Submit Enquiry
  1. Home
  2. Data Analysis
  3. How to report…

Infographic: How to report correlation and regression analyses in a research paper

INFOGRAPHICS and DOWNLOADABLES New
  • Marisha Fonseca
  • Marisha Fonseca

    An editor at heart and perfectionist by disposition, providing solutions for journals, publishers, and universities in areas like alt-text writing and publication consultancy.

    Mar 21, 2023

Reading time
1 min
How to report correlation and regression analyses in a research paper

Default Alt text

Correlation and regression are powerful tools that can help biomedical researchers better understand the complex relationships between variables in their data, and make more informed decisions about treatments and interventions.

What is correlation analysis?

Correlation analysis is a statistical test that quantifies the relationship between two variables. It tells you

  1. How strong the relation is
  2. What direction the relation is (a positive correlation means that as one variable increases, so does the other; a negative correlation means that as one variable increases, the other decreases).

What is regression analysis?

Regression analysis tells us about how much one or more independent variables (“predictors”) have an effect on a dependent variable (“outcome”), e.g., the effect of obesity on cardiovascular mortality.

What is the purpose of correlation or regression analysis?

Through correlation or regression analyses, researchers can

  • analyze the strength and direction of the relationships among different variables,
  • build predictive models,
  • identify potential confounders, and
  • thereby gain insights into the underlying mechanisms of different diseases, treatments, clinical conditions, or health outcomes.

Hence, it’s important to conduct correlation and regression analyses rigorously and report their results appropriately. Here’s a handy infographic on best practices in reporting these analyses.

Best practices in conducting and reporting correlation and regression analyses 01 For Correlation Analysis Always report direction of the correlation. Describe the correlation as positive or inverse, unless you’re including the correlation coefficient (with the – sign if needed). In your abstract, give the correlation coefficient along with using descriptors like “strong” or “moderate”. This gives readers a precise idea of the strength of your findings. In your Methods section, mention the ranges you use to classify correlation coefficients as strong, weak, or moderate. These ranges may vary slightly among studies, so it’s best to state these explicitly in your paper. Be careful about presentation: The “r” in “Pearson’s r” is usually lowercased, and the “ρ” in Spearman’s rho is the Greek letter rho, not the English lowercase letter “p”. 02 For Regression Analysis Always report the regression equation for both simple and multiple regression analyses. Report a measure of the model’s goodness-of-fit to your data, such as r-squared for simple regression and R-squared for multiple regression. For simple linear regression, you can present the results graphically, in a scatter plot showing the regression line and its confidence bounds. In your scatter plot, do not extend the regression line beyond the minimum and maximum values of your data. If you’re conducting multiple regression, report whether the variables were assessed for collinearity and interaction. Specify how the final model was developed (e.g., forward-stepwise). 03 For Both Check that the data meets the underlying assumptions for each test. For example, Pearson’s r is calculated for continuous parametric variables while Spearman’s rho is calculated for continuous nonparametric variables. Report the statistical software used and your significance threshold in the Methods section. Identify the variables used in the analysis and summarize each with descriptive statistics. Report how you handled outliers or missing data, if applicable.

How to report correlation and regression analyses in a research paper

Section Key Points
Correlation Analysis Always report direction (positive/inverse) or include coefficient with sign.
Include correlation coefficient in abstract with descriptors like “strong” or “moderate”.
Define ranges for strong, weak, and moderate correlations in Methods section.
Use correct notation: lowercase r for Pearson’s; Greek ρ for Spearman’s.
Regression Analysis Report regression equation for simple and multiple regression.
Provide goodness-of-fit measures (r² for simple, R² for multiple regression).
Use scatter plots with regression line and confidence bounds.
Do not extend regression line beyond observed data range.
Report collinearity/interaction checks and model-building approach.
Both Ensure assumptions of tests are met (parametric vs nonparametric).
Report statistical software and significance threshold.
Identify variables and summarize with descriptive statistics.
Explain handling of outliers and missing data.

 

Would you like expert advice from a biostatistician when exploring relationships among your study variables? Editage’s Statistical Analysis & Review Services can help!

Best practices in reporting correlations_0.png

Download

Found this useful?

If so, share it with your fellow researchers

Related post

Data Analysis Planning to Write

What researchers should do BEFORE statistical analysis

May 6, 2026

Powerful ways to tackle missing data

May 14, 2024
Data Analysis Data Storage & Management

How to analyze count data in research

November 14, 2023

Related Reading

Full disclosure: Editage Insights is a product of Editage, a global provider of world-class scientific communication solutions. Editage Insights is funded by Editage and endorses services provided by Editage but is editorially independent. English Editing - Editage.com | 英文校正 – Editage.jp | 원어민영문교정 – Editage.co.kr | SCI英文论文发表 – Editage.cn | publicação de artigos – Editage.com.br | 編輯英文 – Editage.com.tw