ERIC Number: ED449217
Record Type: RIE
Publication Date: 2001-Feb
Why Do We Factor Variables When We Care about Types of People? Q and Other Two-Mode Factor Analytic Models.
This paper explains the types of entities that can be analyzed using two-mode factor analysis and the techniques associated with each variation. It also provides an in-depth review and example of one of the most popular variations, the Q-technique. Even when researchers have multiple variables, multiple people, and multiple occasions of measurement, one of the modes is typically held constant while performing a factor analysis on the other two modes, which results in the data collapsing onto a face of the "data box." In factor analysis, the matrix of associations is computed from a raw data matrix. The rows are replicates and the columns are the mode that is to be factored. There are six types of two-mode factor analytic techniques, designated R, Q, O, P, S, and T. The R-technique is the one traditionally thought of when discussing factor analysis. The O-technique and the P-technique hold the individual mode constant, and all the data are from one individual (or an average of all individuals). The S- and T-techniques are rarely used, and they are only generalizable to one variable. The Q-technique seeks to identify different types of people by factoring people over variables, holding occasions constant. The Q-technique often uses a type of ranking of responses known as the Q sort. The Q sort is explained, and a heuristic example is given of the use of Q-technique. (Contains 5 tables and 18 references.) (SLD)
Publication Type: Reports - Descriptive; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A
Note: Paper presented at the Annual Meeting of the Southwest Educational Research Association (New Orleans, LA, February 1-3, 2001).