Validation of the De Jong Gierveld Loneliness Scale in Peruvian old adults: a study based on SEM and IRT multidimensional models.

Loneliness is a public health problem. Its assessment is important to identify older adults who experience greater loneliness and appropriate interventions can be carried out. The De Jong Gierveld Solitude Scale (DJGLS) is one of the most widely used, at least in the European context, to measure loneliness. Although the Spanish version of the DJGLS has shown reliability and validity in Spanish samples of older adults, there is no evidence of adequacy in the Latin American context. Therefore, this study aims to evaluate the psychometric properties of the DJGLS in Peruvian older adults. Specifically, factorial validity, internal consistency and criterion-related validity were studied, based on a double analytical perspective: Classical Test Theory (SEM models) and Item Response Theory. The participants were 235 Peruvian older adults with ages ranging from 61 to 91 years old (Mean = 72.69, SD = 6.68) that assisted to Centers for the Elderly in the city of Trujillo, Peru. Together with the DJGLS, information from the three-item UCLA loneliness scale, Satisfaction with Life Scale, Brief Resilient Coping Scale and the Patient Health Questionnarie-2 was also gathered. The results support a one-dimensional structure of the DJGLS but with the presence of method effects associated to the negatively worded items. Additionally, the IRT multidimensional model applied also indicated the presence of a second dimension related to these negative items. Finally, other evidences of reliability and validity were adequate. In summary, the DJGLS may be considered a reliable and valid instrument to be used in the Peruvian (older adults) context.

Validity of the de Jong Gierveld Loneliness Scale in Spanish older population: competitive structural models and item response theory.

Loneliness is a subjective measure of one's state of mind and the negative feelings about one's level of social contact, which usually involves an unwanted discrepancy between existing and desired relationships. It is measured through self-reported questionnaires, among which a widely used one is the de Jong Gierveld Loneliness Scale (DJGLS). Although it has been previously tested and proven to be a reliable tool for measuring loneliness of Spanish older people, some relevant research questions remain unsolved. Therefore, this article aims to present a validity study of the Spanish version of the DJGLS for its use with older people, by establishing its factorial validity; studying the adequateness of a Rasch model underlying participants' responses; offering evidence of its reliability; and testing for convergent and discriminant validity. Data were obtained in a survey conducted with older adults attending Lifelong Learning programs in Valencia (Spain). The sample consisted of 335 people aged 55 years or older. Sociodemographic data were collected, and loneliness was measured with the DJGLS and the UCLA Loneliness Scale. Results have supported a unidimensional substantive structure, but with minor method effects associated with negatively worded items. In terms of multidimensional IRT analysis, the 2 Parameters Logistic Model fitted the data well and offered better information functions than the Rasch model. Reliability and criterion-related validity estimates were adequate.

The assessment of detachment among university students: Validation of the Recovery Experience Questionnaire in educational contexts / La evaluación del distanciamiento psicológico (detachment) en estudiantes universitarios: Validación del Cuestionario de Experiencia de Recuperación en contextos educativos

The objective of this study was to adapt and validate the Recovery Experience Questionnaire for use in an educational context. So far, this measure of detachment has mainly been used with workers. The study aimed to verify the factor validity and the criterion-related validity through correlation with self-regulation, engagement, coping strategies, and big-five personality dimensions in a sample of 468 Spanish university students. Several models proposed in the literature have been tested through confirmatory factor analysis. The results obtained confirmed the presence of four factors: psychological detachment, relaxation, mastery and control, as observed in previous studies within a workplace setting. Correlations obtained between these factors and the variables considered verified that detachment is significantly and positively associated with self-regulation, engagement, coping strategies, and certain types of personality; on the other side, detachment is significantly and negatively associated with irritation

El objetivo de este estudio fue adaptar y validar el Cuestionario de Experiencias de Recuperación (Recovery Experience Questionnaire) para su uso en un contexto educativo. Hasta el momento, la evaluación del detachment o distanciamiento psicológico ha sido utilizada fundamentalmente en contextos laborales. El estudio tuvo como objetivo verificar la validez factorial y la validez de criterio, a través de la correlación con variables como la autorregulación, el compromiso (engagement), las estrategias de afrontamiento, y las dimensiones de la personalidad contempladas en el modelo de los cinco grandes factores (Big Five Model) en una muestra de 468 estudiantes universitarios españoles. Varios modelos propuestos en la literatura han sido probados a través del análisis factorial confirmatorio. Los resultados obtenidos confirmaron la presencia de los cuatro factores siguientes: el distanciamiento psicológico, la relajación, el dominio, y el control. Estos son los mismos factores que se han observado en estudios previos dentro de los entornos laborales. Las correlaciones obtenidas entre estos cuatro factores y las variables consideradas verifican que el detachment está asociado de forma positiva y significativa con la autorregulación, el compromiso, las estrategias de afrontamiento, y ciertas dimensiones de personalidad. Por otro lado, el detachment mostró una relación significativa y negativa con variables de tipo negativo como la irritación

Traditional scores versus irt estimates on forced-choice tests based on a dominance model / Puntuaciones tradicionales y estimaciones TRI en tests de elección forzosa con un modelo de dominancia

BACKGROUND: Forced-choice tests (FCTs) were proposed to minimize response biases associated with Likert format items. It remains unclear whether scores based on traditional methods for scoring FCTs are appropriate for between-subjects comparisons. Recently, Hontangas et al. (2015) explored the extent to which traditional scoring of FCTs relates to the true scores and IRT estimates. The authors found certain conditions under which traditional scores (TS) can be used with FCTs when the underlying IRT model was an unfolding model. In this study, we examine to what extent the results are preserved when the underlying process becomes a dominance model. METHOD: The independent variables analyzed in a simulation study are: forced-choice format, number of blocks, discrimination of items, polarity of items, variability of intra-block difficulty, range of difficulty, and correlation between dimensions. RESULTS: A similar pattern of results was observed for both models; however, correlations between TS and true thetas are higher and the differences between TS and IRT estimates are less discrepant when a dominance model involved. CONCLUSIONS: A dominance model produces a linear relationship between TS and true scores, and the subjects with extreme thetas are better measured

ANTECEDENTES: los tests de elección forzosa (TEFs) fueron propuestos para reducir los sesgos de respuesta de ítems tipo Likert. Se cuestiona que los métodos de puntuación tradicional (PT) empleados permitan hacer comparaciones entre-sujetos. Recientemente, Hontangas et al. (2015) exploraron cómo las PTs obtenidas con diferentes TEFs se relacionan con sus puntuaciones verdaderas y estimaciones TRI, mostrando las condiciones para ser utilizadas cuando el modelo subyacente es un modelo de unfolding. El objetivo del trabajo actual es comprobar si el patrón de resultados se mantiene con un modelo de dominancia. MÉTODO: las variables independientes del estudio de simulación fueron: formato de elección forzosa, número de bloques, discriminación de los ítems, polaridad de los ítems, variabilidad de la dificultad intrabloque, rango de dificultad del test y correlación entre dimensiones. RESULTADOS: un patrón similar de resultados fue obtenido en ambos modelos, pero en el modelo de dominancia las correlaciones entre PTs y puntuaciones verdaderas son más altas y las diferencias entre PTs y estimaciones TRI se reducen. CONCLUSIONES: un modelo de dominancia produce una relación lineal entre PTs y puntuaciones verdaderas, y los sujetos con puntuaciones extremas son medidos mejor

Traditional scores versus IRT estimates on forced-choice tests based on a dominance model.

BACKGROUND: Forced-choice tests (FCTs) were proposed to minimize response biases associated with Likert format items. It remains unclear whether scores based on traditional methods for scoring FCTs are appropriate for between-subjects comparisons. Recently, Hontangas et al. (2015) explored the extent to which traditional scoring of FCTs relates to the true scores and IRT estimates. The authors found certain conditions under which traditional scores (TS) can be used with FCTs when the underlying IRT model was an unfolding model. In this study, we examine to what extent the results are preserved when the underlying process becomes a dominance model. METHOD: The independent variables analyzed in a simulation study are: forced-choice format, number of blocks, discrimination of items, polarity of items, variability of intra-block difficulty, range of difficulty, and correlation between dimensions. RESULTS: A similar pattern of results was observed for both models; however, correlations between TS and true thetas are higher and the differences between TS and IRT estimates are less discrepant when a dominance model involved. CONCLUSIONS: A dominance model produces a linear relationship between TS and true scores, and the subjects with extreme thetas are better measured.

Comparing Traditional and IRT Scoring of Forced-Choice Tests.

This article explores how traditional scores obtained from different forced-choice (FC) formats relate to their true scores and item response theory (IRT) estimates. Three FC formats are considered from a block of items, and respondents are asked to (a) pick the item that describes them most (PICK), (b) choose the two items that describe them the most and the least (MOLE), or (c) rank all the items in the order of their descriptiveness of the respondents (RANK). The multi-unidimensional pairwise-preference (MUPP) model, which is extended to more than two items per block and different FC formats, is applied to obtain the responses to each item block. Traditional and IRT (i.e., expected a posteriori) scores are computed from each data set and compared. The aim is to clarify the conditions under which simpler traditional scoring procedures for FC formats may be used in place of the more appropriate IRT estimates for the purpose of inter-individual comparisons. Six independent variables are considered: response format, number of items per block, correlation between the dimensions, item discrimination level, and sign-heterogeneity and variability of item difficulty parameters. Results show that the RANK response format outperforms the other formats for both the IRT estimates and traditional scores, although it is only slightly better than the MOLE format. The highest correlations between true and traditional scores are found when the test has a large number of blocks, dimensions assessed are independent, items have high discrimination and highly dispersed location parameters, and the test contains blocks formed by positive and negative items.

Propiedades psicométricas de un test adaptativo informatizado para la medición del ajuste emocional / Psychometric properties of an emotional adjustment computerized adaptive test

En el presente trabajo se describen las propiedades psicométricas de un Test Adaptativo Informatizado para la medición del ajuste emocional de las personas. La revisión de la literatura acerca de la aplicación de los modelos de la teoría de la respuesta a los ítems (TRI) muestra que ésta se ha utilizado más en el trabajo con variables aptitudinales que para la medición de variables de personalidad, sin embargo diversos estudios han mostrado la eficacia de la TRI para la descripción psicométrica de dichas variables. Aun así, pocos trabajos han explorado las características de un Test Adaptativo Informatizado, basado en la TRI, para la medición de una variable de personalidad como es el ajuste emocional. Nuestros resultados muestran la eficiencia del TAI para la evaluación del ajuste emocional, proporcionando una medición válida y precisa, utilizando menor número de elementos de medida en comparación con las escalas de ajuste emocional de instrumentos fuertemente implantados (AU)

In the present work it was described the psychometric properties of an emotional adjustment computerized adaptive test. An examination of Item Response Theory (IRT) research literature indicates that IRT has been mainly used for assessing achievements and ability rather than personality factors. Nevertheless last years have shown several studies which have successfully used IRT to personality assessment instruments. Even so, a few amount of works has inquired the computerized adaptative test features, based on IRT, for the measurement of a personality traits as it’s the emotional adjustment. Our results show the CAT efficiency for the emotional adjustment assessment so this provides a valid and accurate measurement; by using a less number of items in comparison with the emotional adjustment scales from the most strongly established questionnaires (AU)

Análisis factorial confirmatorio de segundo orden y matrices multirrasgo-multimétodo / No disponible

En este trabajo se compara el modelo de análisis factorial confirmatorio de segundo orden, como alternativa a los modelos aditivos tradicionales para analizar la validez convergente y discriminante cuando se utiliza más de un indicador por combinación de rasgo-método. En concreto se comparan el modelo de rasgos correlacionados y métodos correlacionados o modelo completo y el modelo de análisis factorial confirmatorio de segundo orden. Se realizó un estudio de simulación Monte Carlo con estos dos modelos para generar los datos, utilizando una matriz multirrasgo-multimétodo con tres rasgos y tres métodos. Las variables independientes manipuladas son la correlación entre factores de método (0, .2, .4 y .6), el tamaño muestral (200 y 1000) y el modelo de análisis factorial estimado (modelo completo y modelo confirmatorio de segundo orden). En cada condición se utilizaron 100 replicaciones. Las variables dependientes consideradas son la existencia de problemas de estimación (la convergencia del proceso y la aparición de soluciones mal definidas) y el sesgo y la raíz del error cuadrático medio de las estimaciones de las saturaciones de rasgo (validez convergente) y las correlaciones entre rasgos (validez discriminante). Los resultados indican que el modelo completo es superior al modelo confirmatorio de segundo orden. Este modelo presenta muchos problemas de estimación, que lo descartan como alternativa frente a los modelos tradicionales e indican que nunca debe ser utilizado en las condiciones descritas en esta investigación (AU)

Confirmatory factor analysis and MTMM matrices. In this paper, second-order factor analysis is compared with a first-order additive model (complete model) traditionally used to analyse convergent and discriminant validity. A simulation study was performed with the complete model and the second order model as data generators, using multitrait-multimethod matrices with three traits and three methods. Conditions of the study included several indicators per trait-method combination. Manipulated variables included sample size (200 and 1000) and correlation among method factors (0. .2, .4, .6) and the estimated confirmatory model (complete model and second order model). 100 replications were used in each condition. Dependent variables under scrutiny were: a)convergence problems, whether or not the model converged; b) ill-defined solutions; c) bias in the estimation of trait factor loadings (convergent validity) and trait correlations (discriminant validity); and d) the root mean square error in the estimation of the trait factor loadings and trait correlations. Results shown the superiority of the complete model over the second order confirmatory factor analysis. The second order model presented so many estimation problems that almost discard it as an alternative over the first order models in order to analyse multitrait-multimethod matrices included in this study (AU)