Day 17: Factor Analysis in SPSS – Identifying Patterns and Reducing Variables

Day 17: Factor Analysis in SPSS – Identifying Patterns and Reducing Variables

Welcome to Day 17 of your 50-day SPSS learning journey! Today, we’ll explore Factor Analysis, a statistical technique that helps uncover hidden patterns in your data and reduce the number of variables while retaining meaningful information. Factor analysis is widely used in survey design, psychometrics, and exploratory research.

What is Factor Analysis?

Factor analysis identifies underlying dimensions (factors) that explain the patterns of correlations among a set of observed variables. Instead of analyzing each variable individually, factor analysis groups related variables into common factors.

For example, in a customer satisfaction survey with questions about service quality, responsiveness, and value, factor analysis might reveal an overarching factor such as "overall satisfaction."

Types of Factor Analysis

  1. Exploratory Factor Analysis (EFA):

    • Used to explore the underlying structure of data without prior hypotheses.
    • Helps identify how many factors exist and which variables load onto each factor.

  2. Confirmatory Factor Analysis (CFA):

    • Used to test specific hypotheses about factor structure (requires specialized tools like SPSS AMOS).

When to Use Factor Analysis?

Use factor analysis when:

  • You have a large number of variables and want to reduce dimensionality.
  • You suspect that several variables are measuring the same underlying construct.
  • You’re developing or validating a scale or questionnaire.

Key Concepts in Factor Analysis

  1. Communalities:

    • Represents the proportion of each variable's variance explained by the extracted factors.

  2. Eigenvalues:

    • Indicates the amount of variance explained by each factor. Factors with eigenvalues > 1 are typically retained.

  3. Factor Loadings:

    • Shows how strongly each variable correlates with the factor (ranges from -1 to +1).

  4. Rotation:

    • Simplifies the factor structure to make it easier to interpret (e.g., Varimax for orthogonal rotation).

How to Perform Factor Analysis in SPSS

Step 1: Open Your Dataset

For this example, use the following survey dataset:

ID Service Responsiveness Price Quality Convenience Loyalty
1 4 5 3 5 4 4
2 3 4 4 4 3 3
3 5 5 4 5 4 5
4 2 3 3 3 2 2
5 4 4 5 4 5 4

Step 2: Access the Factor Analysis Tool

  1. Go to Analyze > Dimension Reduction > Factor.
  2. A dialog box will appear.

Step 3: Select Variables

  1. Move the variables (Service, Responsiveness, Price, Quality, Convenience, Loyalty) to the Variables box.

Step 4: Customize Extraction Options

  1. Click Extraction:
    • Select Principal Component Analysis (PCA) for extraction.
    • Check Scree Plot to visualize eigenvalues.
    • Retain factors with eigenvalues > 1.
  2. Click Continue.

Step 5: Apply Rotation

  1. Click Rotation:
    • Select Varimax (orthogonal rotation) for easier interpretation.
  2. Click Continue.

Step 6: Run the Analysis

Click OK to generate the factor analysis output.

Interpreting the Output

1. Total Variance Explained Table

  • Shows the eigenvalues and the percentage of variance explained by each factor.
  • Retain factors with eigenvalues > 1.
  • Example: If two factors explain 75% of the total variance, they represent most of the information in the dataset.

2. Scree Plot

  • A graphical representation of eigenvalues.
  • Look for the "elbow" in the plot where eigenvalues level off; factors before this point are retained.

3. Rotated Component Matrix

  • Shows factor loadings (correlations between variables and factors).
  • Example: If Service, Responsiveness, and Quality load highly onto one factor, you might label it "Customer Experience."

Example Output

Suppose you run the analysis and get the following results:

  1. Total Variance Explained Table:

    • Factor 1 explains 45% of the variance.
    • Factor 2 explains 30% of the variance.
    • Combined, they explain 75% of the variance.

  2. Rotated Component Matrix:

Variable Factor 1 (Customer Experience) Factor 2 (Price & Convenience)
Service 0.85 0.10
Responsiveness 0.82 0.15
Quality 0.80 0.12
Price 0.12 0.78
Convenience 0.20 0.84
Loyalty 0.75 0.18

Interpretation:

  • Factor 1 (Customer Experience): Includes Service, Responsiveness, Quality, and Loyalty.
  • Factor 2 (Price & Convenience): Includes Price and Convenience.

Practice Example: Perform Factor Analysis

Use the following dataset:

ID Question_1 Question_2 Question_3 Question_4 Question_5 Question_6
1 4 5 4 3 2 3
2 3 4 3 3 3 4
3 5 5 5 4 4 5
4 2 3 3 2 2 3
5 4 4 4 4 5 5
  1. Perform an exploratory factor analysis using Question_1 to Question_6.
  2. Extract factors with eigenvalues > 1.
  3. Use Varimax rotation and interpret the Rotated Component Matrix.

Common Mistakes to Avoid

  1. Over-Retaining Factors: Retain only factors with eigenvalues > 1 or those before the scree plot elbow.
  2. Ignoring Factor Loadings: Review loadings to ensure variables meaningfully contribute to factors (e.g., > 0.4).
  3. Misinterpreting Factors: Use domain knowledge to label factors meaningfully.

Key Takeaways

  • Factor analysis identifies underlying dimensions in your data and reduces complexity.
  • Retain factors based on eigenvalues, variance explained, and scree plots.
  • Use rotation techniques (e.g., Varimax) to simplify interpretation of factor loadings.

What’s Next?

In Day 18 of your 50-day SPSS learning journey, we’ll explore Cluster Analysis in SPSS. You’ll learn how to group cases based on similarities and uncover hidden patterns in your data. Stay tuned for another essential exploratory tool!


Labels: