Control variable is an extra variable that is not accounted for in research.
The selection, use, and reporting of control variables in international business research: A review and recommendations
Author links open overlay panelBo BernhardNielsenabPersonEnvelopeArpitRaswantaEnvelope Show
ShareShare Cited ByCite https://doi.org/10.1016/j.jwb.2018.05.003Get rights and content AbstractThis study explores the selection, use, and reporting of control variables in studies published in the leading international business (IB) research journals. We review a sample of 246 empirical studies published in the top five IB journals over the period 2012–2015 with particular emphasis on selection, use, and reporting of controls. Approximately 83% of studies included only half of what we consider Minimum Standard of Practice with regards to controls, whereas only 38% of the studies met the 75% threshold. We provide recommendations on how to effectively identify, use and report controls in IB studies. IntroductionControl variables (CVs) constitute a central element of the research design of any empirical study. Confounding variables are likely to covary with the hypothesized focal independent variables thus limiting both the elucidation of causal inference as well as the explanatory power of the model (Pehazur & Schmelkin, 1991; Stone-Romero, 2009). Therefore, researchers must seek to rule out threats to valid inferences in order to determine to what extent the focal independent variables behave as hypothesized. This is typically done by including (controlling for) extraneous variables that are deemed theoretically (or empirically) important but are not focal variables of the study (Kish, 1959). The literature sometimes refers to such variables as covariates, confounding variables, nuisance variables, control variables or simply controls (Atinc, Simmering, & Kroll, 2012; Breaugh, 2008). Researchers need to account for these variables either through experimental design (before the data gathering) or through statistical analysis (after the data gathering process). In this way the researchers are said to account for their effects to avoid a false positive (Type I) error (i.e. falsely concluding that the dependent variables are in a causal relationship with the independent variable). Inadequate attention to controls is a major threat to the validity of inferences made about cause and effect (internal validity).1 One way of controlling by inclusion is to use a matched-group design where particular entities (e.g., state-owned and privately owned firms) that vary in terms of independent and dependent variables are matched on specific criteria (Estrin, Meyer, Nielsen, & Nielsen, 2016). An alternative way of controlling is exclusion by holding particular variables constant, such as limiting a study to emerging market firms only (Buckley, Elia, & Kafouros, 2014). Yet the most common way to control for extraneous influences is via statistical controls. Statistical controls aim at identifying potential sources of influence during study design and including CVs representing these sources of influence during data collection. During data analysis, researchers then control for these extraneous effects by mathematically partialling out variance associated with CVs in calculating relationships between other variables, thereby reducing the risk of Type II errors (Carlson & Wu, 2012; Spector, Zapf, Chen, & Frese, 2000). In this study we focus on IB research that includes statistical controls as non-hypothesized variables in regression type studies. When regressing for instance firm performance (or entry mode) on other variables, IB researchers attempt to establish which specific variables influence the prediction and which do not. This is typically done by considering whether each variable’s contribution remains statistically significant after controlling for other predictors. In multiple regression, when the coefficient of a predictor variable differs significantly from zero, most scholars conclude that this variable makes a “unique” contribution to the outcome. CVs are assumed to be confounding, that is, producing distortions in observed relationships. For this reason, researchers typically clearly assign some variables as being merely controls, or variables of no particular theoretical interest, that need to be somehow removed in their effects on the study. While statistical controls are able to adjust relationships between variables for the action of other variables, this ability is based on certain implicit assumptions about the underlying role of control variables on either the observed measures or the underlying constructs of interest. More generally, the argument seems to be that we decrease the aggregate bias for every additional relevant variable that we include. The inefficiency part of the equation is, however, rarely mentioned, as control variables often do have real effects. Yet, the mathematics of regression analysis do not support the argument that more variables in a regression, even relevant ones, necessarily makes the regression results more accurate (Clarke, 2005). In fact, even small amounts of measurement error in control variables “are magnified as more variables are added to the equation in an attempt to control for other possible sources of bias.” (Griliches, 1977: 12). Control variables are of extreme importance in econometric analyses for a number of reasons. First, the variables included in the analysis drive the results of any statistical analysis of data. Hence, the improper use (inclusion or exclusion) of CVs may distort results and produce misleading findings. Similar to any other variable included in a model (e.g., any predictor or criteria variable), decisions regarding which controls to include affect the significance levels and estimated effect sizes of the other variables. Second, replication and generalizability of results cannot be done without specific knowledge of which factors were controlled, the measurement of these controls, and the specific method utilized for controlling. Finally, inadequate justification and reporting of controls render any extension difficult. This includes meta-analyses, which cannot be conducted on studies where controls are unknown, unjustified, or measurement and descriptive statistics are not reported. In order to advance IB research and build a cumulative body of knowledge about certain phenomena, the correct selection, inclusion and treatment as well as documentation and reporting of CVs is critical since controls often serve as inspiration for new studies of relationships (i.e., as potential moderators/mediators, IVs or even DVs). We build upon insights from previous articles on the role of control variables in social science research (e.g., Atinc et al., 2012; Becker, 2005; Becker et al., 2015; Bernerth & Aguinis, 2015; Breaugh, 2006; Carlson & Wu, 2012; Spector & Brannick, 2011). These studies document the (mis)use of control variables in social science research by analyzing how published work in the top tier management and organizational psychology journals have treated controls inadequately. To the best of our knowledge, however, this is the first comprehensive review of the selection, use, and reporting of control variables in IB research (also see Aguinis, Cascio, & Ramani, 2017). As such, we join an important (recent) conversation within the IB research community which calls for more attention to both methodological rigor in empirical testing and preciseness in presentation and reporting of results (e.g., Andersson, Cuervo-Cazurra, & Nielsen, 2014; Ahlstrom, 2015; Cortina, Köhler & Nielsen, 2015; Cuervo-Cazurra, Andersson, Brannen, Nielsen, & Reuber, 2016; Kingsley, Noordewier, & Bergh, 2017; Welch & Piekkari, 2017). IB research is particularly vulnerable to issues arising from poor treatment in terms of selection, analysis and reporting of control variables due to its complex and multi-disciplinary nature, often spanning multiple countries and contexts (Aguinis et al., 2017; Cuervo-Cazurra et al., 2016). IB studies involve phenomena where country level context (e.g., institutional or cultural) often play a decisive role as boundary conditions for theory development. In fact, what sets IB studies apart from more general strategy, management or organizational research is the cross-border (international) business context in which actors (individuals, teams, firms or even industries) act and interact (Zaheer, Schomaker, & Nachum, 2012). This international context has important implications for use of control variables as it helps establish the boundaries of applicability surrounding a particular empirical argument and rule out alternative or confounding explanations of findings (Teagarden, Von Glinow, & Mellahi, 2018). As noted by Cho and Padmanabhan (2005: 309) “no international business study can be complete unless there is an explicit variable controlling for cultural distance.” This study seeks to investigate the state-of-the-art of treatment of control variables in IB studies. For comparison reasons, we focus on specific issues pertaining to the selection, use and reporting of control variables studied previously, but re-interpret these in terms of specific importance to IB research. Together with our concrete recommendations, this approach is intended to provide IB scholars with a comprehensive yet easy to follow guide to improve their treatment of control variables. In addition, we specifically examine the treatment of country level, contextual variables as controls in IB research, and recommend ways to improve practice with regards to such controls. We start by introducing our sample and method followed by a thorough analysis of the current CV use and reporting in 246 empirical articles published in the top five IB journals during the period 2012–2015. We compare and contrast the use of controls both between the five IB journals and with result from previous studies in other fields. Based on our findings, we provide a set of recommendation to guide future authors, reviewers and editors toward a more consistent and accurate way of controlling for extraneous variables in IB research. Section snippetsSelection of articlesIn an attempt to be comprehensive, we coded all empirical articles published in five top IB journals over the period 2012–2015 with regards to the use and reporting of controls. Given our focus on use and reporting of statistical extraneous CVs,2 FindingsTable 2 reports the percentages of each of the dimensions used to assess the incorporation and reporting of CVs in IB studies. The table illustrates the differences in practices with regards to use and presentation of CVs across IB journals and on average with regards to both individual and aggregated control dimensions. We begin by discussing the overall trends and compare results to findings from four top management journals (Academy of Management Journal, Strategic Management Journal, Journal Recommendations and conclusionIf international business research wishes to increase its ability to inform other research areas through citations, the top journals in the field must strive for higher methodological standards. Three key areas are the selection, use, and reporting of control variables, which has hitherto received little attention in the IB literature. The purpose of our article was to (1) analyze the current treatment of controls in IB research, (2) compare it to other social science published research, and References (60)
From the editors: Explaining interaction effects within and across levels of analysisJournal of International Business Studies(2014) Control variable use and reporting in macro and micro management researchOrganizational Research Methods(2012) Why we need industrial–organizational psychology to fix industrial–organizational psychologyIndustrial and Organizational Psychology(2013) Questions about questionable research practices in the field of management: A guest commentary(2016) The effect of context-related moderators on the internationalization-performance relationship: Evidence from meta-analysisManagement International Review(2007) Statistical control in correlational studies: 10 essential recommendations for organizational researchersJournal of Organizational Behavior(2015) Potential problems in the statistical control of variables in organizational research: A qualitative analysis with recommendationsOrganizational Research Methods(2005) A critical review and best-practice recommendations for control variable usagePersonnel Psychology(2015) Creating repeatable cumulative knowledge in strategic management: A call for a broad and deep conversation among authors, referees, and editorsStrategic Management Journal(2015) Construct measurement in strategic management research: Illusion or reality?Strategic Management Journal(2005) Rethinking the control of nuisance variables in theory testingJournal of Business and Psychology(2006) Acquisitions by emerging market multinationals: Implications for firm performanceJournal of World Business(2014) Conceptualizing and measuring culture in international business and management: From challenges to potential solutionsJournal of International Business Studies(2015) The illusion of statistical control: Control variable practice in management researchOrganizational Research Methods(2012) The phantom menace: Omitted variable bias in econometric researchConflict Management and Peace Science(2005) Suppressor variables, prediction, and the interpretation of psychological relationshipsEducational and Psychological Measurement(1972) Restriction of variance interaction effects and their importance for international business researchJournal of International Business Studies(2015) From the editors: Can I trust your findings? Ruling out alternative explanations in international business researchJournal of International Business Studies(2016) State-owned MNCs and host country expropriation risk: The role of home state soft power and economic gunboat diplomacyJournal of International Business Studies(2014) A content analysis of the content analysis literature in organization studies: Research themes, data sources, and methodological refinementsOrganizational Research Methods(2007) 2023, Journal of Air Transport Management Show abstractNavigate Down This study aims at comprehending the significance of brand experience in impacting brand satisfaction in the aviation sector. Grounded on the theory of experiential marketing, the current research aims to explain the phenomena by which brand experience can be created through marketing communication channels. This study also tests the role of brand love as a mediator on the relationship between brand experience and brand satisfaction, by analyzing 480 responses. This research concludes that brand love partially mediates the relationship between brand experience and brand satisfaction in the aviation sector. Airline managers need to focus their strategies on event marketing and marketing communication as they have a significant effect on brand experience. The findings will also help in understanding the different dimensions of brand experience. 2023, Journal of World Business Show abstractNavigate Down International vertical alliances (IVAs) have garnered increasing scholarly interest in the strategy and international business (IB) literature. Our review of 111 papers published in major IB journals from 2000 to 2020 sheds light on the antecedents, key mediators, moderators and outcomes of IVAs. To generate insights, we juxtaposed forward and backward alliances and compared IVAs with their domestic vertical and horizontal counterparts. In this paper, we highlight key areas for future IVA research, including—but not limited to—broadening the scope of the investigation in order to integrate new theories and methods suited to examine such alliances in the IB field. 2022, Journal of Cleaner Production Show abstractNavigate Down Drawing on the resource-based view (RBV) of the firm and the ecological modernization theory (EMT), the current study examines the mediating effect of green innovation (GI) between green manufacturing practices (GMP) and corporate sustainable performance (CSP). In addition, it investigates the moderating effect of green organizational culture (GOC) on the relationship between GMP and GI. To test the hypothesized model, the data was collected from 328 Saudi manufacturing SMEs and analyzed using the hierarchical regression analysis in SPSS. The empirical results confirmed the effect of GMP on GI, which in turn, has an effect on CSP. The results also confirmed the positive effect of GMP on CSP through GI, which is enhanced by the presence of GOC. Thus, this study expands our understanding of the GMP-CSP relationship, its potential mechanism, and its conditional effect. This study has valuable contributions to the sustainability and green literature, and it has managerial implications for firms seeking to develop effective sustainable performance. 2022, Journal of Business Research Show abstractNavigate Down Owing to privatization and divestment, State Owned Enterprises (SOEs) in many emerging markets are evolving in their form and function. SOEs, increasingly, are hybrid organizations, where the state is one of the owners along with other ownership groups such as institutional investors. Combining the resource-based view and institutional perspectives, we argue that the different ownership groups within the SOE can be instrumental in promoting/deterring the internationalization efforts of SOE. Empirical results from a sample of 1310 firm year observations of 116 unique firms within a 2011–2019 time-period demonstrate that state and foreign institutional ownership impact SOE internationalization negatively whereas domestic institutional ownership has a positive impact on SOE internationalization. Additionally, we examine interactions between different hybrid ownership groups and find that both foreign and domestic institutional investors can offer resource advantages to enable state owners to invest in internationalization activities. This study contributes towards a deeper understanding of ownership hybridity, internationalization challenges and resource mobilization in SOEs from emerging markets. 2022, Journal of Business Research Show abstractNavigate Down This study provides empirical proof that whole organizational innovativeness is rooted in tacit knowledge due to its potency of human capital creation and, that a learning culture composed of a learning climate and mistakes acceptance component fosters human capital development. The main practical implication is that if the IC components are externally rather than internally determined in the particular organization embedded in the specific healthcare system, human capital's power to create an innovative solution is diminished even if the learning culture is developed. So, practically, private healthcare organizations are more innovative than public. Novelty: This study exposed how tacit knowledge creation driven by learning culture and its mistakes acceptance critical component drives next IC components structure, which influences internal performance innovation in the healthcare sector driven by private and public funds. Findings were obtained from a healthcare industry sample composed of 350 cases from Poland and 365 from the United States. Data were analyzed using the structural equation modeling method using Amos and OLS regression using SPSS PROCESS macro. 2022, Journal of World Business Show abstractNavigate Down Many countries are becoming increasingly culturally heterogeneous and diverse. Yet, the cross-cultural research literature has only paid limited attention to the measurement of within-country cultural diversity. Following an analogy from evolutionary biology, this article proposes a novel index of cultural heterozygosity, which takes the individual as a starting point. It develops a distribution-free algorithm, which compares all individuals in a given country sample with each other. This index is different from commonly used fractionalization indices; its relevance for international business research is illustrated with an empirical example regarding country-level differences in corruption. Research article Labour Economics, Volume 21, 2013, pp. 111-121 Show abstractNavigate Down Based on new, exceptionally informative and large German linked employer–employee administrative data, we investigate the question whether the omission of important control variables in matching estimation leads to biased impact estimates of typical active labor market programs for the unemployed. Such biases would lead to false policy conclusions about the cost-effectiveness of these expensive policies. Using newly developed Empirical Monte Carlo Study methods, we find that besides standard personal characteristics, information about the current unemployment spell, regional information, pre-treatment outcomes, and detailed short-term labor market histories remove most of the selection bias. Research article Interactive visualization for research contextualization in international businessJournal of World Business, Volume 53, Issue 3, 2018, pp. 356-372 Show abstractNavigate Down We respond to calls for advances in the contextualization of international business (IB) research by introducing interactive visualization as a methodology for generating contextual insights during the exploratory phases of IB research projects. We suggest that applying interactive visualization early on improves contextualization by means of simultaneous dynamic representations of various phenomena and their respective properties and relationships, even for phenomena that have been widely researched before, like in the cases of international joint ventures and MNE foreign direct investment. The goal of this introduction is to make interactive visualization more accessible to IB scholars. Research article Control variables, discrete instruments, and identification of structural functionsJournal of Econometrics, Volume 222, Issue 1, Part A, 2021, pp. 73-88 Show abstractNavigate Down Control variables provide an important means of controlling for endogeneity in econometric models with nonseparable and/or multidimensional heterogeneity. We allow for discrete instruments, giving identification results under a variety of restrictions on the way the endogenous variable and the control variables affect the outcome. We consider many structural objects of interest, such as average or quantile treatment effects. We illustrate our results with an empirical application to Engel curve estimation. Research article Contextualizing international business research: Enhancing rigor and relevanceJournal of World Business, Volume 53, Issue 3, 2018, pp. 303-306 Show abstractNavigate Down Context differentiates international business (IB) from traditional Business research. Along with many IB scholars, we argue that context should be much more adequately emphasized in IB research. Location differences are commonly ignored; complexity and polycomplexity--and other levels of analysis issues--are rarely acknowledged; and the relevance of models and theory developed in Western contexts is not adequately questioned or explored. This paper suggests contextualization guidelines for scholars to enhance the rigor of their research and to make their IB research more relevant for practitioners. In conclusion we suggest solutions for closing rigor and relevance gaps in IB research by enhancing contextualization. Research article Establishing rigor in mail-survey procedures in international business researchJournal of World Business, Volume 50, Issue 1, 2015, pp. 26-35 Show abstractNavigate Down How rigorous have our data-collection procedures been in international business research? We report the results of a comprehensive content analysis of scholarly work published in four leading international business journals over the past decade. The focus is data-collection procedures used by researchers in mail surveys. The intent is to be self-critical and formulate strategies for enhancing the rigor and success of data-collection procedures in survey-based research. Our findings confirm that international business scholars could significantly improve surveys’ response rates by following more rigorous and well-established methodological practices already established in the social science literature. We also find that, while some continents tend to be oversampled, a large portion of the world remains underrepresented in international business research. The results point to interesting trends in cross-cultural data-collection procedures. Given that primary research will always drive new knowledge creation, scholars are strongly advised to practice best-available procedures for data collection. Research article Synergies in the space of control variables within the equilibrium-point hypothesisNeuroscience, Volume 315, 2016, pp. 150-161 Show abstractNavigate Down We use an approach rooted in the recent theory of synergies to analyze possible co-variation between two hypothetical control variables involved in finger force production based on the equilibrium-point (EP) hypothesis. These control variables are the referent coordinate (R) and apparent stiffness (C) of the finger. We tested a hypothesis that inter-trial co-variation in the {R; C} space during repeated, accurate force production trials stabilizes the fingertip force. This was expected to correspond to a relatively low amount of inter-trial variability affecting force and a high amount of variability keeping the force unchanged. We used the “inverse piano” apparatus to apply small and smooth positional perturbations to fingers during force production tasks. Across trials, R and C showed strong co-variation with the data points lying close to a hyperbolic curve. Hyperbolic regressions accounted for over 99% of the variance in the {R; C} space. Another analysis was conducted by randomizing the original {R; C} data sets and creating surrogate data sets that were then used to compute predicted force values. The surrogate sets always showed much higher force variance compared to the actual data, thus reinforcing the conclusion that finger force control was organized in the {R; C} space, as predicted by the EP hypothesis, and involved co-variation in that space stabilizing total force. What is a control variable in a research?A control variable is any variable that's held constant in a research study. It's not a variable of interest in the study, but it's controlled because it could influence the outcomes. Why are control variables important?
What is a variable that is not accounted for?A confounding variable, in simple terms, refers to a variable that is not accounted for in an experiment. It acts as an external influence that can swiftly change the effect of both dependent and independent research variables; often producing results that differ extremely from what is the case.
What is an extraneous variable in a research study?The whole point of conducting an experiment is to determine whether or not changing the values of some independent variable has an effect on a dependent variable. An extraneous variable is any variable you're not interested in studying that could also have some effect on the dependent variable.
|