Skip to main content

Ordinal Independent Variables

By: Richard A. Williams | Edited by: Paul Atkinson, Sara Delamont, Alexandru Cernat, Joseph W. Sakshaug & Richard A. Williams Published: 2020 | Length:   3 | DOI: |
+- LessMore information
Download PDF

As Richard A. Williams and Christopher Quiroz (2019) noted, there are several procedures that can be used when the dependent variable in an analysis is ordinal. Possible methods include the ordered logit model, the generalized ordered logit model, interval regression, and stage models.

However, researchers also often want to use ordinal measures as independent/explanatory variables in their models. Rightly or wrongly, it is very common to treat such variables as continuous—or, more precisely, as having interval-level measurement with linear effects. When the items use a Likert-type scale (e.g., strongly disagree, disagree, neutral, agree, strongly agree), this may be a reasonable or at least not too controversial practice.

However, many ordinal items use categories that clearly are not equally spaced—for example, the options might be something like “daily,” ...

Looks like you do not have access to this content.

Copy and paste the following HTML into your website