If you’ve ever created a survey using likert scale (“Strongly disagree” → “Strongly agree”), you might get a pile of questions that all seem related—but you’re not sure how they fit together.

That’s where it is factor analysis enter. This is a statistical method that helps you uncover the hidden structure behind your survey items—essentially grouping questions into meaningful categories (called factor).

This guide will walk you through how to run factor analysis step by step.

---

Step 1: Understand the Basics

• What is factor analysis?
  A statistical method that reduces many survey items to a smaller number of factors.
• Why use it?
  To see if your questions measure the same basic concepts. For example, 10 Likert items about “job satisfaction” can actually be grouped into 2 factors: work environment And career growth.

---

Step 2: Prepare Your Data

Before you dive into analysis:

1. Check your sample size. General rule: at least 5–10 responses per item, with a minimum of ~100 participants.
2. Make sure your variables are correlated. Factor analysis works best when the items are moderately correlated.
3. Clean your data. Handle missing values, reverse code negatively worded items, and ensure all Likert responses use the same scale (e.g., 1–5 or 1–7).

---

Step 3: Choose the Right Method

There are two main types:

• Exploratory Factor Analysis (EFA): Use it when you Don’t know previous factor structure.
• Confirmatory Factor Analysis (CFA): Use it when you I already have a theory about how items should be grouped.

💡 For beginners, start with EFF.

---

Step 4: Run Analysis

Here’s how you would do it in practice (example in SPSS, but R, Stata, or Python can do the same):

1. Go to Analysis → Dimensionality Reduction → Factors.
2. Select your Likert scale items.
3. Choose Extraction method (for example, Axis Factoring).
4. Look KMO test (Kaiser-Meyer-Olkin > 0.6 is acceptable).
5. Run Bartlett’s Test of Sphericity (must be significant).
6. Decide how many factors to retain (check for eigenvalues ​​> 1, scree plot, or parallel analysis).
7. Apply rotation (Varimax for independent factors, Oblimin for correlated factors).

---

Step 5: Interpret the Results

• Factor loading: Indicate how strongly each question relates to a factor.
  • In general, loadings above 0.40 are considered significant.
• Cross loading: If an item loads on more than one factor, decide whether to drop or revise it.
• Name the factors: Look at which items are grouped and assign descriptive labels to those groups (e.g., “Work-Life Balance,” “Leadership Support”).

---

Step 6: Validate the Factors

• Inspect Cronbach’s Alpha for internal consistency (α > 0.7 is acceptable).
• If you plan to publish, consider running a Confirmatory Factor Analysis (CFA) on a separate data set to confirm your structure.

---

Example Scenario

Imagine you survey employees with 12 Likert scale questions about job satisfaction. After running EFA, you find:

• Factor 1: Questions about salary, promotions and career growth → “Career Satisfaction”
• Factor 2: Questions about teamwork, leadership, and recognition → “Workplace Support”

Now you can analyze each factor separately instead of having to organize 12 individual items.

---

In short

Performing factor analysis on Likert scale data may sound intimidating, but it is absolutely essential find patterns in your question.

• Use EFF to explore.
• Use rotation to make the results easier to interpret.
• Always check reliability before drawing conclusions.

Once you’ve done it a few times, you’ll find that factor analysis is one of the most powerful tools in your research toolkit.