Hypothesis Testing and Multiplicative Interaction Terms

  title={Hypothesis Testing and Multiplicative Interaction Terms},
  author={Bear F. Braumoeller},
  journal={International Organization},
  pages={807 - 820}
When a statistical equation incorporates a multiplicative term in an attempt to model interaction effects, the statistical significance of the lower-order coefficients is largely useless for the typical purposes of hypothesis testing. This fact remains largely unappreciated in political science, however. This brief article explains this point, provides examples, and offers some suggestions for more meaningful interpretation.I am grateful to Tim McDaniel, Anne Sartori, and Beth Simmons for… 

How Much Should We Trust Estimates from Multiplicative Interaction Models? Simple Tools to Improve Empirical Practice

Replicating nearly 50 interaction effects from 22 recent publications in five top political science journals finds that these core assumptions fail in a majority of cases, suggesting that a large portion of findings across all subfields based on interaction models are modeling artifacts or are at best highly model dependent.


This course will discuss much of the mathematical and statistical theory behind multiple regression and also some potential drawbacks and circumstantial limitations.

Testing for Interaction in Binary Logit and Probit Models: Is a Product Term Essential?

Political scientists presenting binary dependent variable (BDV) models often hypothesize that variables interact to influence the probability of an event, Pr(Y). The current typical approach to

Marginal Effects in Interaction Models: Determining and Controlling the False Positive Rate

When a researcher suspects that the marginal effect of x on y varies with z , a common approach is to plot ∂ y / ∂ x at different values of z along with a pointwise confidence interval generated

Interaction effects in econometrics

We provide practical advice for applied economists regarding robust specification and interpretation of linear regression models with interaction terms. We replicate a number of prominently published

Causal interaction and effect modification: same model, different concepts

Abstract Social scientists use the concept of interactions to study effect dependency. In the causal inference literature, interaction terms may be used in two distinct type of analysis. The first

Using SAS® to Test, Probe and Display Interaction Effects in Linear Models

Interaction effects (or moderated effects) in regression models capture how the effect of an independent variable on the dependent variable varies as a function of a third variable. In general terms,

Reducing Model Misspecification and Bias in the Estimation of Interactions

This paper uses a post-double selection approach that utilizes several lasso estimators to select the interactions to include in the final model, and extends this approach to estimate uncertainty for both interaction and marginal effects.

Understanding Interaction Models: Improving Empirical Analyses

Multiplicative interaction models are common in the quantitative political science literature. This is so for good reason. Institutional arguments frequently imply that the relationship between

A Simple Multivariate Test for Asymmetric Hypotheses

It is argued that claims of necessity and sufficiency involve a type of asymmetric causal claim that is useful in many social scientific contexts and that there is nothing about such asymmetries that should lead scholars to depart from standard social science practice.



Testing for Interaction in Multiple Regression

  • P. Allison
  • Mathematics
    American Journal of Sociology
  • 1977
Contrary to a recent claim, the inclusion of a product term in a multiple regression is a legitimate way to test for interaction. The unstandardized coefficient and the t-test for the product term

In Defense of Multiplicative Terms In Multiple Regression Equations

Though the inclusion of multiplicative terms in multiple regression equations is often prescribed as a method for assessing interaction in multivariate relationships, the technique has been

Beyond linearity by default: Generalized additive models

Social scientists almost always use statistical models positing the dependent variable as a global, linear function of X, despite suspicions that the social and political world is not so simple, or

Causal Complexity and the Study of Politics

Theories that posit complex causation, or multiple causal paths, pervade the study of politics but have yet to find accurate statistical expression. To remedy this situation I derive new econometric

A Guide to Econometrics

The fourth edition of "A Guide to Econometrics" provides an overview of the subject and an intuitive feel for its concepts and techniques without the notation and technical detail often characteristic of econometric textbooks.

Do Democratic Institutions Constrain or Inform? Contrasting Two Institutional Perspectives on Democracy and War

How do domestic political institutions affect the way states interact in international crises? In the last decade we have witnessed an explosion of interest in this question, thanks largely to the

Democratic Transitions, Institutional Strength, and War

The relationship between democratization and war has recently sparked a lively debate. We find that transitions from autocracy that become stalled prior to the establishment of coherent democratic

Trade, Democracy, and the Size of the Public Sector: The Political Underpinnings of Openness

Politics remains prominently absent in the literature showing that higher levels of trade integration lead to a larger public sector. As openness increases, the state, acting as a social planner,

A theory of production