A Brief Review of Contextual-Effect Models and Measurement

Bibliographic Data

ID14633355
AuthorsDouglas L Anderton (University of Massachusetts Amherst), Deborah E Sellers (0000-0003-3265-217X, University of Massachusetts Amherst)
Year1989
Volume22
Issue3
Pages106-115
Publication date1989-07-01
Open AccessNo
TypeARTICLE
VenueHistorical Methods A Journal of Quantitative and Interdisciplinary History (JOURNAL)
Journal identifiersISSN: 0161-5440 • E-ISSN: 1940-1906
PublisherInforma UK Limited (PUBLISHER • GB)
DOI10.1080/01615440.1989.10594184
OpenAlexW2320625529
LanguageEN
Citations received4
References cited9

The intent was to highlight the practical methods and issues of contextual analysis rather than detail the statistical characteristics of any one contextual model. There is too much emphasis in the discussion of contextual effects on forcing theoretical hypotheses into the single-level linear regression model. Similarly there is too much dislike for basic contextual analyses that avoid indiscriminately collapsing reality into a single linear dimension. There is no one correct method for analyzing a contextual effect because there are clearly distinct theoretical and methodological notions of what a contextual effect is. The models reviewed differ primarily according to whether the context is environs characterized by differing types of individuals a characteristic membership of the individual a distinct identifiable environment with unique impact upon individual outcomes or a similar setting with identifiable effects upon causal relationships themselves. The basic question pertains to what a context is and whether it is possible to meaningfully incorporate contexts into explanations of human action. The force of the mechanistic principles embodied in methodologies often accommodate such influences through a forced decomposition of reality into characteristics of the individual rather than those of the environment. Five major linear models represent a general statement of relationship that might also be pursued with a variety of analytical methods other than linear modeling. The first two of these models (individual-effects and macro-effects) are confined to a single level of analysis. The third model is the simplest of those that can be considered a contextual-effect analysis in that both individual and contextual terms are included. The distinction between the simple conditional probability model (Model 3) and the contextual-effects model (Model 4) is the inclusion of both context and individual-level effects. With Model 5 it is possible to model individual-level relationships and contextual effects on the relations

Cognitive psychology · Context analysis · Contextual design · Embodied cognition · Linear model · Machine learning · Multilevel model · Perception · Advanced Text Analysis Techniques · Cognitive Science and Mapping · Computer Science · Mathematics · Multi-Criteria Decision Making · Psychology · Artificial Intelligence

Tools

Open DOISci-Hub