Beyond single equation regression analysis: Path analysis and multi-stage regression analysis

Research output: Contribution to journalArticlepeer-review

34 Scopus citations

Abstract

Multi-stage regression analysis and path analysis provide important complements to the traditional regression analysis. Although regression (covariance) analysis is a useful and common multivariate analysis methodology in pharmacy and many other sciences, there is a problem of limited measurability that only the direct effects of included independent variables can be captured. Furthermore, the traditional regression analysis might yield biased estimates because of the ignored indirect effects in some cases: the compliance and effectiveness studies; the cost of illness studies; or the patient reported outcomes (PRO) studies Therefore multi-stage regression analysis can provide not only a refinement of established conclusions but also a significant improvement to regression analysis. The main purpose of this paper is to high-light the usefulness of multi-stage regression models and path analysis models in a pharmaceutical research setting. This paper can be also used as an introduction to these two models in a research methodology class.

Original languageEnglish
Pages (from-to)37-42
Number of pages6
JournalAmerican Journal of Pharmaceutical Education
Volume66
Issue number1
StatePublished - 2002

Fingerprint

Dive into the research topics of 'Beyond single equation regression analysis: Path analysis and multi-stage regression analysis'. Together they form a unique fingerprint.

Cite this