Evaluación TRI de la escala I-E de Rotter: un nuevo enfoque y algunas consideraciones / An IRT assessment of Rotter’s I-E scale: a new approach and some considerations

Se evalúa el funcionamiento y propiedades de medida de la versión española de la escala de Locus de Control de Rotter; dicha evaluación utiliza un nuevo enfoque basado en la teoría de respuesta al ítem. Nuestro procedimiento permite: (a) detectar el impacto de los sesgos de respuesta (aquiescencia y deseabilidad social); (b) evaluar de forma clara la dimensionalidad y estructura del test; (c) evaluar las posibles causas del deficiente funcionamiento de algunos ítems; y (d) incrementar la precisión de la medida. Globalmente, los resultados no contradicen lo obtenido en estudios previos, pero permiten matizar algunas de las críticas que ha recibido la escala. Se proponen algunas recomendaciones para su uso (AU)

This study assessed the functioning and measurement properties of Rotter’s Locus of Control Scale in its Spanish version by using a new approach based on Item Response Theory. Our procedure allows us: (a) to detect the impact of response biases (acquiescence and social desirability); (b) to evaluate clearly the test dimensionality and structure; (c) to evaluate and understand the causes of some items’ inefficiency; and (d) to increase measurement precision. Globally, the results do not contradict those obtained in previous research, but allow some of the criticisms the scale had received to be qualified. Some recommendations for use are proposed (AU)

[An IRT assessment of Rotter's I-E scale: a new approach and some considerations]. / Evaluación TRI de la escala I-E de Rotter: un nuevo enfoque y algunas consideraciones.

This study assessed the functioning and measurement properties of Rotter's Locus of Control Scale in its Spanish version by using a new approach based on Item Response Theory. Our procedure allows us: (a) to detect the impact of response biases (acquiescence and social desirability); (b) to evaluate clearly the test dimensionality and structure; (c) to evaluate and understand the causes of some items' inefficiency; and (d) to increase measurement precision. Globally, the results do not contradict those obtained in previous research, but allow some of the criticisms the scale had received to be qualified. Some recommendations for use are proposed.

The impact of acquiescence on forced-choice responses: a model-based analysis

El objetivo general del presente estudio es evaluar la potencial utilidad que tiene el formato de elección forzosa para reducir el impacto de la respuesta aquiescente. Las contribuciones con vistas a conseguir este objetivo son de dos tipos: metodológicas y substantivas. Metodológicamente, se propone un procedimiento basado en un modelo de análisis que, a su vez, se desarrolla a partir de un mecanismo de respuesta. Dicho procedimiento permite una evaluación rigurosa del impacto de la aquiescencia. Substantivamente, el procedimiento propuesto se aplica a tres cuestionarios con formato de elección forzosa ampliamente utilizados en personalidad: La escala de Locus de Control de Rotter, la escala de búsqueda de sensaciones de Zuckerman y la escala de aumentaciónreducción de Vando. Los resultados sugieren que el impacto de la aquiescencia es mínimo excepto en el caso de la escala de Rotter. Estos resultados diferenciales podrían explicarse por ciertas características de la escala que se discuten en el artículo(AU)

The general aim of the present study is to assess the potential usefulness of the normative Forced Choice (FC) format for reducing the impact of acquiescent responding (AR). To this end it makes two types of contributions: methodological and substantive. Methodologically, it proposes a model-based procedure, derived from a basic response mechanism, for assessing the impact of AR. Substantively, it applies the procedure in three large datasets, which use well-known normative FC personality questionnaires: Rotter’s Locus of Control Scale, the Sensation Seeking Scale (form V), and Vando’s Reducer-Augmenter Scale. The results suggest that the impact of AR is minimal except for the Locus of Control Scale, so FC-based questionnaires are a good alternative for controlling acquiescence. The different results obtained with the LOC Scale might be explained for some particularities of this scale which are discussed in the paper(AU)

Two SPSS programs for interpreting multiple regression results.

When multiple regression is used in explanation-oriented designs, it is very important to determine both the usefulness of the predictor variables and their relative importance. Standardized regression coefficients are routinely provided by commercial programs. However, they generally function rather poorly as indicators of relative importance, especially in the presence of substantially correlated predictors. We provide two user-friendly SPSS programs that implement currently recommended techniques and recent developments for assessing the relevance of the predictors. The programs also allow the user to take into account the effects of measurement error. The first program, MIMR-Corr.sps, uses a correlation matrix as input, whereas the second program, MIMR-Raw.sps, uses the raw data and computes bootstrap confidence intervals of different statistics. The SPSS syntax, a short manual, and data files related to this article are available as supplemental materials from http://brm.psychonomic-journals.org/content/supplemental.

The external validity of scores based on the twoparameter logistic model: some comparisons between IRT and CTT

Una de las ventajas teóricas de los modelos de la teoría de respuestaal ítem (TRI) es que las estimaciones de los niveles en el rasgo son másprecisas (en términos de información) que cualquier otro tipo de puntuación.Sin embargo, no está claro aún que el uso de estas estimaciones másprecisas permita mejorar la validez externa con respecto a la validezobtenida con las puntuaciones directas. Se presentan algunos resultadosbasados en el modelo logístico de dos parámetros, y se discuten tresaspectos: (a) Cómo deben interpretarse los coeficientes de validez basados en puntuaciones TRI; (b) como pueden hacerse inferencias con respecto aestos coeficientes, y (c) qué mejoras en validez cabe esperar cuando elmodelo es correcto y se usan las puntuaciones TRI en lugar de laspuntuaciones directas. Cuatro estudios empíricos en personalidad aportanevidencia acerca de los resultados que cabe esperar en aplicaciones reales,en las que el modelo no es correcto sino tan sólo una aproximación. Entodos los ejemplos, el resultado general es que los coeficientes de validezbasados en puntuaciones TRI son muy similares a los obtenidos conpuntuaciones directas(AU)

A theoretical advantage of item response theory (IRT) models is that traitestimates based on these models provide more test information than anyother type of test score. It is still unclear, however, whether using IRT traitestimates improves external validity results in comparison with the resultsthat can be obtained by using simple raw scores. This paper discusses somemethodological results based on the 2-parameter logistic model (2PLM) andis concerned with three issues: first, how validity coefficients based on IRTtrait estimates must be interpreted; second, how inferences about thesecoefficients can be made; and third, which differences in external validitycan be expected if the 2PLM is correct for the data and IRT scores are usedin place of raw scores. Four empirical examples in the personality domainprovided further evidence for the results that can be expected in realresearch in which the model is, at best, a good approximation to the data. Ageneral result of these examples was that validity coefficients based on IRTscores were similar to those based on raw scores(AU)

Assessing extraversion and emotional stability in adolescents: development and validation of a questionnaire.

A reduced version of the Five-factor Personality Inventory, which only considers two of the five original dimensions, was constructed to assess Extraversion and Emotional Stability especially for a sample of 581 adolescents. This used Items 6, 9, 11, 19, 21, 39, 41, 46, 54, 59, 61, 66, 84, 89, 91, and 99. Exploratory and confirmatory factor analysis showed that the factor structure of the questionnaire is acceptable, and both scales are reliable and correlated with similar measures as The Eysenck Personality Questionnaire-Revised.

Belief in astrology inventory: development and validation.

After the paper by Mayo, White, and Eysenck in 1978, a considerable number of papers studied the so-called sun-sign-effect predicted by astrology: people born with the sun in a positive sign are supposed to be extraverted, and those with the sun in a negative sign are supposed to be introverted. In these papers, researchers used ad hoc questionnaires with a few questions related to belief, knowledge, experience, or attitude toward astrology. However, an appropriate inventory with known psychometric properties has yet to be developed to assess the belief in astrology. In the present paper, the Belief in Astrology Inventory is presented with some psychometric data. The participants were 743 undergraduates studying Psychology and Social Sciences at a university in Spain. Correlation of scores on Belief in Astrology and Extraversion was small but significant (r = .22; r2 = .04) for positive sun-sign participants. This value accounts for negligible common variance. Women had significandy higher scores on the inventory than men.

Aplicación del psicodiagnóstico miokinético revisado (PMK-R) al estudio de la violencia / Aplicattion of the revised myokinetic psychodiagnostic (MKP-R) test to the study of violence

Este artículo informa de los resultados obtenidos mediante la réplica de un clásico trabajo previo de Emilio Mira (Mira, 1949) sobre la relación existente entre respuestas miokinéticas y violencia. Las medidas miokinéticas se obtuvieron a partir de las acciones que el propio cuerpo genera, o facilitaciones neuromusculares propioceptivas, en los músculos del brazo y hombro, mediante la utilización de una tarea basada en un método gráfico. Los resultados confirman, por primera vez estadísticamente, la hipótesis de Mira de que la irritabilidad, la extroversión y la agresividad son mayores en las personas violentas. Así como que las personas violentas manifiestan menor coherencia entre su mano dominante y no dominante que las personas normales. Además aparecen diferencias estadísticas entre personas violentas y no violentas en su nivel emocional, en contra de lo que había encontrado Mira. Por consiguiente, consideramos que la nueva técnica que hemos utilizado constituye un método más fiable de evaluación no verbal de la personalidad, complementario de los actuales cuestionarios de personalidad (AU)

This paper inform of the results obtained by a replication of a classical work of Emilio Mira (Mira, 1949) on relationship between individual myokinetic responses and violence. The myokinetic measures were obtained from the actions that the body itself generate, or proprioceptive neuromuscular facilitations, in the arm and the shoulder muscles, by means of a graphic method task. This is the first time that the hypothesis of Mira, in the sense of irritability, aggressiveness and extraversion are higher in violent people, is statistically confirmed. Otherwise, that violent people exhibit lesser coherence between dominant and not dominant hand than normal people. Furthermore, statistical differences appear between violent and non violent people on the emotional level, as on the contrary suggested the previous Mira’s work. Consequently, we consider that the new technique that we used in this work could be a more reliable instrument for the non verbal assessment of personality, complementary of current verbal questionnaires of personality (AU)

Unrestricted Factor Analytic Procedures for Assessing Acquiescent Responding in Balanced, Theoretically Unidimensional Personality Scales.

This article describes and proposes an unrestricted factor analytic procedure to: (a) assess the dimensionality and structure of a balanced personality scale taking into account the potential effects of acquiescent responding, and (b) correct the individual trait estimates for acquiescence. The procedure can be considered as an extension of ten Berge's (1999) principal component-based approach, and is intended for scales that attempt to measure a single trait. First, the rationale of the procedure is described for the correlation and covariance cases, and the relations with an alternative restricted model are discussed. Next, additional criteria for supporting the proposed unrestricted solution are suggested. The procedure and the additional criteria are illustrated using empirical data. The usefulness of the procedure in applied personality measurement is discussed.

Propiedades psicométricas del test de optimismo Life Orientation Test / Psychometric properties of the `Life Orientation Test´ (LOT)

El presente estudio evalúa las propiedades psicométricas de un test de optimismo (LOT-R) adaptado al castellano mediante el modelo multidimensional de respuesta graduada (MMGR). El estudio tiene dos vertientes: sustantiva y metodológica. Desde el punto de vista sustantivo interesa evaluar las propiedades de un test que, en su versión original, muestra interesantes propiedades cara a la investigación aplicada en personalidad. Desde le punto de vista metodológico se pretende ilustrar el uso de un modelo que, hasta ahora, no ha sido prácticamente utilizado en investigación aplicada, así como evaluar sus potenciales ventajas teóricas. Los resultados sustantivos indican que la adaptación del LOT-R tiene propiedades muy similares a las de la versión original. Desde el punto de vista metodológico, el MMRG presenta una serie de ventajas respecto al análisis tradicional. Los resultados, tanto sustantivos como metodológicos, se discuten desde un punto de vista aplicado (AU)

An empirical application of the Multidimensional Graded Response Model. This study examines the psychometric properties of the LOT-R Spanish adaptation using the multidimensional graded response model (MGRM). The study has two purposes: substantive and methodological. From a substantive point of view we want to assess the behaviour of a test which, in its original version, has interesting properties in applied personality measurement. From a methodological point of view our study illustrates the application of the MGRM in the personality domain, and allows us to assess whether the theoretical advantages of the MGRM are important from an applied point of view. The substantive results suggest that the properties of the Spanish adaptation are similar to those of the original test. The methodological results show that the model has some advantages with respect to traditional analyses. Both type of results are discussed from an applied point of view (AU)

Adaptación y análisis psicométrico de la escala de deseabilidad social de Marlowe y Crowne / A Spanish version of the Marlowe and Crowne' s social desirability scale

El presente estudio describe la adaptación al castellano de la escala de deseabilidad social de Marlowe y Crowne (SDS) y estudia sus propiedades psicométricas desde el marco de la teoría de respuesta a los ítems (TRI). Se discute también la evolución del constructo de deseabilidad social y la relevancia y utilidad de la SDS. La versión adaptada se muestra como esencialmente unidimensional y posee propiedades psicométricas similares a las de la versión original. El modelo de TRI elegido se muestra apropiado y aporta información de interés respecto a las propiedades de la escala (AU)

A Spanish adaptation of the Marlowe and Crowne’s social desirability scale (SDS) based on the Item Response Theory (IRT) is presented. The historical evolution of both the social desirability construct and the SDS are revised. The Spanish version of the SDS behaves as essentially undimensional and has psychometric properties similar to those of the original version. The IRT model used to scale the SDS fits well the data and provides useful information about the properties of both the items and the total scale (AU)