If you have ever stared at a correlation coefficient in your SPSS output and wondered how to turn it into a sentence your supervisor will accept, you are not alone. Reporting a correlation sounds simple — it is just one number, after all — but examiners and reviewers notice every missing detail: the degrees of freedom, the exact p value, and whether you described the direction and strength of the relationship correctly. This guide shows you how to report Pearson and Spearman correlation results in APA 7th edition style, with worked examples you can adapt to your own thesis.
What APA Style Actually Requires
APA style asks you to give readers everything they need to judge the result without going back to your raw data. For a correlation, that means four pieces of information in every report: the type of correlation coefficient, the degrees of freedom, the value of the coefficient itself, and the p value. You should also state the sample size somewhere nearby, usually in the participants section or in the sentence itself, and describe the direction of the relationship in plain words. A correlation on its own, without context, tells the reader almost nothing.
The general template looks like this: r(df) = value, p = value. The letters are italicised in a finished manuscript: r for Pearson’s correlation, and the degrees of freedom in parentheses. For Pearson’s r, the degrees of freedom equal the sample size minus two (N – 2), because two parameters are estimated from the data. Many students write the sample size in the parentheses by mistake, so double-check this number before you submit.
The Anatomy of a Correlation Sentence
A good APA correlation sentence has a predictable shape. First, name the two variables. Second, say whether the relationship was positive or negative. Third, give the statistic in parentheses or in the sentence flow. Fourth, interpret the strength in words. Consider this example: “There was a strong positive correlation between hours of study and exam scores, r(48) = .52, p < .001.” In one sentence the reader learns the variables, the direction, the strength, the statistic, and the significance level. That is the standard you are aiming for.
Notice two formatting details. APA style omits the leading zero before a correlation coefficient, because a correlation can never exceed 1, so you write r = .52, not r = 0.52. The same rule applies to p values and to Cronbach’s alpha. Also note that the p value is written as p < .001 when it is very small, rather than p = .000, because SPSS prints .000 only as a rounding artefact — a p value is never literally zero.
Step 1: Run the Right Correlation
Before you write anything, make sure you ran the correct test. Pearson’s r is for two continuous variables with a roughly linear relationship, for example revision time and test performance. Spearman’s rho (written as rs) is the rank-based alternative, used when your data are ordinal, such as Likert-scale responses, or when the relationship is monotonic but not linear, or when outliers distort the Pearson result. If one variable is dichotomous, such as pass or fail, and the other is continuous, the correct coefficient is the point-biserial correlation, although many researchers simply report it as r and name the variables clearly. Choosing the wrong coefficient is the most common reason a supervisor sends a results section back for revision, so check your assumptions — linearity, normality of the variables, and absence of extreme outliers — before you report.
Step 2: Worked Example — Pearson’s r
Imagine you surveyed 50 undergraduate students. You want to know whether the number of hours they studied each week relates to their final exam score. You run a Pearson correlation in SPSS and obtain r = .52, p < .001. Your sample size is 50, so the degrees of freedom are 50 – 2 = 48.
Here is how to report it in your results section: “A Pearson correlation was conducted to examine the relationship between weekly study hours and final exam scores. There was a strong positive correlation between the two variables, r(48) = .52, p < .001. As weekly study hours increased, final exam scores also tended to increase.”
That is the complete report for a single correlation. You do not need to add the means and standard deviations in the same sentence, but it is good practice to report them in a nearby table or in the descriptive statistics section so readers can see the scale of each variable.
How to Describe the Strength of a Correlation
Readers need a word, not just a number, for how strong the relationship is. The most widely used benchmarks come from Jacob Cohen’s conventional guidelines: an absolute value of around .10 is small, around .30 is medium, and around .50 or above is large. These are rough guides, not laws of nature — a correlation of .25 can be highly meaningful in a field where variables are hard to measure, and a correlation of .60 can be trivial if it merely confirms the obvious. Use Cohen’s labels as a starting point, then explain in your own words why the size of the relationship matters for your research question.
Also report the direction carefully. A positive correlation means that as one variable increases, the other tends to increase as well. A negative correlation means that as one variable increases, the other tends to decrease. Sign errors are surprisingly common when students copy numbers from SPSS output, so read the sign twice before you write the sentence.
Reporting a Non-Significant Correlation
Not every correlation you run will be significant, and that is fine — a non-significant result is still a result. The format is the same, but you report the exact p value rather than a threshold. For example: “There was no significant correlation between daily social media use and sleep quality, r(78) = .14, p = .21.” Notice the exact p value of .21, the word “no significant correlation” rather than “no correlation” (the sample correlation was .14, after all), and the absence of any claim that the variables are unrelated in the population. A non-significant result means your study did not find convincing evidence of a relationship, not that you proved there is none.
Worked Example — Spearman’s Rho
Now suppose you asked students to rank their confidence in statistics on a five-point scale and compared it with their rank in the class. Because the confidence measure is ordinal, you run a Spearman correlation and obtain rs = .38, p = .004, with 60 students. The degrees of freedom are reported in the same way, as N – 2 = 58.
Report it like this: “A Spearman rank-order correlation was conducted to examine the relationship between self-rated statistics confidence and class rank. There was a moderate positive correlation between the two variables, rs(58) = .38, p = .004.” If you are reporting several Spearman correlations, you can write rs throughout, and readers will understand that you used the rank-based method. Some journals prefer the Greek letter rho, but rs is the standard APA form in student work and is unambiguous.
Reporting Several Correlations in a Table
When you have more than two or three correlations, a sentence for each one becomes unreadable. APA style then calls for a correlation matrix table: variables down the rows and across the columns, with the correlation coefficients in the cells and asterisks marking significance levels. Under the table, add a note explaining the asterisks, for example: “Note. *p < .05. **p < .01. ***p < .001.” In the text, highlight only the correlations that matter for your argument and refer readers to the table for the rest, like this: “As shown in Table 1, study hours correlated positively with exam scores, whereas test anxiety correlated negatively with both variables.” A table never replaces interpretation — it supports it.
Effect Size and Shared Variance
A correlation is itself an effect size, which is convenient: you do not need a separate statistic. You can also square it to get the coefficient of determination, r-squared, which tells you the proportion of variance in one variable that is shared with the other. In our first example, r = .52 gives r-squared = .27, meaning the two variables share about 27% of their variance. Reporting r-squared is optional in APA style, but it helps readers who think in percentages, and supervisors often appreciate it. Write it as: “The two variables shared 27% of their variance (r-squared = .27).” Do not confuse this with a claim about causation — shared variance is still just association.
Common Mistakes to Avoid
First, do not write “the correlation proves that X causes Y.” Correlation never establishes causation on its own, no matter how large the coefficient or how small the p value. Use language like “associated with” or “related to” rather than “leads to” or “improves.”
Second, do not round your coefficient to one decimal place. Report correlations to two decimal places, as in r = .52, so that readers can judge the strength accurately.
Third, do not report p = .000. SPSS prints this when the p value rounds to zero at three decimal places; the correct APA form is p < .001.
Fourth, do not forget the degrees of freedom. A bare statement like “r = .52, p < .001” is incomplete, because the reader cannot tell how large your sample was.
Fifth, do not report a Pearson correlation for clearly non-linear data. If your scatterplot curves, Pearson’s r will underestimate the relationship — check the plot first and switch to Spearman if the pattern is monotonic but curved.
Quick Template You Can Copy
Use this template and fill in your own numbers: “A Pearson correlation was conducted to examine the relationship between [variable 1] and [variable 2]. There was a [weak/moderate/strong] [positive/negative] correlation between the two variables, r([N – 2]) = [value], p = [value].” For Spearman, replace “Pearson correlation” with “Spearman rank-order correlation” and r with rs. Keep one copy of this template beside you while you write your results chapter, and every correlation you report will follow the same clean format.
Frequently Asked Questions
Do I report the degrees of freedom for a correlation?
Yes. For Pearson’s r and Spearman’s rs, the degrees of freedom are N – 2, where N is your sample size. For example, with 50 participants you write r(48). Without the degrees of freedom, the reader cannot evaluate the result properly.
Should I write r = 0.52 or r = .52?
Write r = .52. APA style drops the leading zero for statistics that cannot exceed 1, which includes correlations, p values, and proportions. This is one of the easiest formatting marks to lose, so check every coefficient.
What if my correlation is significant but very small?
Report it honestly and interpret it in context. A correlation of .15 with a large sample can be statistically significant yet practically unimportant. Say so plainly: describe the effect as small, and explain whether it still matters for theory or practice rather than hiding behind the p value.
Can I report a correlation from a table without a sentence?
You can, but you should still discuss the key correlations in the text. Tables present the numbers; the text tells the reader which ones matter and why. A results section that is only tables reads like an appendix, not an argument.
Final Checklist Before You Submit
Before you send your results section to your supervisor, run through this checklist. Have you named both variables? Have you stated the direction and strength of the relationship? Have you given the correct coefficient, the degrees of freedom as N – 2, and the p value in APA format? Have you avoided causal language? And have you checked the scatterplot for linearity and outliers? If you can answer yes to all five, your correlation reporting is ready. Getting these details right is a small effort that signals careful, professional work — exactly the impression you want your results chapter to make.