Constructing Personality Maps, Mapping Personality Constructs: Multidimensional Scaling Recovers the Big Five Factors from Internal and External Structure

Abstract

This report examines the structure of similarities underlying the lexicon of personality-trait description, when "similarity" is defined and measured in terms of (a) semantic judgment and (b) covariance in actual use. A lexicon of 60 trait adjectives was examined, using several procedures for collecting semantic judgments. Similarity data of both kinds were analyzed with multidimensional scaling (MDS) to provide a parsimonious representation of underlying structure. The convergence between semantic judgments and covariance within trait-attribution data was substantial; both kinds of data evinced the same structure when collected for subsets of adjectives. Canonical correlation was employed to find the number of dimensions shared across MDS solutions. Interpretation of the results was facilitated by individual-differences MDS, which can select an optimal set of underlying dimensions, and at the same time accommodate the differences between data sets that arise when data-collection procedures differ in the relative emphasis they place upon those dimensions. We interpret the small number and shared nature of the dimensions by arguing that the lexicon's structure relates to trait perception rather than personality structure per se, even when probed with trait-attribution covariance. Keywords: Big Five, trait perception, internal structure, multidimensional scaling Este trabajo examina la estructura de las similitudes subyacentes al léxico de la descripción de los rasgos de personalidad, cuando "similitud" se define y se mide en términos de: (a) juicio semántico y (b) covarianza en el uso actual. Se examinó un léxico de 60 adjetivos de rasgos, empleando varios procedimientos para recoger juicios semánticos. Los datos de similitud de ambos tipos se analizaron con escalonamiento multidimensional (EMD) para obtener una representación parsimoniosa de la estructura subyacente. La convergencia entre los juicios semánticos y la covarianza rasgo-datos atribucionales era sustancial; ambos tipos de datos mostraban la misma estructura cuando se recogían para subconjuntos de adjetivos. Se empleó la correlación canónica para encontrar el número de dimensiones compartidas por las soluciones EMD. La EMD de diferencias individuales facilitó la interpretación de los resultados porque puede seleccionar un conjunto óptimo de dimensiones subyacentes y, al mismo tiempo, adaptar las diferencias entre los conjuntos de datos que emergen cuando los procedimientos de recogida de datos difieren con respecto al énfasis relativo que se concede a dichas dimensiones. Nosotros interpretamos el pequeño número y la naturaleza compartida de las dimensiones arguyendo que la estructura del léxico se relaciona más con la percepción de los rasgos que con la estructure de la personalidad en sí, incluso cuando se analiza mediante la covarianza rasgo-atribución. 2007, Vol. 10, No. 1, 68-83 ISSN 1138 Acknowledgements: This research would have been impossible without the informants, including students who found the task of arranging words into groups more co...

Convergence of Internal and External Structure for the California Child Q-set

“…However, the fourth dimension or distinction made among CCQ items (replicated in a 55-item subset) is not so easily assimilated to the D4 found for adjectives (glossed in Bimler & Kirkland, 2007, as ‘emotionality’ or ‘abstract/concrete’). In both cases the fourth dimension was of minor importance.…”

“…For these reasons, we opted to concentrate on four-dimensional solutions such as G 4 for subsequent analyses. Retaining four dimensions from MDS will also facilitate a comparison with the internal structure of trait adjectives (Bimler & Kirkland, 2007). …”

“…Individual entries were calculated according to the algorithm in Bimler and Kirkland (2007, Appendix B).…”

“…Multidimensional scaling of the adjusted item set reveals a tightly-clustered arrangement (e.g. Bimler & Kirkland, 2007) in which factor poles are isolated by these artificial or exaggerated voids. In contrast, the CCQ was designed to be broadly inclusive, so that theory-driven expectations would not restrict its capacity to capture data.…”

“…If they prove to be compatible, the various matrices can be combined into a single consensus map of trait space, and we will be able to pose the secondary research question noted earlier: how far that map’s dimensions correspond to those found in MDS studies of single-word descriptors (e.g. Bimler & Kirkland, 2007). …”