How to deal with 5-point Likert Scale variables in regressions

[Eleonora Facibeni]

Hi everyone. I would really appreciate it if you could help me deal with two 5-point Likert scale independent variables, ranging from 1=strongly disagree to 5=strongly agree. They refer to employees' satisfaction with the introduction and usefulness of a new business model. In binary logistic regressions and multiple linear regressions, should I treat Likert scale variables as categorical regressors or continuous ones? I have already consulted some academic articles (e.g., Norman, 2010; Harpe, 2015), but I'm not sure what can be the most suitable approach. Thank you for your help.

Norman G., Likert scales, levels of measurement and the "laws" of statistics. Adv Health Sci Educ Theory Pract. 2010 Dec;15(5):625-32, DOI: 10.1007/s10459-010-9222-y.
Spencer E. Harpe, How to analyze Likert and other rating scale data, Currents in Pharmacy Teaching and Learning, Volume 7, Issue 6, 2015, Pages 836-850, https://doi.org/10.1016/j.cptl.2015.08.001.

[Nick Cox]

Different people do different things here, as you have found. One agnostic answer is to try both and see how much difference it makes to modelling success. Also, the pattern of coefficient estimates if you treat a Likert item as categorical should be informative.

Detail: Please see advice in the FAQ on giving more than bare name (date) references.

[Eleonora Facibeni]

Thank you Nick Cox