It is also brilliantly supported by Lütkepohl, H. In contrast, JMulTi is a specialist package designed to close the gap between theoretical developments and availability to practitioners in a narrow range of time-series estimators.
Betraying these origins, it is still patchy in its coverage of time-series econometrics.
Stata is a general econometrics package – the best, in my opinion – but is rooted in cross-section and panel econometrics and medical statistics. An advantage is that you can also analyze polytomous, not just dichotomous items. You end up with a set of items that is valid in the sense of an ordinal measure, and you get different indices on the degree of scalability. The resulting scale satisfies the assumptions of item response analysis, that are also also inherent in Guttman scaling (i.e., unidimensionality, monotonicity). For instance, the MSP module (commands trace, msp, loevh sj 11-1 st0216) performs Mokken Scale Analysis, which aims to evaluate response patterns to scales designed to be cumulative. I am not aware of procedures for specific scalogram analysis implemented in Stata, but there are commands following a very similar goal.
The general purpose of such optimal scaling is to construct a one-dimensional continuum for the concept of interest underlying your set of items. What you are probably referring to by “scalogram analysis” is iirc known also as Guttman or optimal scaling (I think Kristian is also right in calling MCA a synonemous).
0 kommentar(er)
