Software

Linear Regression in Jamovi: Step-by-Step Guide for Thesis Students

Q: Can Jamovi run multiple regression with more than two predictors?

Yes. Add all predictors to the Covariates box. Jamovi automatically includes all in a single model. Check Adjusted R² (not R²) to evaluate model fit - R² always increases when you add predictors, even if they add no meaningful information.

Q: What is the difference between B and β in Jamovi regression output?

B (unstandardised coefficient) shows the raw change in Y for a 1-unit increase in X. β (standardised coefficient) shows the same relationship in standard deviation units, making it comparable across predictors with different scales. Report B for interpretability and β for comparing the relative importance of predictors.

Q: How do I report a non-significant regression predictor?

Report it the same way as a significant predictor, but note it is not significant. Example: 'Study hours was not a significant predictor of exam score, B = 0.12, β = .08, p = .412.' Never omit non-significant predictors from your results table - selective reporting is considered poor practice.

Q: What sample size do I need for linear regression in my thesis?

A common guideline is at least 10–20 participants per predictor variable. For a thesis with 3 predictors, you need at least 30–60 participants. A power analysis using G*Power (F tests → Linear Multiple Regression) will give the exact required N for your effect size and desired power (80% is standard).

Q: Can Jamovi run logistic regression for binary outcomes?

Yes. Go to Analyses → Regression → Logistic Regression. Use this when your outcome variable is binary (pass/fail, yes/no, 0/1). Jamovi produces odds ratios - report them as: 'For each unit increase in X, the odds of [outcome] increase by a factor of [OR].'

April 20265 min read

Jamovi makes linear regression as fast as clicking through three menus. But fast execution and correct execution are different things. This guide shows you exactly how to run simple and multiple linear regression in Jamovi - from loading your data to interpreting the output and copy-pasting APA-formatted results into your thesis.

Free sample chapter

Data Analysis From Survey to Results

Step-by-step guidance for choosing the right test, running it, and writing up APA results - in plain language, not theory. Get the free sample chapter when you join the waitlist.

Statistical Analysis Step by Step book cover

Key takeaways

Setup: outcome (Y) → Dependent Variable box; predictors (X) → Covariates (metric) or Factors (categorical).
R² vs Adjusted R²: use Adjusted R² when you have more than one predictor - R² always increases with added variables.
B vs β: report B for practical interpretation; report β to compare the relative strength of multiple predictors.
Always check the Q-Q plot (residual normality) and Cook's distance (influential outliers) before interpreting results.
VIF > 10 in collinearity statistics signals serious multicollinearity - remove or combine the affected predictors.

When to Use Linear Regression in Your Thesis

Linear regression answers the question: does variable X predict variable Y, and by how much?

Use simple linear regression when you have one predictor variable.
Use multiple linear regression when you have two or more predictors.
Use logistic regression when your outcome variable is binary (yes/no, pass/fail) - Jamovi supports this too.

Your outcome variable (Y) must be metric (continuous, measured on an interval or ratio scale). Predictors can be metric or dummy-coded categorical variables.

Running Linear Regression in Jamovi: Step by Step

Step 1: Open your data file (File → Open).
Step 2: Click Analyses → Regression → Linear Regression.
Step 3: Drag your outcome variable (Y) to the Dependent Variable box.
Step 4: Drag predictor(s) to the Covariates box (metric) or the Factors box (categorical).
Step 5: Click Model Coefficients → check 'Standardised estimate' to get β values.
Step 6: Click Assumption Checks → check 'Residuals Q-Q plot' and 'Cook's distance'.
Step 7: Click Model Fit → check R² and Adjusted R².

Understanding the Jamovi Regression Output

Jamovi produces three key output tables. Here is what each tells you and what to report in your thesis.

Output	What It Tells You	Report in Thesis
R²	% of Y variance explained by your model	R² = .42 (42% of variance explained)
Adjusted R²	R² corrected for number of predictors	Use for multiple regression
F (model)	Whether the model is statistically significant	F(df1, df2) = X.XX, p = .XXX
B (unstandardised)	Raw slope: Y change per 1-unit X increase	Report for practical interpretation
β (standardised)	Slope in SD units - comparable across predictors	Report to compare predictor importance
p (per predictor)	Whether each predictor is significant	p = .XXX for each predictor

Checking Regression Assumptions in Jamovi

Run these checks before interpreting your results:

Normality of residuals: Check the Q-Q plot under Assumption Checks. Points should follow the diagonal line closely.

Homoscedasticity: The fitted values vs. residuals plot should show no funnel shape - equal spread across all values of X.

Multicollinearity (multiple regression only): Under Assumption Checks → Collinearity statistics. VIF below 5 is fine; above 10 is a problem.

Outliers: Cook's distance values above 1.0 (some use 0.5) may be influential outliers worth investigating.

⚠️

Never interpret regression coefficients without running assumption checks first. A single high-leverage outlier can flip the sign of your β coefficients.

APA Format for Reporting Regression Results

Copy-paste template for simple linear regression:

"A simple linear regression was conducted to examine the relationship between [predictor] and [outcome]. The model was significant, F(1, 98) = 24.3, p < .001, R² = .20. [Predictor] was a significant predictor of [outcome], B = 0.54, β = .45, p < .001, indicating that a one-unit increase in [predictor] was associated with a 0.54-unit increase in [outcome]."

For multiple regression, report each predictor separately:

"Multiple linear regression revealed that [predictor 1] (β = .38, p = .003) and [predictor 2] (β = .22, p = .041) significantly predicted [outcome], while [predictor 3] did not (β = .09, p = .234). The model accounted for 34% of the variance in [outcome], R² = .34, F(3, 96) = 16.4, p < .001."

Frequently asked questions

Can Jamovi run multiple regression with more than two predictors?

▾

Yes. Add all predictors to the Covariates box. Jamovi automatically includes all in a single model. Check Adjusted R² (not R²) to evaluate model fit - R² always increases when you add predictors, even if they add no meaningful information.

What is the difference between B and β in Jamovi regression output?

▾

B (unstandardised coefficient) shows the raw change in Y for a 1-unit increase in X. β (standardised coefficient) shows the same relationship in standard deviation units, making it comparable across predictors with different scales. Report B for interpretability and β for comparing the relative importance of predictors.

How do I report a non-significant regression predictor?

▾

Report it the same way as a significant predictor, but note it is not significant. Example: 'Study hours was not a significant predictor of exam score, B = 0.12, β = .08, p = .412.' Never omit non-significant predictors from your results table - selective reporting is considered poor practice.

What sample size do I need for linear regression in my thesis?

▾

A common guideline is at least 10–20 participants per predictor variable. For a thesis with 3 predictors, you need at least 30–60 participants. A power analysis using G*Power (F tests → Linear Multiple Regression) will give the exact required N for your effect size and desired power (80% is standard).

Can Jamovi run logistic regression for binary outcomes?

▾

Yes. Go to Analyses → Regression → Logistic Regression. Use this when your outcome variable is binary (pass/fail, yes/no, 0/1). Jamovi produces odds ratios - report them as: 'For each unit increase in X, the odds of [outcome] increase by a factor of [OR].'

Not sure which statistical test to use?

Answer 5 quick questions about your research design and get the right test - with an explanation of why - in under two minutes.

Statoria Team

Statistics educators & software developers

We build Statoria to help bachelor and master students get through their thesis data analysis without stress. Our guides are written by researchers with experience in social science statistics and student supervision.

Related guides

Test selection