|
Activity Number:
|
249
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #310379 |
|
Title:
|
Methodological Aspect of Interpretation of Principal Components
|
|
Author(s):
|
Lev Sverdlov*+
|
|
Companies:
|
|
|
Address:
|
200 Baxter Blvd 3, Portland, ME, 04101,
|
|
Keywords:
|
Factor analysis ; Principal components ; Interpretation ; Content ; Meaning
|
|
Abstract:
|
This report describes a strategy facilitating translation of statistical results of principle components analysis (PCA) into a language of content and meaning (i.e., "surmising" the content and meaning of the statistically identified principal component). The strategy includes 1) presenting variables in a comparative way ("younger" vs. "older" instead of "age"); 2) the development of two contrasting (positive and negative) but logically equivalent presentations of each retained component; 3) sorting absolute values of loadings in a descending order, and 4) contrasting positive and negative presentations of the factor. The graphical presentation can be helpful for the interpretation. The strategy is illustrated by the results of studies (Sverdlov et al., 1993, 2004). Psychological and philosophical aspects of the interpretation of PC are discussed.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2007 program |