Simple but not simpler

Coen, Corinne A.

doi:10.1007/s10588-008-9041-3

Cited by 17 publications

(12 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Model Extensionsmentioning

confidence: 99%

“…In addition to those conceptual targets, there are a number of parameters and functional relationships that could be built upon the parameters listed in Table B1 to introduce and/or represent more complex relationships of interest. In general, there are important tradeoffs when adding new complexities to a computational model (Coen, 2009); however, considering how and where to build upon or integrate new ideas into a theoretical model can often prove a useful exercise for refining and improving computational theory.For example, elements related to motivation and/or characteristics of the testing environment could be introduced into SiRJ through both the Tmax and sjtMax parameters. Tmax is a parameter of Thomas et al’s (2008) HyGene model that determines the maximum number of retrieval failures an individual will endure before terminating semantic memory search for plausible interpretations during conditional reasoning.…”

Section: Model Operationalizationsmentioning

confidence: 99%

See 1 more Smart Citation

A general response process theory for situational judgment tests.

Grand

2020

Journal of Applied Psychology

View full text Add to dashboard Cite

Situational judgment tests (SJTs) have emerged as a staple of assessment methodologies for organizational practitioners and researchers. Despite their prevalence, many questions regarding how to interpret respondent choices or how variations in item construction and instruction influence the nature of observed responses remain. Existing conceptual and empirical efforts to explore these questions have largely been rooted in reflexive psychometric measurement models that describe participant responses as indicative of (usually multiple) latent constructs. However, some have suggested that a key to better understanding SJT responses lies in unpacking the judgment and decision-making processes employed by respondents and the psychological and contextual factors that shape how those processes play out. To this end, the present paper advances an integrative and generalizable process-oriented theory of SJT responding. The framework, labeled Situated Reasoning and Judgment (SiRJ), proposes that respondents engage in a series of conditional reasoning, similarity, and preference accumulation judgments when deciding how to evaluate and respond to an SJT item. To evaluate the theory's plausibility and utility, the SiRJ framework is translated into a formal computational model and results from a simulation study are analyzed using neural network and Bayesian survival analytic techniques that demonstrate its capability to replicate existing and new empirical effects, suggest insights into SJT interpretation and development, and stimulate new directions for future research. An interactive web application that allows users to explore the computational model

show abstract

Section: Model Extensionsmentioning

confidence: 99%

Section: Model Operationalizationsmentioning

confidence: 99%

A general response process theory for situational judgment tests.

Grand

2020

Journal of Applied Psychology

View full text Add to dashboard Cite

show abstract

“…(p. 109)Although an emphasis on simple rules in developing theory for emergent phenomena may fit well with the general suggestion of increased use of computational/agent-based models (Kozlowski, 2015; Kozlowski et al, 2013), we believe more specific and systematic guidance is needed about how researchers can deploy this perspective in theory development (cf. Coen, 2009; Davis, Eisenhardt, & Bingham, 2007; Harrison, Lin, Carroll, & Carley, 2007). Moreover, theory building techniques emphasizing simple rules can also likely be leveraged without using computational/agent-based models.…”

Section: The Futurementioning

confidence: 99%

Multilevel Methods and Statistics: The Next Frontier

Eckardt

Yammarino

Dionne

et al. 2020

Organizational Research Methods

View full text Add to dashboard Cite

The purpose of this article is to take stock of extant multilevel methodological and statistical work and highlight needed areas for future research. A basic overview of the history and progression of multilevel methods and statistics in the organizational sciences is provided, as well as a discussion of recent developments to summarize the current state of the science. The eight articles in the current feature topic are also summarized and integrated to depict several themes and directions for the next wave of multilevel methods and statistics. Last, to highlight what still needs to be accomplished in the field, several unresolved issues and future research topics are noted and an agenda related to future multilevel work is discussed.

show abstract

“…That comparison seems us irrelevant; in fact, this generator should rather be compared with other descriptive models of real networks, like p * models, that can hardly be applied in agent-based modelling. "Simple, but not simpler" (Coen, 2009); the model as to be simpler than the reality, but shouldn't become too simple and failing to reach its purpose. Our model was built especially for helping modellers during their modelling process, and we tried to find a good tradeoff between complexity and descriptivity.…”

Section: A Complex Solutionmentioning

confidence: 99%

Generate Descriptive Social Networks for Large Populations from Available Observations: A Novel Methodology and a Generator

Thiriot¹

2020

Preprint

View full text Add to dashboard Cite

When modeling a social dynamics with an agent-oriented approach, researchers have to describe the structure of interactions within the population. Given the intractability of extensive network collecting, they rely on random network generators that are supposed to explore the space of plausible networks. We first identify the needs of modelers, including placing heterogeneous agents on the network given their attributes and differentiating the various types of social links that lead to different interactions. We point out the existence of data in the form of scattered statistics and qualitative observations, that should be used to parameter the generator. We propose a new approach peculiar to agent-based modeling, in which we will generate social links from individuals' observed attributes, and return them as a multiplex network. Interdependencies between socioeconomic attributes, and generative rules, are encoded as Bayesian networks. A methodology guides modelers through the formalization of these parameters. This approach is illustrated by describing the structure of interactions that supports diffusion of contraceptive solutions in rural Kenya.

show abstract

Simple but not simpler

Cited by 17 publications

References 1 publication

A general response process theory for situational judgment tests.

A general response process theory for situational judgment tests.

Multilevel Methods and Statistics: The Next Frontier

Generate Descriptive Social Networks for Large Populations from Available Observations: A Novel Methodology and a Generator

Contact Info

Product

Resources

About