Analyses

Moderation

Test whether one predictor’s slope depends on a second predictor.

Edit on GitHub

When to use it

Use moderation when you think the slope of X changes across levels of M. For example: does an extra study hour buy more exam points when anxiety is low than when anxiety is high? The test is the product term, hours × anxiety.

Assumptions

Same straight-line model as linear regression, plus the product term. Tensr centers the predictors by default before building that product, so the hours slope is the slope at average anxiety.

Running it in Tensr

Analyze → Regression → Moderation. In chat: “Does anxiety moderate the effect of hours on score?”
Outcome, predictor, and moderator are three numeric columns.
Center variables starts on. Confidence level starts at 0.95.

Options

Prop

Type

Reading the output

Hours predicting exam score, with practice quizzes as the moderator. Practice was built with no link to the slope, so the interaction is the result to read.

TermBSEp
Intercept71.3720.815< .001
hours2.1080.706.004
practice-0.2470.851.772
hours × practice-0.3070.655.641
Moderator levelSlope
Low M (−1 SD)2.405
Mean M2.108
High M (+1 SD)1.811

The interaction between hours and practice is not statistically significant at the 95% level, suggesting moderation is not supported. The primary result is not significant (p = .641). This is not large enough to treat the comparison this page is about as a reliable association. The result is non-significant: the data are still compatible with no effect. Metrics: R² = 0.11; Adj R² = 0.081; F = 3.776; p (model) = .013; p (interaction) = .641.

Reporting (APA 7)

The interaction between hours and practice is not statistically significant at the 95% level, suggesting moderation is not supported. This result is not significant (p = .641). Report the estimate with that p, and do not describe the pattern as a reliable effect.

Coming from SPSS

Analyze → Regression → Moderation.

SPSS does not ship a single Moderation dialog in the base menus. The usual path is Linear Regression with a computed product term. Tensr builds that product and the simple slopes.

Entering hours and anxiety without a product term is linear regression.