Visualize Variable Relationships at a Glance

A correlation matrix displays Pearson correlation coefficients (r) between all pairs of variables in a dataset. It helps identify multicollinearity, explore patterns, and guide further statistical modelling.

  • Displays Pearson r values between all variable pairs simultaneously
  • Colour-coded cells reveal relationship strength and direction instantly
  • Star notation (*) indicates statistical significance levels (p < .05, .01, .001)
Pearson Correlation Interpretation
.70 to 1.00
Strong positive correlation
.40 to .69
Moderate positive correlation
.10 to .39
Weak positive correlation
-.09 to .09
Negligible / No correlation
-.40 to -.69
Moderate negative correlation
-.70 to -1.00
Strong negative correlation

Creating a Correlation Matrix in Your Software

Step-by-step instructions for common statistical software packages

1
Open Bivariate Correlations

Navigate to Analyze → Correlate → Bivariate from the main SPSS menu bar.

2
Move Variables to the List

Select all continuous variables and move them to the Variables box on the right side.

3
Select Pearson & Flag Significance

Ensure Pearson is checked as correlation coefficient. Tick "Flag significant correlations" to show asterisks.

4
Click OK and Interpret Output

Review the correlation matrix in the Output window. Look for asterisks indicating p < .05, .01, or .001.

1
Load Required Packages

Install and load corrplot for visualisation and psych for significance testing.

install.packages(c("corrplot", "psych"))
library(corrplot)
library(psych)
2
Compute the Correlation Matrix

Use the cor() function with pairwise deletion for missing data.

cor_matrix <- cor(your_data, use = "pairwise.complete.obs")
3
Visualize with corrplot

Create a colour-coded heatmap with correlation coefficients displayed.

corrplot(cor_matrix, method = "color", type = "upper",
addCoef.col = "black", tl.cex = 0.8)
4
Test Significance with psych

Use corr.test() to obtain p-values for each correlation coefficient.

cor_results <- corr.test(your_data)
cor_results$r # correlation matrix
cor_results$p # p-value matrix
1
Enable the Analysis ToolPak

Go to File → Options → Add-ins → Manage Excel Add-ins → check Analysis ToolPak → OK.

2
Open Correlation Tool

Click Data tab → Data Analysis → Correlation → OK to open the dialog box.

3
Select Your Data Range

Select the cell range containing your variables. Check "Labels in first row" if headers are included.

4
Format with Conditional Formatting

Apply a 3-color scale conditional format (Home → Conditional Formatting → Color Scales) to visualise correlation heatmap.

Types of Correlation Relationships

Visual examples of different correlation strengths and directions

r = +0.82
Strong Positive
Strong Positive Correlation

As one variable increases, the other consistently increases. Points cluster tightly around an upward-sloping line.

r = +0.51
Moderate Positive
Moderate Positive Correlation

A clear upward trend exists but with more scatter. The relationship is visible but not perfect.

r = -0.47
Moderate Negative
Moderate Negative Correlation

As one variable increases, the other tends to decrease. The downward trend is discernible but scattered.

r = +0.06
Negligible
Negligible / No Correlation

No systematic relationship between variables. Points appear randomly scattered with no trend.

Tip: Correlation does not imply causation. Always consider potential confounding variables.