
80 Ordered Regression Models
the data-generating process. For example, one should choose a cumulative or adjacent
model rather than a stage model to analyze subjective social class because the assumption
of a sequential decision-making process is not applicable. One does not have to identify
with the “lower” or “working” classes before identifying with the “middle” or “upper
middle” classes.
3.6.1 Guidelines for Choosing a Partial or Parallel Ordered Regression Model
Researchers should consider two main factors when deciding whether to use a par-
tial or parallel ordered regression model to analyze an ordinal outcome. First, is there
any ambigu ...