If you’re working with survey results, test scores, sales numbers, or any dataset with numeric variables, correlation is often the first analysis you run. It helps you understand whether two variables move together and how strongly they’re related. This guide walks you through how to do correlation in SPSS step by step, using the most common method: bivariate correlations (especially Pearson’s correlation). You’ll also learn how to interpret correlation matrix outputs, and how to read a pearson r table, pearson correlation coefficient table, and pearson correlation table without confusion. Get expert help with correlation
What correlation means
Correlation measures the strength and direction of a relationship between two variables:
Direction
Positive correlation: as X increases, Y tends to increase.
Negative correlation: as X increases, Y tends to decrease.
Strength
Values closer to +1 or -1 indicate a stronger relationship.
Values near 0 suggest little to no linear relationship.
The most common statistic you’ll see in SPSS is Pearson’s r (the Pearson correlation coefficient).
Before you start in SPSS: check your data
To get reliable results, do these quick checks:
Variable type
Pearson correlation is best for scale/continuous variables (interval/ratio).
2. Missing values
SPSS can exclude missing values pairwise or listwise; your choice affects results.
3. Outliers
Extreme values can inflate or hide correlations.
4. Linearity
Pearson’s r captures linear relationships. If the relationship is curved, r may be low even if variables are clearly related.
If your data is ordinal (ranked) or not normally distributed, you may prefer Spearman’s rho. But since most users want Pearson in SPSS, we’ll focus there.
How to do correlation in SPSS (step-by-step)
Here’s the exact path in SPSS to run correlations:
Step 1: Open your dataset
Go to File > Open > Data and select your file (e.g., .sav, .xlsx, .csv).
Step 2: Go to Bivariate Correlations
Click Analyze > Correlate > Bivariate…
This is the standard workflow for bivariate correlations in SPSS.
Step 3: Choose variables
In the left panel, select the variables you want to correlate.
Click the arrow to move them into the Variables box.
Tip: You can select more than two variables. SPSS will output a full correlation matrix.
Step 4: Select correlation type
In the correlation coefficients section, choose:
✅ Pearson (most common)
Optional:Spearman (rank-based)
Kendall’s tau-b (often for small samples/ordinal data)
Step 5: Choose significance options
Test of Significance
Two-tailed is most common (checks for any relationship, positive or negative).
One-tailed only if you have a strong directional hypothesis.
Tick Flag significant correlations so SPSS marks significant results with asterisks (*).
Step 6: Handle missing data
Under “Missing Values,” choose:
Pairwise: uses all available data for each pair (more data retained, but N can vary).
Listwise: uses only cases with no missing values across selected variables (consistent N, but may reduce sample).
Step 7: Run the analysis
Click OK.
SPSS will output a correlation table (often called a pearson correlation table or pearson r table) in the Output Viewer.
Understanding the SPSS output: correlation matrix explained
When you include multiple variables, SPSS gives a correlation matrix—a table showing correlations for every pair of variables.
This section is the core of how to interpret correlation matrix results.
What you’ll see in the correlation matrix
For each variable pair, SPSS usually reports:
Pearson Correlation
This is the r value (ranges from -1 to +1).
Sig. (2-tailed)
This is the p-value. It tells you whether the correlation is statistically significant.
N
The number of cases used for that correlation.
So, the pearson correlation coefficient table in SPSS is basically telling you:
“How strong is the relationship (r), is it significant (p), and how many data points were used (N)?”
How to read a pearson r table correctly
A pearson r table (or pearson correlation table) can look intimidating, but it’s repetitive and predictable.
Key rules to remember
The diagonal values are always 1.000 because a variable correlates perfectly with itself.
The table is symmetrical:
Correlation of X with Y equals correlation of Y with X.
SPSS may mark significance with:
* for p < .05
** for p < .01
Example interpretation (simple)
If you see:
Pearson Correlation = 0.62
Sig. (2-tailed) = 0.003
N = 50
You can interpret it as:
There’s a moderate-to-strong positive correlation between the variables.
It’s statistically significant (p = 0.003).
Based on 50 observations.
Bivariate correlations: what they are and when to use them
Bivariate correlations simply mean you are analyzing the relationship between two variables at a time. In SPSS, the Bivariate Correlations procedure can still produce a matrix for many variables, but the logic remains pair-by-pair.
When bivariate correlations are useful
Checking relationships before regression
Screening variables for multicollinearity
Exploring survey constructs (e.g., satisfaction vs. loyalty)
Validating assumptions (e.g., “does income rise with education?”)
There’s no single universal rule for “weak” vs “strong,” but these practical guidelines are common:
0.00 to 0.19: very weak
0.20 to 0.39: weak
0.40 to 0.59: moderate
0.60 to 0.79: strong
0.80 to 1.00: very strong
Use the absolute value to judge strength (ignore the sign for strength).
The sign (+/-) only tells direction.
Common mistakes when using Pearson correlation in SPSS
Avoid these errors to keep your correlation analysis accurate:
1) Confusing correlation with causation
A correlation does not prove that one variable causes the other. A third variable could influence both.
2) Ignoring outliers
Outliers can heavily distort Pearson correlations. Always check scatterplots.
3) Using Pearson on ordinal categories
If your data is ranked or Likert-style with limited categories, consider Spearman (though Pearson is often used with Likert scales in practice).
4) Interpreting non-significant correlations as meaningful
If p is high (e.g., p = .40), don’t treat r as a reliable relationship in that sample.
Reporting correlation results
Here’s a clean way to report results in writing:
Example:
“A Pearson correlation was conducted to assess the relationship between study time and exam score. Results showed a significant positive correlation, r = .62, p = .003 (two-tailed), N = 50.”
This is exactly what your Pearson correlation coefficient table supports.
Extra tip: Create a scatterplot to support your interpretation
If you want to strengthen your analysis:
Go to Graphs > Chart Builder
Choose Scatter/Dot
Select a Simple Scatter
Put one variable on X-axis and the other on Y-axis
Scatterplots help confirm whether the relationship is linear and whether outliers are present.
FAQ
What is the difference between a pearson r table and a correlation matrix?
A pearson r table often refers to the output table listing Pearson correlations. A correlation matrix is that same idea when multiple variables are included (a grid of all pairwise correlations).
What does “Sig. (2-tailed)” mean in SPSS?
It’s the p-value testing whether the correlation differs from zero in either direction (positive or negative).
Can SPSS run correlations for more than two variables?
Yes—select multiple variables in the bivariate correlations dialog, and SPSS outputs a full matrix.
Final checklist
Use Analyze > Correlate > Bivariate
Select your variables
Choose Pearson
Use Two-tailed (usually)
Decide on pairwise vs listwise missing values
Interpret the output using:
r (direction + strength)
p-value (significance)
N (sample size)
Support results with a scatterplot if needed
If you follow these steps, you’ll not only know how to do correlation in SPSS, but you’ll also confidently explain how to interpret correlation matrix outputs and read any pearson correlation table like a pro.