m 6
 


     

P g      
    
 


 

m 6

  

 



  
 



   

  

 


m 6  

  

 









 




1. Selecting and Measuring a set of variables in a
given domain
2. Data screening Ȃ Prepare correlation matrix
3. Factor Extraction
4. Factor Rotation to increase interpretability
5. Interpretation
6. Validation and Reliability of the measures
m A good factor:
P Àakes sense
P will be easy to interpret
P simple structure
P Lacks complex loadings
m Ônlike many of the analyses so far
there is no statistical criterion to
compare the linear combination to
P n ÀANOVA we create linear
combinations that maximally
differentiate groups
P n Canonical correlation one linear
combination is used to correlate with
another
m t is more art than science
P ghere are a number of extraction
methods (PCA, FA, etc.)
P ghere are a number of rotation methods
(Orthogonal, Oblique)
P Number of factors to extract
P Communality estimates
P EgCǥ
m ghis is what makes it greatǥ
m Life (researcher) saver
P Often when nothing else can be salvaged from
research a FA or PCA will be conducted
m Exploratory FA
P ©ummarizing data by grouping correlated
variables
P nvestigating sets of measured variables related
to theoretical constructs
P Ôsually done near the onset of research
P ghe type of FA and PCA we are talking about in
this chapter
m Confirmatory FA
P Àore advanced technique
P When factor structure is known or at least
theorized
P gesting generalization of factor structure to new
data, etc.
P ghis is tested through ©EÀ methods discussed in
the next chapter
m Observed Correlation Àatrix
m Reproduced Correlation Àatrix
m Residual Correlation Àatrix
m Orthogonal Rotation
P Loading Àatrix Ȃ correlation between each
variable and the factor
m Oblique Rotation
P Factor Correlation Àatrix Ȃ correlation between
the factors
P ©tructure Àatrix Ȃ correlation between factors
and variables
P Pattern Àatrix Ȃ unique relationship between
each factor and variable uncontaminated by
overlap between the factors
m Factor Coefficient matrix Ȃ coefficients used
to calculate factor scores (like regression
coefficients)
m FA produces factors; PCA produces
components
m Factors cause variables; components are
aggregates of the variables
m ghree general goals: data reduction, describe
relationships and test theories about
relationships (next chapter)
m How many interpretable factors exist in the
data? or How many factors are needed to
summarize the pattern of correlations?
m What does each factor mean?
nterpretation?
m What is the percentage of variance in
the data accounted for by the factors?
m Which factors account for the most variance?
m How well does the factor structure fit a given
theory?
m What would each subjectǯs score be if they
could be measured directly on the factors?
m Hypotheses about factors believed to underlie
a domain
P ©hould have 6 or more for stable solution
m nclude marker variables
P Pure variables Ȃ correlated with only one factor
P ghey define the factor clearly
P Complex variables load on more than on factor and
muddy the water
m Àake sure the sample chosen is spread out on
possible scores on the variables and the
factors being measured
m Factors are known to change across samples
and time points, so samples should be tested
before being pooled together