Modeling consensus emergence in groups using longitudinal multilevel methods

Abstract

Organizational researchers have long been interested in studying bottom-up multilevel processes where lower level units (e.g., employees) in organizations interact to jointly create characteristics of higher level units (e.g., work groups). This article contributes to the literature on bottom-up processes by detailing a statistical approach-the consensus emergence model (CEM)-that allows researchers to study emergence of shared perceptions and feelings or climates in groups over time. The described methodological approach extends standard multilevel methodology by examining residual variances within a growth model to account for dynamic change in group consensus. The CEM provides a formal test for consensus emergence. The approach also allows researchers to test explanatory models of consensus emergence by including personlevel, group-level, and observation-level predictors. We illustrate the CEM by applying the method to data from two longitudinal studies of work units. The first study investigated job satisfaction in military companies. Our second study examined professional archeologists working in groups on a field excavation mission and focused on fatigue at the end of the work day. Our analyses demonstrate the CEM's ability to detect and study emergence, and suggest that the CEM may be a valuable tool to help extend the study of emergence in organizational research.

The emergence of team resilience: A multilevel conceptual model of facilitating factors

“…In such cases, team members’ perceptions should cluster around the mean and therefore represent a meaningful representation of a higher‐level construct (James, ). The consensus emergence model is a statistical approach developed recently to facilitate such tests of bottom‐up emergent processes (see Lang & Bliese, in press; Lang, Bliese, & Voogt, ).…”

Specifying Location–Scale Models for Heterogeneous Variances as Multilevel SEMs

“…To demonstrate how specifying the model in the multilevel SEM framework can help address difficulties that currently exist in organizational research, consider the case study where repeated measures are nested within people who are further nested within groups. One such example appears in Lang et al (2018), who describe a three-level model with this structure for consensus emergence. One…”

“…Table 3 Comparison of estimates from the Model 3c in Lang et al (2019), the reproduction of that model as a Bayesian multilevel SEM, and the extension of the model that allows for scale random effects estimated as a multilevel SEM. Note: Model 3c = the model originally presented in Table 9 of Lang et al (2018). ML-SEM = Multilevel Structural Equation Model.…”

“…Estimates from models estimated with Bayesian methods are the median of the posterior distribution for each parameter. 1 Covariances are reported in Lang et al (2018), so we provide covariances estimates for ease of comparison rather than correlation as we present in previous tables.…”

“…Despite the frequency of questions about cohesion and heterogeneity in the organizational literature, Lang et al (2018Lang et al ( , 2019 for a heterogeneous variance model in a mixed effect framework and would likely require a multistage model whose estimates would require Croon's correction (Croon, 2002) in order to be unbiased (Devlieger & Rosseel, 2017;Kelcey, Cox, & Dong, 2019), mediation is another straightforward extension if modeling the random scale effects in the multivariate multilevel SEM framework (Hoffman, 2019). McNeish and Hamaker (2020) note such advantages of multilevel SEM over mixed effects models, noting that multilevel SEM allows any hypothesized structural relations to be modeled between random effects.…”