40-1 EFA - 2019 - Upld

EXPLORATORY FACTOR
ANALYSIS
DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

AARHUS UNIVERSITY
MANAGEMENT OCTOBER 2019 ASSISTANT PROFESSOR
AGENDA
2x 45 min- 15 min break
• How do we develop scales?

• Taxonomy of scales
• Latent/reflective
• Formative
• Profile
• Exploratory factor analysis
• What does it do?
• What is the procedure?
• Difficult decisions
• Example
• How do I report it?

AARHUS UNIVERSITY
Where do constructs come from?
To the extent that a variable is abstract and latent rather than concrete and
observable (such as the rating itself), it is called a “construct.” Such a variable is
literally something that scientists “construct” (put together from their
own imaginations) and which does not exist as an observable
dimension of behavior… Nearly all theories concern statements about
constructs rather than about specific, observable variables because constructs are
more general than specific behaviors by definition.
(MacKenzie, S., Podsakoff, P., Podsakoff, N., Construct measurement and validation procedures in MIS
and behavioral research: integrating new and existing techniques, p. 298)

AARHUS UNIVERSITY
EFA- MAIN JOB

AARHUS UNIVERSITY
DEVELOPING CONSTRUCT
• Exploratory factor analysis is crucial tool

FOR CONSTRUCT CREATION
• Selection of appropriate items/questions
• DEFINE WHAT YOU WANT TO MEASURE

• GENERATE QUESTIONS = ITEMS
• DECIDE WHAT TYPE OF MEASURE IT IS
• COLLECT DATA
• EFA

AARHUS UNIVERSITY
MEASUREMENT- DIMENSIONALITY
Construct dimensionality
• Not every construct can be captured by one dimension, many have distinguishable facets that in composite measure the
original construct of interest
• Firms performance- operational excellence, customer relationships, revenue growth
• The sub-dimensions can be (sub)constructs themselves!
HOW DO WE DECIDE?
Structure of the entire construct and relationship between the dimension
• Sub-dimensions are manifestation of the overall construct or defining characteristics of it?
• Does the construct exist separately on a deeper level than its sub-dimensions?
• Is the change with the overall construct associated with change in all of the sub-dimensions or is it possible that it is
associated only with one/few of them?

AARHUS UNIVERSITY
Taxonomy of multidimensional constructs (according to Law, Wong, Mobley, 1998)
YOU DECIDE! NOT SPSS
LATENT
AGGREGATE
PROFILE

AARHUS UNIVERSITY
LATENT MODEL
• The main construct exists at a deeper level than the sub-constructs=dimensions
• Higher-order abstraction- commonality among the dimensions
• The sub-dimensions are reflected in the items, the main construct is reflected in the sub-dimensions
EXAMPLE- OCB
• Organizational citizenship behaviour- displaying extra-role behaviour at work (not officially required
to do and not paid for)- individuals that display high OCB will be scoring high on sub-dimensions
which are altruism, sportsmanship, civic virtue, courtesy and conscientiousness, these sub-
dimensions are captured via reflective constructs
• OCB exists at a deeper level than the sub-dimensions and we expect an individual who scores high
on OCB to score high on all 5 sub-dimensions (they are related) as they together reflect/manifest this
personal quality of the employee

AARHUS UNIVERSITY
AGGREGATE MODEL
• The main construct is formed from its sub-dimensions
• The main construct does not exist separately on a deeper level
• Each dimension can weigh in on the entire construct differently or equally, but taking one away would
severely impact the meaning of the overall construct (that is the totality of all the sub-dimensions)
• We do not expect a person that scores high on the overall construct to score necessarily high on each sub-
dimension
EXAMPLE- JOB SATISFACTION

• Job satisfaction- being highly satisfied with ones job, associated with low turnover intentions is a
composition of satisfaction with pay, co-workers, work itself, supervisor and promotion.
• Job satisfaction exist on the same level as the separate sub-dimensions which are satisfaction with the
different elements of the job and we do not expect that an individual who scores high on job satisfaction
necessarily scores high on all 5 sub-dimensions. It is possible that person satisfied with job overall actually
dislikes the supervisor but likes the pay, the co-workers and the job itself.

AARHUS UNIVERSITY
PROFILE MODEL
• Dimensions are at the same level as the construct but the constructs cannot
be expressed as a function of these dimensions
• Researchers therefore create profiles that have some specific levels of the
dimensions
• Theoretical reason for combinations
EXAMPLE- TYPOLOGY OF GLOBAL LEADERSHIP

• Global leaders can be categorized based on the complexity they have to
face
• High/high, High/low, Low/low, Low/high = different type of leader
• Used for recruitment- we cannot just say score high/how on each
dimension- combination of the dimensions create a unique profile

AARHUS UNIVERSITY
EXPLANATORY FACTOR ANALYSIS
• Selecting the final set of items and in settling on the final structure of the measure
• Identify coherent subset of variables
• Helps to discover structure within the construct- can we see the subdimensions that we have defined theoretically?
• Factors= representatives of the subdimensions or constructs
• Often it is reiterative process
BOTH PRINCIPAL COMPONENT ANALYSIS AND EFA SIMPLIFY DATA- DATA REDUCTION
BUT THEY ARE NOT THE SAME METHOD AND HAVE DISTINCT
PURPOSE!!!
• Components- just simplified data
• Factors- theoretically determined- they are operationalizations of the underlying processes

AARHUS UNIVERSITY
EFA
• Mathematically factor is a linear combination of the observed variables

• It is essentially explorative technique- there is no objectively one best/most
correct solution
• It is the researcher that needs to make the final decision on which rotation is
the final one, and therefore how the factors will be defined and interpreted e1_1= 0.32 * factor + ε1
• Factor loadings (the strength of association between observed variable and the e1_2= 0.47 * factor + ε2
factor) help to define the factor- give it meaning- help to name the pattern we e1_3= 0.52 * factor + ε3
have uncovered e1_4= 0.27 * factor + ε4
• Interpretability and stability of the factors is therefore how we can judge e1_5= 0.42 * factor + ε5
the final “quality” of the analysis
Observed
variables/items
Factor loadings
Measurement error for each

observed variable
AARHUS UNIVERSITY
EFA- CLEAN UP STRUCTURE
YOU NAME THE

NEW VARIABLE
= CONSTRUCT

AARHUS UNIVERSITY
WHAT DOES IT DO?
Repackage correlation matrix through eigenvalues/eigenvectors to factor loadings
• Eigenvalue is collapsed variance from above and under diagonal in correlation matrix
• Through eigenvalues and eigenvectors we calculate factor loadings
• Factor loadings connect invisible factors to visible questions = items

AARHUS UNIVERSITY
EFA- THE PROCEDURE
• PREPARE- Checking assumptions and the correlation matrix of the items we have selected
• PERFORM- extract factors- the core task is to repackage the correlation matrix and find the factors and its association
with each variable- factor loadings
• Initially we have as many factors as variables
• We extract only the interesting factors and force all items to be expressed through them
• ROTATE- Rotation of factors to increase interpretability-we want to find which items belong to which factor- where
is the highest factor loadings, we then delete variables that had low association with the retained factors
• INTERPRETATION- Interpret the results, save factors for future analysis- ending up with clear structure (no
overlap between factors) we interpret the factors (name the underlying pattern) and save the factor scores for future
analysis

AARHUS UNIVERSITY
PERFORMING EFA- PREPARE
Sample size and study design
• Factor should be over-determined- at least 3-5 items for each factor
Sample size and missing data/suitability of the data

• Rules of thumb- 5 obs. - 10 obs. per item, but these are not set in stone
• Sample size depends on whether the correlations are expected to be strong in the population and whether we expect
the factors to be strong and independent- smaller sample will be ok (specific cut off points in Fabrigar et.al, 1999)
• 100 and less- well-determined factors
• 100-200 obs- well-determined factors with lower common correlation between variables
• 300 obs- medium level of common correlation
• 500 obs- low level of correlations and a lot of poorly defined factors

AARHUS UNIVERSITY
PERFORMING EFA - PREPARE
Check assumptions and factorability of R

Normality/case and variable outliers
• Not a necessity especially if we simply want to reduce data, but enhances solution
• Normality tests in SPSS are quite sensitive for large samples- Kolmogorov-Smirnov (>2000 obs) Shapiro-Wilk (<2000 obs)
• Values of skewness and kurtosis should be close 0 (+/- 2)
Outliers
• Visual tools- histograms, q-q plots, box plots
• Variables with low squared multiple correlation and low correlation with the factors are outliers and should be removed (you’ll find them
during the analysis)
Linearity, multicollinearity but NOT singularity

• Correlations (our starting point) measures linear relationships, so we should have only variables/items that relate to each other in a linear
way, even very strong correlation- muticollinearity is fine
• Singularity (or extreme multicollinearity) should be avoided

AARHUS UNIVERSITY
PERFORMING EFA- PREPARE
Factorability of R
• Check correlation matrix for correlations higher than 0,3

• In linear algebra, the matrix is singular if and only if its determinant is zero
• There should be patterns of correlations- so more than a pair of variables
correlated- check partial correlation matrices
• SPSS- anti-image correlation matrix- to check the partial correlations between
variables
• Bartlett’s test of sphericity testing whether the correlations in correlation matrix
are zero, too sensitive
• Kaiser’s measure of sampling adequacy- ratio of sum of squared correlations-
SSC/(SSC-SSPartialC)- so if partial correlations are small the ratio approaches 1
• Values above 0.6 acceptable

AARHUS UNIVERSITY
PERFORMING EFA- PERFORM
Factor-extraction procedure (available in SPSS)- different methods how to estimate the parameters of factor model- factor loadings and
unique variances
• Principal components- tries to explain all variance in data not separate what items have in common and what is unique and error
• Principal axis factoring- estimates communalities- squared multiple corr of the items- tries to explain just this variance with factors
Maximum likelihood- estimates the factor loadings and specific variances for each variable through maximizing the log likelihood
function

AARHUS UNIVERSITY
CRUCIAL DECISION- PCA OR EFA
• Mathematically the difference is in the positive

diagonal of the correlation matrix that will be
analysed, for PCA it will be 1, since we are
interested in capturing all the variance in the data,
in EFA we are interested only in capturing the
variance that is shared among the variables (SPSS
doesn’t show this- don’t get confused)
• In PCA each variable adds equally to the

variance studied, if we retain all components of
PCA we can reproduce backwards the
correlation matrix of the real data perfectly

AARHUS UNIVERSITY
EFA OR PCA
FROM:
https
://stats.idre.ucla.edu/spss/seminars/efa-
spss/

AARHUS UNIVERSITY
CRUCIAL DECISION- PCA OR EFA
• In EFA the positive diagonal of correlation matrix has therefore communalities (squared multiple correlation of the
variable) on the positive diagonal- in SPSS you can see them as initial communalities
• All communalities add up to the variance that is studied by EFA, which is less than the total variance of the data,
therefore factors do not retain the totality of the data and cannot reproduce back the entire correlation matrix only
approximation of it
• Communalities
Initial communalities in EFA are squared multiple correlations of each variable (shared variance with other variables), in
PCA it is not calculated it is set to 1 (100%) so all the variance of the variable
Extracted communalities (in EFA and PCA)- sum of squared loadings for a variable across all extracted factors
(components), also shows how much variance on that particular variable has been explained by the extracted factors
(components)

AARHUS UNIVERSITY
CRUCIAL DECISION- NUMBER OF
FACTORS Selecting number of factors
• Balancing the need for parsimony with the need for accuracy- essentially we should retain factors until additional
factors account for a very small amount of variance, their added value is minimal
• Underfactoring (not extracting enough factors) is usually introducing more problems than overfactoring
• When overfactoring, the main factors remain stable over several rotations, so the smaller unstable factors can be
removed
Rules of thumb?
• Kaiser criterion- we should keep all the factors that have eigenvalues higher than 1
• Scree plot- eigenvalues plotted against factors, shows how much variance is explained in each factor, usually ordered in
descending order, so we are looking for the turning point when the slope changes
• Variance explained- the total factors that are retained should be explaining together about 60% of variance and more

AARHUS UNIVERSITY
CRUCIAL DECISION- NUMBER OF
FACTORS SPSS is cheating a bit… it displays initial eigenvalues for factor analysis like it was PCA and retains automatically factors based on PCA
eigenvalues (and it does this in screeplot too), that is why in PCA you will see no difference between initial and extraction % of variance
explained, but there will be difference in EFA

AARHUS UNIVERSITY
CRUCIAL DECISION ROTATION
• For any given solution with two or more factors, there exists an
infinite number of alternative orientations of the factors in
multidimensional space
• Simple structure
• Factor loadings are then coordinates marking the position of
the variable in relation to the factor- they therefore change
during rotation
• How much variance factors explain does not change though
• Point of rotation is to change the factor loadings so they fulfil the
criteria of simple structure

AARHUS UNIVERSITY
PERFORM- ROTATION
Orthogonal
• Varimax- simplify columns of loading matrix by maximizing variance of loadings on each factor , the spread of
loadings within factor should be as wide as possible, high loadings after extraction become higher after rotation
and vice versa.
• Quartimax- simplify rows of loading matrix by maximizing variance of loadings within variables
• Equimax- combination of the above that tries to simplify both rows and columns of the loading matrix
Oblique rotation
• Oblimin- minimizes cross-products of loadings you can choose the level of correlation between the factors by
choosing the level of Delta
• Promax- an orthogonal solution is rotated again to allow correlations among factors, the orthogonal loadings are
raised to powers (2,3,6)
• It is not guaranteed that the factors are correlated and although you allow for oblique rotations, if there is little
correlation between factors, the solution will be very similar to orthogonal rotation

AARHUS UNIVERSITY
PERFORM- ROTATION
• Rotations create new matrices

• Orthogonal rotation- interpret the rotated matrix
• Oblique rotation- Instead of rotated factor matrix we get structure matrix which is a product of pattern and factor
correlation matrix
• The pattern matrix is representing the “clean” amount of unique variation that the factor explain for each variable
• The structure matrix includes the shared variance caused by the overlap between factors
• Interpret the pattern matrix because it is easier and consider variables that have higher loading than 0.32, best
would actually be higher than 0.71

AARHUS UNIVERSITY
HELP- PLOT
• Plot the factors
• Can help to see if you should
use different rotation
• Extract extra factor
• We want variables at the end of
the axis not inbetween
• Axis should go through the
variable clusters
• Variables in the center do not
distinguish either of the factors
well

AARHUS UNIVERSITY
MOST CRUCIAL- INTERPRETATION
• Pattern matrix- clean association between factors and items
• Look at the items that have the highest association
• Marker item- higher than 0.8
• Interpret the results- give factors name
• Decide on the structure

AARHUS UNIVERSITY
SAVE FACTOR SCORES
• We are doing EFA so we find out how to represent best our constructs
• We can select the “best items” and simply calculate sum scales or average
score out of them
• Directly save factor scores after finding EFA solution
• These are all estimation techniques to assign a score to every individual on the
underlying construct that is now variable in its own right
• Regression- new variable mean of 0, variance equal to SMC between
factors and variables
• Barlett
• Anderson-Rubin- standardized variable

AARHUS UNIVERSITY
MENTI-QUESTIONS
Students want to capture attitudes of consumers to bubble gum- they ask following questions in their survey:
Agree/disagree on 7-point Likert scale with following statements
TASTE
• This gum tastes like heaven in my mouth
• This gum is disgusting (R)
• I love the taste of this gum
• This gum tastes great even after some fish and onion
BRAND
• I find the commercials for this gum amusing
• This packaging always grabs my attention
• I feel proud to take out this brand of gum out of my pocket before some kissing action
PRICE
• This gum is more expensive than toothpaste
• I need to save up to buy this gum
• I try to limit the use of this gum due to price

AARHUS UNIVERSITY
MENTI-QUESTIONS
What is this?

AARHUS UNIVERSITY
EFA-EXAMPLE
Data on 1406 academic employees in Scandinavia
41% females
Average age 38
16 ITEMS FOR FACTOR ANALYSIS (attitudes towards job- 7-point Likert scale, anchors- “strongly disagree/strongly agree”)
• I have felt uneasy and nervous as a result of my job. • When I am working, I forget everything else around me.
• My job gets to me more than it should. • Time flies when I am working.
• There are lots of times when my job drives me up the wall. • I get carried away when I am working.
• I feel relaxed when I am at work. • It is difficult to detach myself from my job.
• Sometimes when I think of my job I get a tight feeling in my chest. • I am immersed in my work.
• I feel guilty when I take time off from my job. • I feel happy when I am working intensely.
• We have too much work and too little time to do it.
• Working makes it hard for team members to spend time with their family.
• We have enough time for our work.
• Working here leaves little time for other activities.

AARHUS UNIVERSITY
TYPE OF ANALYSIS
What type of analysis are we performing? ASSUMPTIONS
• We assume that there are two underlying factors that should be part of one multi-dimensional construct, there is also one more factor that
should be only uni-dimensional construct distinguished from the multi-dimensional construct
• We assume that two factors of multi-dimensional construct should correlate, the uni-dimensional construct should not correlate
highly with the other factors
Find structure
• Exploratory factor analysis not PCA is therefore chosen
• Oblique rotations should be selected due to expected correlation
Sample size and missing data/suitability of the data
• 16 items that should predict 3 factors (advised 3-5)
• Missing data was deleted- missing at random
Randomly split the sample so other half can be used for confirmation of the found structure

AARHUS UNIVERSITY
CHECK DATA- PREPARE
Linearity/Multicollinearity/Singularity/Factorability of R
• Scatterplots of correlations between variable pairs do not suggest nonlinear relations (F2 highest negative skewness, F11 highest
positive skewness)
• There are some weak correlations in the matrix (but we have more than 500 obs and many items for just 3 expected factors), but
many above 0.3 and significant
• No variable has extremely low correlations with all the others (although there are some bivariate relations close to 0)
• Determinant of the correlation matrix is not zero
• Anti-image correlation matrix- in relation to correlation matrix shows lower values, so there should be some patterns not just
bivariate correlation
• Bartlett’s test of sphericity significant
• Kaiser’s measure of sampling adequacy is above 0,6

AARHUS UNIVERSITY
PERFORMING EFA
Correlation matrix vs. Anti-image matrix

AARHUS UNIVERSITY
FACTOR EXTRACTION
Starting off with principal axis factoring
• Eigenvalues and screeplot would suggest 3 factors
• However cumulative variance would suggest 4-5, plus
eigenvalues would be close to 1 for the 4th and 5th factor
• Forcing extraction of 6 factors, the real cumulative variance
explained is just above 50%
• Other extraction techniques produce minuscule differences
Eigenvalue/trace=
%of variance explained
Trace is the total of eigenvalues
which in this case equal to the
number of variables

AARHUS UNIVERSITY
COMMUNALITIES
• Extracted communality for “when I am working, I forget everything else
around me” = (0.097)^2 + (0.499)^2 + (0.074)^2 = 0.263
• It can be calculated as sum of squared factor loadings for the variables
• Communality is the proportion of the variable’s variance that can be
explained by the factors
• Initial vs extraction
• Unfortunately some of the communalities are quite low (<0.5), but already the
initial communalities are quite low- you can check that the increase in
extracted communalities is very small when we force extraction of more
factors
• Each factor is well defined, although doesn’t have a proper marker variable
with coefficient higher than 0.8, this might change with rotation

AARHUS UNIVERSITY
FACTOR MATRIX Coefficients ordered by strength and lower than 0.3 are supressed
Compare the factor matrix for

solution with forced overfactoring-
in this case the extra factors
(extracted after the third one) would
be weakly defined by low factor
loadings with few variables, better
stay with just 3

AARHUS UNIVERSITY
ROTATION Equamax Quartimax Varimax
• Orthogonal rotation produced very

similar solutions, only the last factor
cross-loaded in quartimax solution
that is designed to achieve variable
rather than factor clarity
• With rotation the loadings are

much clearer- marker variable
• All reversed items have charge as
expected

AARHUS UNIVERSITY
ROTATION
• Oblique rotation- Oblimin (delta=0) has produced again
similar solution
• Pattern matrix shows “clean” factor loadings- good for
interpretation
• Structure matrix shows the shared variance caused by
overlap of factors and just like factor correlation matrix,
shows there is quite high correlation (>0.3) between the 1st
and 3rd factor

AARHUS UNIVERSITY
INFERENCE
Highest loadings:
• Factor 2- “I am immersed in my work.” (0.671)
• Absorption as part of engagement scale by Schaufeli. et.al (2006)
• Factor 1- “There are lots of times when my job drives me up the wall.”
(0.807)
• Anxiety as part of work stress by Parker and DeCotiis’ (1983)
• Factor 3- “We have enough time for our work” (0,811)

• All the other items negative- Work pressure as a part of work stress by
Parker and DeCotiis’ (1983)
• The correlation between factor 1 and 3 indicated that they are part of the same
underlying construct- work stress

AARHUS UNIVERSITY
WHAT IF FORMATIVE CONTRUCT?
• If we believed work stress to be a formative construct, we could use the EFA to see what are the
components within the index- in this case it would be time pressure at work and anxiety
• We decide- formative index composed of two reflective scales? or just index?
• If we decide that pressure and anxiety are adding up to formative overall measure- work stress
• The items related to having time off are connected with anxiety- they should be removed
and two items from anxiety should be removed as well to reduce the overlap between the
two as much as possible
• The level of anxiety and the level of stress then can be added up to create stress index
• If work stress was reflective- feelings of work pressure and anxiety are just reflections of being stressed-
overlapped between them is actually welcomed

AARHUS UNIVERSITY
WHAT TO REPORT?
• Reporting the structure after rotation in terms of factor loadings
• Report perhaps more than one extraction method and rotation method
• If you have over-factored describe the results
• Remember exploratory method- description and transparency vital

• Ultimately you want to showcase stability of the results

AARHUS UNIVERSITY
DEPARTMENT OF MANAGEMENT
AARHUS UNIVERSITY

40-1 EFA - 2019 - Upld

Uploaded by

Copyright:

Available Formats

40-1 EFA - 2019 - Upld

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

40-1 EFA - 2019 - Upld

Uploaded by

Copyright:

Available Formats

EXPLORATORY FACTOR

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

2x 45 min- 15 min break

• How do we develop scales?

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

• Exploratory factor analysis is crucial tool

• DEFINE WHAT YOU WANT TO MEASURE

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

EXAMPLE- JOB SATISFACTION

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

EXAMPLE- TYPOLOGY OF GLOBAL LEADERSHIP

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

• Mathematically factor is a linear combination of the observed variables

Measurement error for each

YOU NAME THE

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

Sample size and missing data/suitability of the data

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

Check assumptions and factorability of R

Linearity, multicollinearity but NOT singularity

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

• Check correlation matrix for correlations higher than 0,3

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

• Mathematically the difference is in the positive

• In PCA each variable adds equally to the

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

• Rotations create new matrices

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

Compare the factor matrix for

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA

• Orthogonal rotation produced very

• With rotation the loadings are

DEPARTMENT OF ANNAMARIA KUBOVCIKOVA