey. Field Methods, 34(2), 125–142. https://doi.org/10.1177/1525822X211012259
Matthews, L. (2017). Applying multigroup analysis in PLS-SEM: A step-by-step process. In H. Latan & R. Noonan (Eds.), Partial Least Squares Path Modeling: Basic Concepts, Methodological Issues and Applications (pp. 219–243). Springer International Publishing. https://doi.org/10.1007/978-3-319-64069-3_10
Salzberger, T., & Koller, M. (2019). The direction of the response scale matters – accounting for the unit of measurement. European Journal of Marketing, 53(5), 871–891. https://doi.org/10.1108/EJM-08-2017-0539
Good luck,
David Morse Jhunar John Tauy Thank you so much for your time and guidance.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India