Adding a quadratic term to a regression changes linear coefficients due to collinearity and omitted variable bias
the verdict
SUPPORTED
the evidence backs this
refutedsupported
the weight of evidence
1 source for · 0 against
AS REPORTEDno primary record reached; this is what the reporting says
Retrieved evidence demonstrates that including quadratic terms and non-linear specifications in regression models affects parameter estimation and accounts for omitted non-linearities.
# Problems with products? Control strategies for models with interaction and quadratic effects
Political Science Research and Methods. Published: 2020-05-18. 33 citations.
## Authors
- Janina Beiser‐McGrath (University of Konstanz): h-index 5; 103 citations; corresponding author
- Liam F. Beiser‐McGrath (University of Konstanz): h-index 15; 1,108 citations
## Topics
- Electoral Systems and Political Participation
- Advanced Causal Inference Techniques
- Economic Policies and Impacts
---
# Problems with products? Control strategies for models with interaction and quadratic effects*
## Abstract
Models testing interactive and quadratic hypotheses are common in Political Science but control strategies for these models have received little attention. Common practice is to simply include additive control variables, without relevant product terms, into models with interaction or quadratic terms. In this paper, we show in Monte Carlos that interaction terms can absorb the effects of other un-modeled interaction and non-linear effects and analogously, that included quadratic terms can reflect omitted interactions and non-linearities. This problem even occurs when included and omitt