Modeling and Interpreting Interactions in Tradeoff Research
Jeffrey J. Burks
University of Notre Dame
David W. Randolph
Jim A. Seida
University of Notre Dame - Department of Accountancy
October 30, 2013
We examine the use of interaction terms in studies of tax-based tradeoffs that managers face in financial reporting choices. Interactions allow the researcher to test whether the rate at which firms trade tax benefits for financial reporting benefits is dependent on the magnitude of the benefits. Although we focus on a tax setting, many of our insights apply to research designs that utilize interaction terms. We begin by defining the term tradeoff and distinguish a linear-additive model, with no interactions, from a linear-multiplicative model. We then discuss statistical issues that affect the interpretation and design of multiplicative models. Although we advocate the use of multiplicative models in tradeoff research, we show via simulation that noise levels common in tradeoff research greatly reduce the ability to statistically detect interaction effects. Given the difficulty of detecting interaction effects, we discuss potential solutions and the need to provide a balanced discussion of unconditional and conditional effects so readers appreciate that Type II error may be hiding systematic differences in an effect across firms.
Number of Pages in PDF File: 49
Keywords: tradeoff, tax, non-tax costs, interaction, linear-multiplicative
JEL Classification: M40, C50, H25working papers series
Date posted: October 31, 2013
© 2014 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo8 in 0.500 seconds