s-news
[Top] [All Lists]

Advice on factor analysis

To: S-news <s-news@lists.biostat.wustl.edu>
Subject: Advice on factor analysis
From: "Walter R. Paczkowski" <dataanalytics@earthlink.net>
Date: Thu, 26 Feb 2004 09:05:34 -0600
User-agent: Mozilla/5.0 (Windows; U; Win 9x 4.90; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0 (nscd2)
Hello,

I need some advice on factor analysis, a topic that's not a strong point for me. First, some background: I have a client who has a questionnaire that asks respondents to rate on a 5-point scale their intent to see a movie. The questionnaire also asks them to rate on a 5-point scale their level of agreement/disagreement with 28 statements about movies in general . I'd like to factor analyze the 28 statements to get, say, 4 factors and then use those four in an ordinal regression model for the intent question. My questions are: 1. Can I use the factor scores as explanatory variables in the ordinal regression? 2. How do I interpret a factor score (not the factor itself)? For example, the SPLUS manual for 6.0 (V. 2, page 81) shows a predicted factor score of 1.45 and then -0.0011 for factor 1 for two people for the math scores example. What does 1.45 and -0.0011 mean? How do I interpret something like this for my client, and me?

Thanks for any help,

Walt Paczkowski



<Prev in Thread] Current Thread [Next in Thread>
  • Advice on factor analysis, Walter R. Paczkowski <=