Correlation Analysis
Correlation measures the strength and direction of the relationship between two continuous variables. The most common measure is Pearson's correlation coefficient (r).
Interpreting Pearson's r
| r Value | Strength | Direction |
|---|---|---|
| 0.00 – 0.19 | Very Weak | Positive (+) = variables move together Negative (-) = variables move inversely |
| 0.20 – 0.39 | Weak | |
| 0.40 – 0.59 | Moderate | |
| 0.60 – 0.79 | Strong | |
| 0.80 – 1.00 | Very Strong |
SPSS Steps: Pearson Correlation
- Analyze → Correlate → Bivariate
- Move variables of interest to the "Variables" box
- Ensure Pearson is checked
- Check Two-tailed test of significance
- Check Flag significant correlations
- Click OK
Reading the Correlation Matrix
| Satisfaction | Work Environment | Development | |
|---|---|---|---|
| Satisfaction | 1 | .654** | .487** |
| Work Environment | .654** | 1 | .523** |
| Development | .487** | .523** | 1 |
** Correlation is significant at the 0.01 level (2-tailed)
The strongest relationship is between Satisfaction and Work Environment (r = .654, strong positive correlation).
Simple Linear Regression
Regression goes beyond correlation — it predicts the value of one variable based on another.
Example Hypothesis:
"Work environment has a statistically significant impact on job satisfaction."
SPSS Steps:
- Analyze → Regression → Linear
- Move the dependent variable (Satisfaction) to "Dependent"
- Move the independent variable (Work_Environment) to "Independent(s)"
- Click OK
Reading Regression Output
1. Model Summary
| R | R Square | Adjusted R Square |
|---|---|---|
| .654 | .428 | .422 |
R² = .428 means work environment explains 42.8% of the variance in job satisfaction.
2. ANOVA Table
Check the Sig. value. If < .05, the model is significant.
3. Coefficients Table
| Model | B | Beta | t | Sig. |
|---|---|---|---|---|
| (Constant) | 1.234 | 3.456 | .001 | |
| Work Environment | .672 | .654 | 8.234 | .000 |
B = .672: For every 1-unit increase in work environment score, satisfaction increases by 0.672 units.
Beta = .654: Standardized coefficient showing the relative importance.
Multiple Regression
To test the impact of multiple independent variables simultaneously, add more variables to the "Independent(s)" box. Use the Stepwise method to find the best predictors.
Important: Correlation does NOT imply causation. A significant correlation between two variables does not mean one causes the other.