This tutorial provides an introduction to conducting exploratory factor analyses (EFAs) using SPSS. The SPSS data, syntax and output is available for each analysis, along with screencasts on youtube.
General steps
Here are the recommended steps when conducting an exploratory factor analysis using SPSS:
Get to know the data:
Univariate distributions: Explore the distribution and central tendency for each of the variables statistically (e.g., use descriptive statistics to examine the M, SD, skewness, and kurtosis) and graphically (e.g., histograms)).
Bivariate distributions: Examine the bivariate correlations statistically (e.g., correlation matrix) and graphically (e.g., scatterplots)
Examine EFA assumptions:
Sample size
Linearity
Homoscedasticity
Factorability
Conduct EFA (repeat until a good model is identified - may require many different analyses involving different numbers of factors, different types of rotation, and different sets of items):
Via Pull-down menus: Analyze - Data Reduction - Factor Analysis - Put target variables into the Variables box (order doesn't matter, but the output will be neater to read if the variables are sequentially ordered from 1 to X)
Descriptives - check these options:
Initial solution
Correlation matrix
Coefficients
KMO and Bartlett's
Anti-image
Extraction - check these options:
Analyse - Correlation matrix
Display - Screen plot
Extract - Eigenvalues over 1 - or usually better is to specify the number of factors
Rotation - Varimax (uncorrelated factors) or Direct Oblimin (correlated factors)
Options - check these options:
Sorted by size
Suppress values less than .20
After checking internal consistency and creating composite scores, work out the correlations between the factors: In SPSS - Analyze - Correlate - Bivariate - and put all of the composite factor scores in.
On a scale from 1 to 100, where 1 represents "strongly disagree" and 100 represents "strongly agree"; please indicate the extent with which you agree with each of the following statements.
Items
1
I think smoking is acceptable.
2
I don't care if people smoke around me.
3
I don't think people should smoke in restaurants.
4
I think people should have the right to smoke.
5
I don't think people should smoke around food.
Solution
PAF, Varimax or Oblimin rotation (5 items; N = 107), 2 factors:
Teachers, for each of the following paired behavioral statements, please mark a cross over the dot which is nearest the statement that best describes the TYPICAL behavior of THIS student at school
Items
1
CONCENTRATES
Cannot concentrate on any particular task; easily distracted ↔ Can concentrate on any task; not easily distracted
2
CURIOUS
Shows little curiousity or motivation ↔ Eager to learn; curious and enquiring
3
PERSEVERES
Lacks perseverance; is impatient with difficult or challenging work ↔ Perseveres in the face of difficult or challenging work
4
EVEN-TEMPERED
Irritable,'touchy', 'cranky' ↔ Even-tempered
5
PLACID
Easily excited; gets 'high' ↔ Not easily excited; placid
6
COMPLIANT
Demanding and argumentative ↔ Patient and compliant
7
SELF-CONTROLLED
Has difficulty controlling own behaviour ↔ Is able to control own behaviour
8
RELATES-WARMLY
Provocative; disruptive; short attention span ↔ Relates warmly to others
9
SUSTAINED ATTENTION
Easily frustrated; short attention span ↔ Persistent, sustained attention span
10
COMMUNICATIVE
Difficult to reason and communicate with ↔ Easy to reason and communicate with
11
RELAXED
Restless; fidgety; can't sit still ↔ Relaxed; can sit still
12
CALM
On the go; lively; always moving ↔ Settled, calm
13
PURPOSEFUL ACTIVITY
Aimless activity ↔ Purposeful activity
14
COOPERATIVE
Disputes, fights over sharing ↔ Co-operative; shared with others
15
CONTENTED
Is easily upset; unhappy ↔ Contented; happy
Solution
Solution: 3-factor PAF, Oblimin rotation (13 items; N = 1500), Drop items 6 and 7
