Statistical Methods

Factor Analysis

Uncover the hidden structure in your data. Identify clusters, reduce complexity, and find the latent variables that matter.

What Is Factor Analysis?

It is a statistical method used to uncover the hidden structure in a set of variables. It helps researchers identify clusters of related items (called factors) and reduce complex data into simpler, interpretable dimensions.

"Factor analysis identifies latent variables that explain patterns of correlations within observed variables."
Methodologies

Types of Factor Analysis

Exploratory (EFA)

Discover underlying factor structure without prior assumptions.

✅ SPSS

Confirmatory (CFA)

Test if data fit a hypothesized model you already have in mind.

⚠️ AMOS R Mplus

Key Concepts

Factor Loadings

Correlation between a variable and a factor (-1 to +1).

Threshold: > 0.40

Eigenvalues

Indicate variance explained by each factor.

Rule: Keep > 1

Communality

How much of a variable's variance is explained by the extracted factors.

Rotation

Optimizes structure for interpretation.

e.g. Varimax
Visual Analysis

The Scree Plot

The "elbow" point helps determine how many factors to retain. Factors above the break point are usually retained, while those in the rubble are discarded.

Interact with the graph points to see eigenvalues.

Pre-Analysis Checklist

SPSS Workflow

1. Navigate

Analyze > Dimension Reduction > Factor

2. Settings

Choose Extraction (Principal Axis) and set Rotation to Varimax.

3. Review

Check the Rotated Component Matrix.

Interpreting Output

Rotated Component Matrix

Item Factor 1 Factor 2
Q1 .74 .10
Q2 .70 .12
Q3 .13 .81
Q4 .18 .77
Q1 & Q2 load on Factor 1, Q3 & Q4 on Factor 2
Reduce items Identify constructs Validate structure Prep for reliability