In general, there are two different methods for obtaining Rasch measures from a
dataset using Likert-scale items: employing (1) a Rating Scale Model (RSM), or (2)
a Partial Credit Model (PCM). In the RSM, one single, consistent, and unchanging
measurement scale is assumed to characterize all items in an instrument (Wu et al.,
2007). The PCM, on the other hand, originates from a desire to make use of the
advantages offered by Rasch analysis. This model assumes that different measurement
scales characterize different items in an instrument. That is, it is not assumed
that each pair of adjacent categories (e.g., strongly agree, agree) is equidistant from
one another among all items. Thus, PCM takes into account scale differences that
may occur among items. The PCM is mostly used when ‘richer data than the dichotomous
data that are typically generated by traditional assessment practices are available’
(Wu et al., 2007, p. 3).