Department of Economics and Business Economics

Statistical Myths about Log-Transformed Dependent Variables and How to Better Estimate Exponential Models

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Standard

Statistical Myths about Log-Transformed Dependent Variables and How to Better Estimate Exponential Models. / Villadsen, Anders Ryom; Wulff, Jesper.

In: British Journal of Management, Vol. 32, No. 3, 07.2021, p. 779-796.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Harvard

APA

CBE

MLA

Vancouver

Author

Bibtex

@article{8c49d0f2c2e9435a8f9bcec584da7306,
title = "Statistical Myths about Log-Transformed Dependent Variables and How to Better Estimate Exponential Models",
abstract = "We review 10 years of research published in the Strategic Management Journal (SMJ) and find the wide use of log-transformed dependent variables (LTDVs) to be based on statistical myths, with possible detrimental effects for the validity of research findings. We find that many researchers use LTDVs for the wrong reasons, and very often in a way that is misaligned with the hypothesis they intend to examine. Researchers also appear unaware of the severe shortcomings of LTDVs. Using LTDVs implies estimating an exponential model, which represents a non-linear relationship. We identify three myths that are widely followed by researchers: (1) LTDVs should be used to make distributions more normal; (2) linear hypotheses can be tested with LTDVs; and (3) LTDVs are the best way to estimate an exponential model. We call on researchers to exhibit caution when planning to use LTDVs and recommend instead the use of generalized linear models (GLMs) with quasi-maximum likelihood estimation. The superiority of GLMs is demonstrated by two empirical examples from recently published studies.",
keywords = "DECADE, ERRORS, FLEXIBLE LINK, KNOWLEDGE, MARKET, METHODOLOGICAL MYTHS, STRATEGIC MANAGEMENT, VALUES, ZERO",
author = "Villadsen, {Anders Ryom} and Jesper Wulff",
year = "2021",
month = jul,
doi = "10.1111/1467-8551.12431",
language = "English",
volume = "32",
pages = "779--796",
journal = "British Journal of Management",
issn = "1045-3172",
publisher = "Wiley-Blackwell Publishing Ltd.",
number = "3",

}

RIS

TY - JOUR

T1 - Statistical Myths about Log-Transformed Dependent Variables and How to Better Estimate Exponential Models

AU - Villadsen, Anders Ryom

AU - Wulff, Jesper

PY - 2021/7

Y1 - 2021/7

N2 - We review 10 years of research published in the Strategic Management Journal (SMJ) and find the wide use of log-transformed dependent variables (LTDVs) to be based on statistical myths, with possible detrimental effects for the validity of research findings. We find that many researchers use LTDVs for the wrong reasons, and very often in a way that is misaligned with the hypothesis they intend to examine. Researchers also appear unaware of the severe shortcomings of LTDVs. Using LTDVs implies estimating an exponential model, which represents a non-linear relationship. We identify three myths that are widely followed by researchers: (1) LTDVs should be used to make distributions more normal; (2) linear hypotheses can be tested with LTDVs; and (3) LTDVs are the best way to estimate an exponential model. We call on researchers to exhibit caution when planning to use LTDVs and recommend instead the use of generalized linear models (GLMs) with quasi-maximum likelihood estimation. The superiority of GLMs is demonstrated by two empirical examples from recently published studies.

AB - We review 10 years of research published in the Strategic Management Journal (SMJ) and find the wide use of log-transformed dependent variables (LTDVs) to be based on statistical myths, with possible detrimental effects for the validity of research findings. We find that many researchers use LTDVs for the wrong reasons, and very often in a way that is misaligned with the hypothesis they intend to examine. Researchers also appear unaware of the severe shortcomings of LTDVs. Using LTDVs implies estimating an exponential model, which represents a non-linear relationship. We identify three myths that are widely followed by researchers: (1) LTDVs should be used to make distributions more normal; (2) linear hypotheses can be tested with LTDVs; and (3) LTDVs are the best way to estimate an exponential model. We call on researchers to exhibit caution when planning to use LTDVs and recommend instead the use of generalized linear models (GLMs) with quasi-maximum likelihood estimation. The superiority of GLMs is demonstrated by two empirical examples from recently published studies.

KW - DECADE

KW - ERRORS

KW - FLEXIBLE LINK

KW - KNOWLEDGE

KW - MARKET

KW - METHODOLOGICAL MYTHS

KW - STRATEGIC MANAGEMENT

KW - VALUES

KW - ZERO

UR - http://www.scopus.com/inward/record.url?scp=85090138603&partnerID=8YFLogxK

U2 - 10.1111/1467-8551.12431

DO - 10.1111/1467-8551.12431

M3 - Journal article

VL - 32

SP - 779

EP - 796

JO - British Journal of Management

JF - British Journal of Management

SN - 1045-3172

IS - 3

ER -