Análisis diferencial de los indicadores psicométricos de la escala multidimensional de trastornos afectivos (EMTA) desde los modelos de la teoría clásica de los test y la teoría de respuesta al ítem
Fonseca Consuegra, Liz
Show full item record
Show full item record
AbstractThis graduation project was developed under Ps. Omar Cortes, the guidances, to support the research work of the Ps .Daniella Abello (2011), aimed to analyze the levels of reliability, validity and psychometric indicators associated Multidimensional Scale of Affective Disorders (EMTA) from the classical theory models of the Test and Item Response Theory in a sample of 384 students from three Higher Education Institutions in Barranquilla, women and men, aged between 17 and 26 years. This study is framed in the empirical-analytical paradigm, the theoretical approach is analytical-synthetic and is considered as an instrumental study, due to its purpose. The data were collected through the application of the following instruments : a) The Multidimensional Scale Affective Disorder ( EMTA ) b ) The Beck Depression Inventory ( BDI ) (α: 0.791 ), and c ) The self-report scale of Altman (Altman Self -Rating Mania Scale) adapted to the Spanish sphere (α: 0,566 ) . After analyzing EMTA Scale, satisfactory results were reported from both models. From TCT results, the reliability indices of Alfa Cronbach were reported high for both subscales ( α : , 942L for Depression subscale and α : , 864 for the Mania subscale ) , high internal consistency for all items , an strong validity construct and criteria validity significantly higher for the Depression subscale , but a slightly lower than expected for the Mania subscale : however, it is considered that the weakness of this aspect underlies the reliability of the test report as used as reference for the Mania subscale , future research is suggested to use a more reliable standard . Also, from the TRI, was reported high levels of reliability from both subscales, consistent with the results obtained from the TCT: in the case of depression to 0.88 for individuals and 0.99 for items , while Mania rates were 0.84 for individuals and 0.87 for items . Throughout the report the results enjoyed an increasing monotonic fit for both items to the two subscales, which confirms that the principle of invariance in latent trait measurement although the report item performance and people, tended to be more uniform for the Mania subscale as the Depression subscale. The results are in accordance to theoretical parameters significantly in both models, responding well to guidelines. However, the contrast with other studies was not possible because there were no investigations homologous to the analysis, since the market do not have scales to measure components simultaneously Mood Disorders and Depression test / Manía most recognized (bobes, J., Hatch, M., Bascarán, M., Saiz, P., Bousoño, M., 2004, Sanchez, R., Velasquez, J., Navarro, A., 2011) have been standardized from one TCT model, probably because these studies date from a time before the date of the rise of the TRI.