ArticlePDF Available

Interpreting Dummy Variables and Their Interaction Effects in Strategy Research

February 2007
Strategic Organization 5(1):13-30

DOI:10.1177/1476127006073512

Authors:

Paul S. L. Yip

Eric W. K. Tsang

The University of Texas at Dallas

Download full-text PDF Read full-text

Download full-text PDF

Read full-text

Download citationCopy link

Citations (117)References (27)Figures (1)

Abstract and Figures

Dummy variables have been employed frequently in strategy research to capture the influence of categorical variables. However, misinterpretation of results may arise, especially when interaction effects between dummy variables and other explanatory variables are involved in a regression. We discuss two approaches of entering dummy variables into a regression and their associated interpretations. We discuss some common mistakes of interpretation and hypothesis testing found in two recently published strategy papers, and highlight the advantages of our recommended approach over the approach usually adopted by management researchers.

OLS regression results with the partition and base approaches a

…

Figures – uploaded by Eric W. K. Tsang

Author content

Content may be subject to copyright.

Discover the world’s research

25+ million members
160+ million publication pages
2.3+ billion citations

Join for free

Public Full-text 1

Content uploaded by Eric W. K. Tsang

Author content

Content may be subject to copyright.

STRATEGIC ORGANIZATION Vol 5(1):13–30

DOI: 10.1177/1476127006073512

http://so.sagepub.com

ARTICLES

Interpreting dummy variables and their

interaction effects in strategy research

Paul S. L.Yip Nanyang Technological University, Singapore

Eric W. K.Tsang Wayne State University,USA

Abstract

Dummy variables have been employed frequently in strategy research to capture the influence

of categorical variables. However, misinterpretation of results may arise, especially when inter-

action effects between dummy variables and other explanatory variables are involved in a

regression. We discuss two approaches of entering dummy variables into a regression and

their associated interpretations. We discuss some common mistakes of interpretation and

hypothesis testing found in two recently published strategy papers, and highlight the advan-

tages of our recommended approach over the approach usually adopted by management

researchers.

Key words • base approach •dummy variable •interaction effect •partition approach

Dummy variables have been frequently used by management researchers to cap-

ture the influence of categorical variables. In particular, strategy researchers

often make use of dummy variables to study strategic responses or orientations.

For example, McGahan and Mitchell (2003) discuss how firms change in the

face of constraints to change. They argue that instead of examining the different

ways in which firms respond to constraints, it is worthwhile investigating

whether responses are path-independent or path-dependent. Each of these two

types of responses can be represented by a dummy variable. Another example is

the Miles and Snow (1978) typology, which is a comprehensive way of describ-

ing strategic orientations. Rajagopalan (1996), for instance, examines the per-

formance implications of the fit between strategic orientations and incentive

plan characteristics and uses dummy variables to represent strategic orientations

in her empirical analysis.

Another crucial area of strategy research is to study how discrete strategic

choices affect firm performance, and these choices are represented by dummy

at Universidad de Alicante on November 4, 2008 http://soq.sagepub.comDownloaded from

Citations (117)

References (27)

… La base de datos está compuesta por un total de 26 variables, para efectos de este artículo se mencionarán aquellas relevantes para comprender la asociación entre las denuncias VIF de pareja y las medidas cautelares en Chile, con algunas variables sociodemográficas relevantes, para las cuales, se crearon variables algunas dummies como denuncia por VIF (Sí / No) y estado de relación con el victimario (Vigente / Terminada) (Yip y Tsang 2007). …

… En primer lugar, se procedió a armonizar la información, operacionalizar variables y crear variables dummies para poder proceder al análisis de los datos (Yip y Tsang 2007). Una vez realizada la etapa de limpieza de datos se efectuó una exploración descriptiva de los datos, para comprender tendencias y posteriormente hacer un análisis en profundidad. …

Caracterización y análisis de las medidas cautelares de los casos de femicidios en Chile (2008-2022)

Article

Full-text available

Dec 2024

View

Caracterización y análisis de las medidas cautelares de los casos de femicidios en Chile (2008-2022)

Article

Full-text available

Nov 2024

ViewShow abstract

… Correspondingly, we construct the variable physical_low (transition_low), coded 1 if the industry-specific physical biodiversity risk (transition biodiversity risk) is lower than the median, and 0 otherwise. We construct both variables (i.e., high and low) for each risk dimension (i.e., physical and transition) since we apply the partition approach in our moderation analysis (Goettsche et al., 2016;Yip & Tsang, 2007). 20 …

Pricing firms’ biodiversity risk exposure: Empirical evidence from audit fees

Article

Full-text available

Mar 2025
J IND ECOL

ViewShow abstract

… The regression model can be expressed as follows (equation 1 Dummy variable regression is valuable for incorporating categorical predictors into a regression model, such as geographic location (Yip & Tsang, 2007 …

CEOs’ and Directors’ perspective towards environmental sustainability and climate change

Article

Feb 2025
J CLEAN PROD

View

… We measured the joint effects through a series of dummy variables representing each pretrial detention status and mode of conviction pairing; the detained and trial-convicted group served as the reference to test our hypothesis of whether these defendants are sentenced most severely. We also elected to use this “partition approach” (Yip & Tsang, 2007) to assess the joint effects given recent criticisms surrounding the interpretation of interaction term coefficients in nonlinear models (Mize, 2019). While the propensity score analysis addressed most covariate imbalances across detention status, we also conducted “doubly robust” outcome analyses by including the covariates in the sentencing models to address any lingering imbalances (Griffin et al., 2023). …

Is There a (Pre)Trial Penalty? Assessing the Joint Impacts of Pretrial Detention and Trial Conviction on Sentencing

Article

Feb 2025

ViewShow abstract

… Data transformation using the MSI method was carried out on the variable of the highest education completed by the Village Head. Meanwhile, dummy variable is a categorical scale data transformation method [66]. Dummy variables can only be represented by the values “0” and “1”, which a value of “1” for situations that are considered good and others are given a value of “0” [67]. …

Smart Village Concept in Indonesia: ICT as Determining Factor

Article

Full-text available

Jan 2025

ViewShow abstract

… Cox and Schechter (2019); Yip and Tsang (2007) Note. Authors’ compilation. …

Does Profitability Moderate the Relationship Between the Leverage and Dividend Policy of Manufacturing Firms in Nigeria and South Africa?

Article

Full-text available

Dec 2024

ViewShow abstract

… where β ij is the estimated regression coefficient for age group i, A i is a dummy variable for age with value 1 for age group i and 0 otherwise, while x j is one of the explanatory constructs A, A risk , IN, DN, and PBC. We specified the dummy moderator variables following the partition approach to ease the interpretation of dummy variables and their interaction effects (Yip and Tsang, 2007). In this model, the regression coefficient for the construct x j is made dependent on the age group, as in a conventional interaction, but without including the effect of age in the intercept of the model. …

Not the average farmer: Heterogeneity in Dutch arable farmers’ intentions to reduce pesticide use

Article

Dec 2024
ENVIRON SCI POLICY

View

… We also define three groups according to internal R&D intensity groups: (i) no internal R&D 24 , (ii) internal R&D intensity being positive but lower than €5,000/employee, (iii) internal R&D intensity higher than €5,000/employee 25 . These groups are interacted with the COOPUNI and COOPCTKIB variables following the partition approach (Yip & Tsang, 2007) for simplicity of interpretation (the coefficients directly reporting the effect of cooperation with universities/KIBS for these particular groups of firms, that is, the difference in the expected value of the dependent variable for two firms that belong to a specific group when one of them cooperates with universities/KIBS and the other does not, holding all other covariates equal). Table 5 reports the results for the interactions with size and R&D intensity. …

Does Cooperation with Universities and Knowledge Intensive Business Services Matter? Firm-level Evidence from Spain

Article

Full-text available

Nov 2024

ViewShow abstract

Foreign identity and organizational crises: Evidence in the U.S. automobile industry

Article

Aug 2024
J WORLD BUS

ViewShow abstractShow more

https://imasdk.googleapis.com/js/core/bridge3.692.0_en.html#fid=goog_1946775505

115M

202

Learn how to best influence lab equipment purchasing behavior

Recommended publications

Discover more

Article

Gender and Governors’ Policy Agendas

January 1970

Barbara E Ferrara

The rise of women governors is one of the most significant developments in U.S. politics in recent decades, one fraught with potential policy effects that have only recently begun to be explored. I study a crucial aspect of these potential effects, gubernatorial agenda setting, by testing the hypothesis drawn from the women and legislatures literature that male and female governors have different … [Show full abstract]

Article

Testing hypothesis on the equality of regression functions

January 1996

Article

Full-text available

Predictive Capability of Bedload Equations Using Flume Data

March 2012 · Journal of Hydrology and Hydromechanics

The study on bedload transport behaviour is widely explored from the last few decades and many semi-empirical or empirical equilibrium transport equations are developed. The phenomenon is a very complex due to its varied physical properties like velocity, depth, slope, particle size in the alluvial system. In practi-cal applications, these formulae have appreciable deviation from each other in … [Show full abstract]

View full-text

Article

The inapplicability of formulae to determine vertical dimension

March 1979 · Australian Dental Journal

The accuracy and precision of certain external cephalometric measurements were investigated on a subject presenting optimal conditions, using ten operators. Due to difficulties in identifying landmarks and tissue compressibility, the standard deviation of a measurement for ten replications ranged from 0 . 3 to 1 . 7 mm, while the “between operators” standard deviation was no better than 2 . 0 mm. … [Show full abstract]

Last Updated: 22 Oct 2024

Discover by subject area