Factor Analysis - Factor Analysis (StudeerSnel)

8 important questions on Factor Analysis - Factor Analysis (StudeerSnel)

What is step 1? (requirements)

  1. Sample size: 4/5 per variable with a minimum of 50 variables.
  2. Measurement properties: interval or ratio
  3. Missing values: <10%

Difference between exploratory and confirmatory?

  • Exploratory: find underlying structure, reveal interrelationships, generate hypothesis.
  • Confirmatory: a prior idea, testing of hypothesis, testing relationships before factor analysis.

What is step 2? (preparations)

  1. Skewness and kurtosis: between -3 and 3
  2. Multicollinearity
    • Correlation matrix: >0,3 or <-0,3
    • Bartlett's test of sphericity: Significant when there are correlations between the variables
  3. Sample adequacy: Kaiser-Meyer Olkin >0,5 (higher is better)
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

What is step 3? (determine method)

  • Principal component analysis: data reduction: total variance considered
  • Common factor analysis: data summarization: error variance considered

What is step 4? (determine number of factors)

  1. A priori determination: based on prior knowledge/theory
  2. Based on eigenvalues: factors with variance >1 included
  3. Based on the percentage of variance: should be account for at least 60%
  4. Based on scree plot: bending point of the plot


2, 3, and 4 are most often used

What is step 5? (rotate factors)

  • Orthogonal (varimax): axes stay in right (90 degrees) angle. If correlation >-0,3 or <0,3
  • Oblique (oblimin): axes change. If correlation <-0,3 or >0,3

What is step 6? (delete variables)

After running the data, the set needs to be cleaned up
  1. Factor loading: the correlation between a variable and a factor, this is visible in the table of commonalities at extraction. 


Rule --> delete correlations <0,3

2. Cross loader: variable that correlates with two or more factors.


Rule --> there is cross-loading when the differences between the highest factor loading of a variable on a factor, and the second-highest factor loading of a variable on a factor is smaller than <0,2. The variable with the highest cross load should be deleted.

What is step 7? (interpret factors)

Run down final criteria:
  • KMO and Bartlett's sufficient: KMO >0,5 & Bartlett's: significant
  • Communalities: all above 0,2
  • No cross loadings: Difference <0,2
  • Explained variance: >60%
  • Factor loadings: minimal 0,3 (350) or 0,4 (200)

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo