Abstract
Whereas variable-centered approaches (e.g., multiple regression, CFA, SEM) assume that all individuals from a sample are drawn from a single population for which a single set of “averaged” parameters can be estimated, person-centered approaches (e.g., mixture models, latent profile analyses, latent class analyses, mixture regression, growth mixture analyses) relax this assumption and considers the possibility that the sample might include multiple subpopulations characterized by different sets of parameters. Person-centered approaches thus provide a very rich complement to traditional variable-centered methods, allowing researchers to model complex processes in a more heuristic way (Wang & Hanges, 2011). Such methods are powerful tools to model interactions between large set of variables in an effective way. They identify naturally occurring subpopulations and allow for further comparisons between them.
The raising popularity of person-centered approaches brings an opportunity for a more in-depth understanding of work-related phenomena. However, it also presents multiple challenges. For example, variable- and person-centered approaches are often framed as distinct and complementary approaches to research, yet this complementary nature has seldom been systematically explored. A key issue that has yet to be explored in research is the impact of key decisions made in preliminary variable-centered applications on later person-centered results. Another key challenge is to further explore the full flexibility of the generalized structural equation modeling (GSEM) framework, which provides a very high level of flexibility to person-centered analyses over and above their classical formulations (e.g., correlated uniquenesses, heteroscedasticity, parameter constraints). Indeed, very little research carefully attempts to understand the meaning, in terms of theoretical implication, of these more flexible specifications made possible by GSEM. Another unexplored critical issue is that most applications of person-centered methods have been limited to providing a description of the results obtained in a specific sample, with at most an eyeball comparison of solutions obtained in multiple samples. This leaves open the question of whether (and how) results obtained from different studies using person-centered methodologies may be replicated, compared, or aggregated in a more quantitative manner? Finally, although it has often been argued that person-centered approaches should support theory development, strategies for both exploratory and confirmatory applications are needed.
Aims of this Feature Topic
The aim of this Feature Topic is to highlight the benefits and limitations presented by person-centered research methods within the managerial and organizational research, and to illustrate recent developments in this methodological area. In line with Organizational Research Methods’ policies, we equally welcome proposals with a dominant methodological focus, proposals with a dominant pedagogical focus, as well as proposals with a dominant substantive (or rather, substantive-methodological synergy) focus. These later types of papers should aim to illustrate how these methods may be used to address important research questions. Submissions to this special issue could cover, but are not restricted to, the following topics: The complementarity of variable- and person-centered approaches in organizational research. The complementarity of exploratory and confirmatory approaches to person-centered analyses. The theoretical meaning of alternative parameterizations of person-centered analyses estimated within the Generalized Structural Equation Modeling Framework. Cross-validation of subpopulations across diverse samples (e.g., investigating whether similar subpopulations occur across cultures, gender groups, and organizations). Examining the stability and change of subpopulations over time (e.g., examining transitions between latent or observed subpopulations). Using person-centered approaches in a psychometric validation process, to explore the underlying multidimensionality of psychometric constructs. Using person-centered approaches to establish and validate diagnostic classification methods for workplace phenomena. Extending the use of meta-analytic procedures to person-centered research. Small sample approaches to person-centered analyses. Using person-centered approaches to validate interaction effects (e.g., whether and how a combination of variables predicts important outcomes).
Editorial procedure
In order to be considered for publication in this feature topic, a 5-7 page (double-spaced) proposal/summary should be sent by email to
The Guest Editors for this Feature Topic are:
Alexandre J.S. Morin, Institute for Positive Psychology and Education, Australian Catholic University (
Marylene Gagne, Department of Psychology, University of Western Australia (
Aleksandra Bujacz, Karolinska Institute, Sweden (
Timeline
Announce the Call for White Papers, October 15, 2015
White Papers Submitted by March 31, 2016
Feedback on White Papers by June 15, 2016
Complete Manuscripts Submitted by October 31, 2016
Feedback on Complete Manuscripts by January 31, 2017
Revised Manuscripts Submitted by May 31, 2017
Decisions on Revised Manuscripts by August 31, 2017