Class notes 9 Dec 2004
Last class! Overview of factor
analysis...
EPSY
6306
9 Dec 2004
Path analysis takes multiple regression one step further, it is looking for causality
Factor analysis, you are looking for structure, you are looking for DVs
historically for publication, if significance was not found many things were not published
- this trap esp happened in psychology
When you collect data, always go to the highest level of measurement you can!
Intro to Factor Analysis
you can measure attitudes, behaviors, and beliefs
- when you measure attitudes, that is not as good because they can change day to day
- behaviors and beliefs are more permanent, better to measure
stay away from surveys like computers to attitudes
instruments usually have a theoretical background (a theory that explains it)
enduring traits, behaviors, beliefs, are what we are often looking for in research
basic idea of factor analysis: if 2 item are highly correlated, they must represent the same phenomenon, they tell us about the same underlying variance, combining them to form a single measure is reasonable for 2 reasons:
- parsimony (simplicity of the phenomenon)
- reduction in error
9 Dec 2004
Path analysis takes multiple regression one step further, it is looking for causality
Factor analysis, you are looking for structure, you are looking for DVs
historically for publication, if significance was not found many things were not published
- this trap esp happened in psychology
When you collect data, always go to the highest level of measurement you can!
Intro to Factor Analysis
you can measure attitudes, behaviors, and beliefs
- when you measure attitudes, that is not as good because they can change day to day
- behaviors and beliefs are more permanent, better to measure
stay away from surveys like computers to attitudes
instruments usually have a theoretical background (a theory that explains it)
enduring traits, behaviors, beliefs, are what we are often looking for in research
basic idea of factor analysis: if 2 item are highly correlated, they must represent the same phenomenon, they tell us about the same underlying variance, combining them to form a single measure is reasonable for 2 reasons:
- parsimony (simplicity of the phenomenon)
- reduction in error
Posted: Fri - December 10, 2004 at 10:47 AM |
Quick Links
Calendar
| Sun | Mon | Tue | Wed | Thu | Fri | Sat |
Categories
Archives
XML/RSS Feed
Get More Info
Comments and Trackbacks powered by HaloScan.com.
Statistics
Total entries in this blog:
Total entries in this category: 12
Published On: Dec 23, 2004 02:45 AM
Total entries in this category: 12
Published On: Dec 23, 2004 02:45 AM
