Scribd
Upload
2K views3 pages

Steps in Conducting Factor Analysis

Factor analysis involves 4 main steps: 1) computing a correlation matrix of all variables, 2) extracting initial factors using principal component analysis, 3) rotating factors to achieve a simple structure, and 4) making a final decision on the number of underlying factors. In the first step, variables are examined to ensure they are correlated and appropriate for factor analysis. Statistical tests like Bartlett's test of sphericity and the Kaiser-Meyer-Olkin measure of sampling adequacy are used. In the second step, principal component analysis extracts initial factors to estimate the number of underlying factors based on eigenvalues above 1 and a scree plot analysis.

Uploaded by

nantzsi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
2K views3 pages

Steps in Conducting Factor Analysis

Factor analysis involves 4 main steps: 1) computing a correlation matrix of all variables, 2) extracting initial factors using principal component analysis, 3) rotating factors to achieve a simple structure, and 4) making a final decision on the number of underlying factors. In the first step, variables are examined to ensure they are correlated and appropriate for factor analysis. Statistical tests like Bartlett's test of sphericity and the Kaiser-Meyer-Olkin measure of sampling adequacy are used. In the second step, principal component analysis extracts initial factors to estimate the number of underlying factors based on eigenvalues above 1 and a scree plot analysis.

Uploaded by

nantzsi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd

Steps in Factor Analysis

Factor analysis usually proceeds in four steps:


1st Step : the correlation matrix for all variables is computed
2nd Step : Factor extraction
3rd Step : Factor rotation
4th Step : Make final decisions about the number of underlying factors
1st Step : the correlation matrix
enerate a correlation matrix for all variables
!dentify variables not related to other variables
!f the correlation bet"een variables are small# it is unlikely that they share common
factors $variables must be related to each other for the factor model to be
appropriate%&
'hink of correlations in absolute value&
(orrelation coefficients greater than )&3 in absolute value are indicative of
acceptable correlations&
*xamine visually the appropriateness of the factor model&
Bartlett Test of Sphericity:
used to test the hypothesis the correlation matrix is an identity matrix $all
diagonal terms are 1 and all off+diagonal terms are )%&
!f the value of the test statistic for sphericity is large and the associated
significance level is small# it is unlikely that the population correlation
matrix is an identity&
!f the hypothesis that the population correlation matrix is an identity cannot be
re,ected because the observed significance level is large# the use of the factor model
should be reconsidered&
The Kaiser-eyer-!lkin "K!# measure of sampling ade$uacy:
is an index for comparing the magnitude of the observed correlation
coefficients to the magnitude of the partial correlation coefficients&
'he closer the -M. measure to 1 indicate a si/eable sampling ade0uacy $&1
and higher are great# &2 is acceptable# &3 is mediocre# less than &4 is
unaccaptable %&
5easonably large values are needed for a good factor analysis& Small -M.
values indicate that a factor analysis of the variables may not be a good idea&
%nd Step : Factor extraction
'he primary ob,ective of this stage is to determine the factors&
!nitial decisions can be made here about the number of factors underlying a
set of measured variables&
*stimates of initial factors are obtained using &rincipal components
analysis&
'he principal components analysis is the most commonly used extraction
method & .ther factor extraction methods include:
Maximum likelihood method
6rincipal axis factoring
7lpha method
8n"eighted lease s0uares method
enerali/ed least s0uare method
!mage factoring&
!n principal components analysis# linear com'inations of the observed variables are
formed&
'he 1st principal component is the combination that accounts for the largest amount of
variance in the sample $1st extracted factor%&
'he 2nd principle component accounts for the next largest amount of variance and is
uncorrelated (ith the first $2nd extracted factor%&
Successive components explain progressively smaller portions of the total sample variance#
and all are uncorrelated "ith each other&
'o decide on ho" many factors "e
need to represent the data# "e use 2
statistical criteria:
)igen *alues+ and
'he Scree &lot&
'he determination of the number of
factors is usually done by considering
only factors "ith *igen values greater
than 1&
Factors "ith a variance less than 1 are
no better than a single variable# since
each variable is expected to have a
variance of 1&
'he examination of the
Scree plot provides a visual
of the total variance
associated "ith each factor&
'he steep slope sho"s the
large factors&
'he gradual trailing off
$scree% sho"s the rest of the
factors usually lo"er than an
*igen value of 1&
!n choosing the number of
factors# in addition to the
statistical criteria# one should make initial decisions based on conceptual and theoretical
grounds&
7t this stage# the decision about the number of factors is not final&

You might also like