Premium
Validity tests for componential analyses of conceptual domains: A cross‐cultural study in methodology
Author(s) -
Tzeng Oliver C. S.,
Osgood Charles E.
Publication year - 1976
Publication title -
behavioral science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.371
H-Index - 45
eISSN - 1099-1743
pISSN - 0005-7940
DOI - 10.1002/bs.3830210202
Subject(s) - psychology , attribution , affect (linguistics) , social psychology , cognition , construct (python library) , function (biology) , mediation , set (abstract data type) , cognitive psychology , computer science , communication , neuroscience , evolutionary biology , biology , programming language , political science , law
An analytic prediction model for evaluation of the validities of intuited denotative features in componential analyses is outlined and illustrated for a set of concepts referring to time. The intuited components function as independent variables and the measured affective meanings of the time‐related concepts function as dependent variables, both represented in Euclidean spaces. Predictions of attributions of affect from the denotative components proved to have very high predictive, construct and content validities for both the cross‐cultural (mean) and the majority of the culture‐indigenous interconcept squared distance matrices. Although this model for validity checking is limited to those components which are predictive of affect attribution, the external criterion, it represents a type of social system analysis in which objective culture as reflected in the denotative components for a conceptual domain is related to subjective culture, i.e., evaluative, potency and activity feelings, via the mediation of the cognitive systems of the individual human beings who make the semantic differential judgments. Similarities and differences in both objective (weights given to components) and subjective (variations in affect attribution) cultures can be quantitatively expressed.