What are the best practices for presenting regression analysis results in quantitative research publications?

1 Choose the right model

Before you present your results, you need to choose the right regression model for your research question and data. Depending on the type and level of your variables, the nature and size of your sample, and the assumptions and hypotheses you want to test, you may opt for different types of regression models, such as linear, logistic, or multilevel. You should justify your choice of model based on theoretical and empirical grounds, and check for potential issues such as multicollinearity, heteroskedasticity, endogeneity, or outliers. You should also report any tests or corrections you performed to ensure the validity and robustness of your model.

2 Use tables wisely

Tables are an essential tool for presenting regression analysis results in quantitative research publications. They allow you to display the coefficients, standard errors, significance levels, and other statistics of your model in a concise and organized format. However, if not well-designed and explained, tables can be overwhelming and confusing. To ensure effective use of tables, make sure to follow the style and conventions of your journal or discipline. Only include relevant and necessary information, avoiding cluttering with too many rows, columns, or decimals. Label variables, models, and statistics clearly, providing definitions and sources in the notes or captions. Highlight key results and findings with symbols, colors, or fonts to indicate the significance and magnitude of your coefficients. Finally, provide a descriptive title for each table and a brief summary of its main message in the text.

3 Supplement with graphs

Graphs can be a great way to supplement tables by providing a visual representation of regression analysis results. They can help illustrate the relationship between variables, the distribution of data, the fit and prediction of your model, and the effect and confidence of coefficients. However, graphs can also be misleading and distracting if not well-crafted and integrated. To ensure your graphs are effective, use appropriate and consistent types of graphs for your data and model, such as scatter plots, line charts, bar charts, or histograms. Make sure to include clear and accurate scales, axes, labels, and legends on your graphs. Avoid unnecessary or distorted elements, such as 3D effects, gridlines, or backgrounds. Additionally, emphasize the key results and findings with symbols, colors or lines to show the regression line, confidence intervals or marginal effects. Finally, provide a descriptive title for each graph and a brief explanation of its main purpose and implication in the text.

4 Write clearly and concisely

When presenting your regression analysis results in quantitative research publications, it is not enough to rely on tables and graphs. You need to write clearly and concisely about your results in order to interpret and discuss their meaning, significance, and implications. Avoid simply repeating the numbers and statistics from your tables and graphs; instead, explain what they tell you about your research question, hypothesis, and theory. To write clearly and concisely, use plain language that avoids jargon, acronyms, or technical terms. Use active and direct voice to state your results, transitioning and connecting them logically. Finally, provide evidence and citations to support your results; do not make claims or conclusions that are not backed up by your data or literature.

5 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?