Knowledge that Transforms

Given the prevalence of statistical techniques that use probability to quantify uncertainty, the aim of this article is to highlight the theoretical aspects and implications of the major current probability interpretations that justify the development and use of such techniques. After briefly sketching the origins and development of the notion of probability, its theoretical interpretations will be outlined. Two main trends will be distinguished: one epistemic and one empirical, corresponding to the twofold meaning characterizing probability. The epistemic type embodies the so-called classical theory put forward by Laplace as well as the logical and subjective approaches. By contrast, the frequency and propensity theories are, in theory, empirical in character. This way of understanding probability contrasts with both the tenet that there is a “Bayesian interpretation” of probability and the tendency to conflate Bayesian probability with the subjective interpretation, both of which are misleading for reasons that emerge from the following discussion. The final section of the paper addresses the question of which type of probability is best suited for the organization sciences and suggests the subjective interpretation as the best option by virtue of its pluralism and awareness of context.

Job analysis has a central role in virtually every aspect of HR and is one of several high performance work practices thought to underlie firm performance. Given its ubiquity and importance, it is not surprising that considerable effort has been devoted to developing comprehensive job analysis systems and methodologies. Yet, the complexity inherent in collecting detailed and specific “decomposed” information has led some to pursue “holistic” strategies designed to focus on more general and abstract job analysis information. It is not clear, however, if these two different strategies yield comparable information, nor if respondents are equally capable of generating equivalent information. Drawing from cognitive psychology research, we suggest that experienced and careless job analysis respondents are less likely to evidence convergence in their decomposed and holistic job analysis judgments. In a field sample of professional managers, we found that three different types of task-related work experience moderated the relationship between decomposed and holistic ratings, accounting for an average ΔR 2 of 4.7%. Three other more general types of work experience, however, did not moderate this relationship, supporting predictions that only experience directly related to work tasks would prove to be a liability when making judgments. We also found that respondent carelessness moderated the relationship between decomposed and holistic ratings, accounting for an average ΔR 2 of 6.2%. These results link cognitive limitations to important job analysis respondent differences and suggest a number of theoretical and practical implications when collecting holistic job analysis data.

This paper draws from a case study of decision analysis—a discipline rooted in Bayesianism aimed at supporting managerial decision making—to inform the current discussion on the adoption of Bayesian modes of thinking in management research and practice. Relying on concepts from the science, technology, and society field of study and actor-network theory, we approach the production of scientific knowledge as a cultural, practical, and material affair. Specifically, we analyze the activities deployed by decision analysts to overcome the challenges of making a discipline built on Bayes’ legacy scientifically acceptable, managerially relevant, and long lasting. As a novel contribution to the discussion on the “Bayesian revolution,” our study goes beyond institutional accounts of the legitimation of Bayesianism to highlight the role of politics and material artifacts in past and current attempts at importing Bayesianism. Our study also shows the importance of historical continuity in the promotion of Bayesian methods in management.

Bayesian estimation and inference remain infrequently used in organizational science research. Despite innumerable warnings regarding the entrenched frequentist paradigm, our field has yet to embrace the Bayesian “revolution” that seems to be sweeping through so many other disciplines. With this context as a backdrop, we address a simple yet difficult question: What is the likelihood that Bayesian methodologies eventually will supplement or even supplant traditional frequentist methodologies in the organizational science community? We draw on institutional theory to address this question, highlighting the cultural-cognitive, normative, and regulative forces that play important roles. As novel contributions to the discussion, we go beyond our own ideas and previously published opinions on the subject to report the opinions of 26 institutional elites (current and former officers of academic associations, editors, and editorial board members). These leading scholars help us shed light not only on the likelihood that Bayesianism will take root in the field but also on practical steps that could be taken to assist in this process. In some ways, we build Bayesian priors about Bayesian analysis, where those priors will be qualified on the basis of future events and outcomes.

Assessing the relative importance of predictors has been of historical importance in a variety of disciplines including management, medicine, economics, and psychology. When approaching hypotheses on the relative ordering of the magnitude of predicted effects (e.g., the effects of discrimination from managers and coworkers are larger than that from clients), one quickly runs into problems within a traditional frequentist framework. Null hypothesis significance testing does not allow researchers to directly map research hypotheses on to results and suffers from a multiple testing problem that leads to low statistical power. Furthermore, all traditional structural equation modeling fit indices lose much of their suitability for model comparison, because order hypotheses are not countable in terms of degrees of freedom. To adequately tackle order hypotheses, we advocate a Bayesian method that provides a single internally consistent solution for estimation and inference. The key element in the proposed model comparison approach is the use of the Bayes factor and the incorporation of order constraints by means of a smart formulation of prior distributions. An easy-to-use software package BIEMS (Bayesian inequality and equality constrained model selection) is introduced and two empirical examples in the organizational behavior area are provided to showcase the method, both offering new findings that have implications for theory: the first on the differential impact of discrimination in the workplace from insiders and outsiders to the organization on employees’ well-being, and the second on Karasek’s stressor–strain theory about how the relative order of magnitude of the effects of job control and demands depends on the specific well-being outcome dimension.

Although interpersonal interactions are the mainstay of many assessment center exercises, little is known about how these interactions unfold and affect participant behavior and performance. More specifically, participants interact with role players who have been instructed to demonstrate behavior reflecting specific dispositions as part of the exercise. This study focuses on role player portrayed disposition as a potentially important social demand relevant to participant behavior and performance in interpersonal simulations. We integrate interpersonal theory and trait activation theory to formulate hypotheses about the effects of role player portrayed disposition on participant behavior and performance in 184 interpersonal simulations. A significant effect of portrayed disposition was found for participant relationship building and directive communication behavior. Furthermore, portrayed disposition moderated the relationship between participant use of these behaviors and performance ratings. Conceptually, this study sheds light on the complementary mechanisms and social demands that produce participant performance differences across exercises. At a practical level, this study provides valuable evidence-based guidance for developing interpersonal simulations.

We examine how firms’ resources and the power of their governmental owners influence the likelihood of privatization of state-owned enterprises. Using data on 206 Chinese pharmaceutical firms over the period of 2000 to 2007, we found that firm financial performance (a proxy for firm resources) increased the likelihood of privatization. In addition to firm resources, we investigated how a heterogeneous body of decision makers within the governmental hierarchy influences the likelihood of firms’ privatization. We found that provincial governmental owners’ willingness to privatize firms increased when they had higher fiscal power. The results of this study also indicated a negative moderating influence of provincial fiscal power: Higher fiscal power of provincial governmental owners weakened the relationship between firm financial performance and the likelihood of privatization.

Social information exchange (SIE) in organizations has long been an area of interest for management scholars; however, in recent years, this literature has become fragmented and widely dispersed. As communication and transfer of information increasingly occur between individuals and aggregates of widely varying national and regional cultures, a reconsideration and review of the topic is appropriate, including identification of key issues in this research domain and an integration and synthesis of what we currently know about SIE across cultural boundaries. We examine the last 13 years of cross-cultural SIE research at the country, organization and subunit, team and dyad, and individual levels; provide a basic analytic framework; and provide propositions and direction to guide future research. Our review notes key findings based on three general topics in the literature: (1) antecedents to SIE, (2) process and relational outcomes of SIE, and (3) performance outcomes of SIE. We conclude that this area of research would benefit from increased focus on the nature of the relationship between the exchange partners, the broader social context in which exchanges are embedded, consideration of the capabilities of the actors and their task requirements, and timing of events. Issues regarding SIE quality and fidelity, motivations, cultural distance, and uncertainty are discussed. These research directions can potentially enhance diverse literatures, such as interpersonal interactions, team decision making, knowledge transfer, and corporate governance.

Bayesian estimation and inference have been core features of scientific knowledge generation since the work of Sir Thomas Bayes was built upon by Pierre-Simone Laplace from the late 1700s through the early 1800s. Although present-day statistical analysis in organizational research is “frequentist” in nature (due to the influence of scholars such as Sir Ronald Fisher and Jerzey Neyman), the past 20 years has seen a veritable explosion of Bayesian applications across the social and physical sciences. This special issue highlights these applications and the many opportunities they carry, including precise and flexible methods for testing hypotheses and very intuitive ways of describing results. For this special issue, three editorial commentaries were solicited from world-renowned experts in statistics, probability, and their historical and current applications. These papers offer a view from outside management, giving fresh insight into topics that are rarely covered in management research, including critical perspectives on existing paradigms in our field and recommendations for improvements in statistical methods and research design. The topics covered relate to (a) the apparent desire for universal or default methods of inquiry and inference—whether Bayesian or frequentist—which narrows researchers’ focus and reduces their ability to develop and deploy a larger “toolbox” of methods and approaches; (b) the many limitations of existing frequentist tools, which tend to be underestimated or ignored because of their institutionalized and habitual nature; and (c) the existence and importance of

In the management literature, generalizability theory (GT) has been typically used to investigate the reliability of assessment center and job performance ratings. However, the management field has yet to take full advantage of the information GT can offer regarding the reliability of measurement. It is likely that GT has not been adopted because of the complexities involved with its notation and practical application. Moreover, current methods for obtaining accurate interval estimates around estimated variance components or their reliability coefficients are not easily implementable. Alternatively, Bayesian methods provide a different method for estimating GT variance components. Bayesian methods enable management researchers to estimate the posterior distributions of each GT variance component as well as the GT reliability coefficients. From these posterior distributions, researchers can easily obtain the interval estimates for each variance component and the corresponding reliability estimates. Conducting two studies, the authors examine what priors should be used when conducting a Bayesian GT analysis and what estimates should be used to summarize a variance component’s posterior distribution. Additionally, the authors find that under certain conditions, Bayesian methods perform better than frequentist methods.