The perils of endogeneity and instrumental variables in strategy research

Understanding through simulations

Matthew Semadeni, Michael C. Withers, Samuel Certo

Research output: Contribution to journalArticle

139 Citations (Scopus)

Abstract

In this paper we use simulations to examine how endogeneity biases the results reported by ordinary least squares (OLS) regression. In addition, we examine how instrumental variable techniques help to alleviate such bias. Our results demonstrate severe bias even at low levels of endogeneity. Our results also illustrate how instrumental variables produce unbiased coefficient estimates, but instrumental variables are associated with extremely low levels of statistical power. Finally, our simulations highlight how stronger instruments improve statistical power and that endogenous instruments can report results that are inferior to those reported by OLS regression. Based on our results, we provide a series of recommendations for scholars dealing with endogeneity.

Original languageEnglish (US)
Pages (from-to)1070-1079
Number of pages10
JournalStrategic Management Journal
Volume35
Issue number7
DOIs
StatePublished - 2014

Fingerprint

Strategy research
Simulation
Endogeneity
Instrumental variables
Ordinary least squares
Statistical power
Endogeneity bias
Coefficients

Keywords

  • endogeneity
  • instrumental variables
  • ordinary least squares
  • simulations
  • strategy research

ASJC Scopus subject areas

  • Business and International Management
  • Strategy and Management

Cite this

The perils of endogeneity and instrumental variables in strategy research : Understanding through simulations. / Semadeni, Matthew; Withers, Michael C.; Certo, Samuel.

In: Strategic Management Journal, Vol. 35, No. 7, 2014, p. 1070-1079.

Research output: Contribution to journalArticle

@article{09c6d760597d44c78ac71b0254255838,
title = "The perils of endogeneity and instrumental variables in strategy research: Understanding through simulations",
abstract = "In this paper we use simulations to examine how endogeneity biases the results reported by ordinary least squares (OLS) regression. In addition, we examine how instrumental variable techniques help to alleviate such bias. Our results demonstrate severe bias even at low levels of endogeneity. Our results also illustrate how instrumental variables produce unbiased coefficient estimates, but instrumental variables are associated with extremely low levels of statistical power. Finally, our simulations highlight how stronger instruments improve statistical power and that endogenous instruments can report results that are inferior to those reported by OLS regression. Based on our results, we provide a series of recommendations for scholars dealing with endogeneity.",
keywords = "endogeneity, instrumental variables, ordinary least squares, simulations, strategy research",
author = "Matthew Semadeni and Withers, {Michael C.} and Samuel Certo",
year = "2014",
doi = "10.1002/smj.2136",
language = "English (US)",
volume = "35",
pages = "1070--1079",
journal = "Strategic Management Journal",
issn = "0143-2095",
publisher = "John Wiley and Sons Ltd",
number = "7",

}

TY - JOUR

T1 - The perils of endogeneity and instrumental variables in strategy research

T2 - Understanding through simulations

AU - Semadeni, Matthew

AU - Withers, Michael C.

AU - Certo, Samuel

PY - 2014

Y1 - 2014

N2 - In this paper we use simulations to examine how endogeneity biases the results reported by ordinary least squares (OLS) regression. In addition, we examine how instrumental variable techniques help to alleviate such bias. Our results demonstrate severe bias even at low levels of endogeneity. Our results also illustrate how instrumental variables produce unbiased coefficient estimates, but instrumental variables are associated with extremely low levels of statistical power. Finally, our simulations highlight how stronger instruments improve statistical power and that endogenous instruments can report results that are inferior to those reported by OLS regression. Based on our results, we provide a series of recommendations for scholars dealing with endogeneity.

AB - In this paper we use simulations to examine how endogeneity biases the results reported by ordinary least squares (OLS) regression. In addition, we examine how instrumental variable techniques help to alleviate such bias. Our results demonstrate severe bias even at low levels of endogeneity. Our results also illustrate how instrumental variables produce unbiased coefficient estimates, but instrumental variables are associated with extremely low levels of statistical power. Finally, our simulations highlight how stronger instruments improve statistical power and that endogenous instruments can report results that are inferior to those reported by OLS regression. Based on our results, we provide a series of recommendations for scholars dealing with endogeneity.

KW - endogeneity

KW - instrumental variables

KW - ordinary least squares

KW - simulations

KW - strategy research

UR - http://www.scopus.com/inward/record.url?scp=84898903847&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84898903847&partnerID=8YFLogxK

U2 - 10.1002/smj.2136

DO - 10.1002/smj.2136

M3 - Article

VL - 35

SP - 1070

EP - 1079

JO - Strategic Management Journal

JF - Strategic Management Journal

SN - 0143-2095

IS - 7

ER -