Partial scalar invariance and observed differences across gender in a reasoning test battery / Invarianza escalar parcial y diferencias observadas entre sexos en una batería de tests de razonamiento

BACKGROUND: The substantive basis of the Reasoning Test Battery (BPR) is the theory of the hierarchical organization of cognitive abilities and therefore, it combines a general cognitive factor and specific factors associated with abstract, numerical, verbal, practical, spatial and mechanical reasoning. The battery has three forms, covering an age range from 9 to 22 years. METHOD: The present study analyzes the internal structure of the Basque version of the battery using exploratory and confirmatory factor analyses. Factorial invariance studies across gender were performed and partial differences observed were analyzed in a sample of 1,923 students. RESULT: The results concluded: (a) the presence of one general reasoning factor in each of the forms, (b) partial scalar invariance across gender affecting mechanical reasoning and numerical reasoning, (c) no differences in the general reasoning factor, and (d) negligible observed differences in partial scales. CONCLUSIONS: Tests for measurement invariance indicate differences in factor intercepts, cautioning that comparisons of observed g scores across gender are not appropriate

ANTECEDENTES: construida sobre la teoría de la organización jerárquica de las habilidades cognitivas, la Batería de Pruebas de Razonamiento (BPR) combina un factor de razonamiento general y factores específicos asociados con el razonamiento abstracto, numérico, verbal, práctico, espacial y mecánico. La batería tiene 3 Formas que cubren un rango de edad entre 9 y 22 años. MÉTODO: se analizó la estructura interna de la versión en euskera de la batería por medio de análisis factoriales exploratorios y confirmatorios. Se llevaron a cabo estudios de invarianza factorial en función del sexo y se analizaron las diferencias observadas en las escalas parciales en una muestra de 1.923 estudiantes. RESULTADOS: los resultados concluyeron: (a) la presencia de un factor de razonamiento general en cada una de las formas, (b) la invarianza escalar parcial que afectan a las escalas de razonamiento mecánico y razonamiento numérico, (c) la no diferenciación entre sexos en el factor general, y (d) diferencias mínimas en las escalas parciales. CONCLUSIONES: los resultados del test de invarianza factorial apuntaron la presencia de valores interceptales diferentes, lo cual desaconseja la comparación de puntuaciones observadas g en función del sexo

Internal structure and partial invariance across gender in the Spanish version of the reasoning test battery

The Reasoning Test Battery (BPR) is an instrument built on theories of the hierarchical organization of cognitive abilities and therefore consists of different tasks related with abstract, numerical, verbal, practical, spatial and mechanical reasoning. It was originally created in Belgium and later adapted to Portuguese. There are three forms of the battery consisting of different items and scales which cover an age range from 9 to 22. This paper focuses on the adaptation of the BPR to Spanish, and analyzes different aspects of its internal structure: (a) exploratory item factor analysis was applied to assess the presence of a dominant factor for each partial scale; (b) the general underlined model was evaluated through confirmatory factor analysis, and (c) factorial invariance across gender was studied. The sample consisted of 2624 Spanish students. The results concluded the presence of a general factor beyond the scales, with equivalent values for men and women, and gender differences in the factorial structure which affect the numerical reasoning, abstract reasoning and mechanical reasoning scales(AU)

No disponible

Procedimientos analítico-racionales en la adaptación de tests. Adaptación al español de la batería de pruebas de razonamiento: Adaptation to Spanish of the Reasoning Tests Battery / Judgmental-analytical procedures for adapting tests

Adaptar un test entre culturas o idiomas requiere considerar aspectos legales, culturales, lingüísticos, métricos y de uso. La atención prestada por la literatura especializada a los aspectos metodológicos implicados en el estudio de la equivalencia métrica ha sido significativamente mayor que la dedicada a los procedimientos analítico-racionales previos a la fase de confirmación empírica. Sin embargo, la consideración de estos últimos es crucial en el proceso de adaptación. En esta línea, el objetivo de este trabajo es incidir en la relevancia de las etapas previas y describirlas, ofreciendo una sistematización que considera diez apartados. Todos ellos contribuyen a garantizar y favorecer la construcción de un test adaptado y equivalente, en la medida de lo posible, al test original. El proceso es ejemplificado con la adaptación al español de una prueba cognitiva construida en portugués para la población portuguesa, la Bateria de Provas de Raciocínio.

Adapting a test between cultures or languages requires taking into account legal, linguistic, metric, and use-related considerations. Significantly more attention has been paid to the methodological aspects involved in the study of metric equivalence than to judgmental-analytical procedures prior to the empirical confirmation stage. However, considering the latter is crucial in the adaptation process. Along these lines, this paper seeks to describe and focus on the relevance of the previous stages, thereby offering a systematization process that comprises ten sections. This approach contributes to ensuring the construction of a test adapted and equivalent in as much as possible to the original. This process is exemplified by means of a Spanish language adaptation of a cognitive test originally designed in Portuguese for the Portuguese population, the Reasoning Test Battery.