Publicación: Índices de complejidad cognitiva desde la Técnica de la Rejilla: exploración de su validez convergente y de sus relaciones con variables clínicas y de personalidad
Cargando...
Fecha
2014-10-14
Autores
Editor/a
Director/a
Tutor/a
Coordinador/a
Prologuista
Revisor/a
Ilustrador/a
Derechos de acceso
info:eu-repo/semantics/openAccess
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Nacional de Educación a Distancia (España). Facultad de Psicología. Departamento de Psicología Social y de las Organizaciones
Resumen
Landfield (1977) propuso que la complejidad cognitiva (CC) estaría compuesta de dos procesos independientes: diferenciación cognitiva e integración cognitiva. Numerosos investigadores han desarrollado diferentes métodos basados en la técnica de la rejilla (Kelly, 1955) para medir estos procesos. El objetivo de este trabajo es explorar dos aspectos concretos de la CC. Primero, examinar la validez convergente de los principales índices de diferenciación, en concreto PVEPF, Intensidad, Bieri, FIC y SQMdif, y de los principales índices de integración, a saber, Ordenación y SQMint. En segundo lugar, este trabajo explora las relaciones de cada uno de estos índices con variables demográficas y psicológicas, concretamente edad, nivel de estudios, género, número de dilemas implicativos en la rejilla (Feixas y Saúl, 2004), problemas clínicos, personalidad, apego y ansiedad. 1- Validez convergente Se emplea metodología correlacional y factorial para analizar una muestra de 100 individuos no clínicos de nacionalidad española, los cuales completaron un protocolo semiestructurado de rejilla administrado por estudiantes de Grado. El método de extracción fue el de componentes principales con rotación Varimax. En aras al carácter exploratorio del estudio se eligió la solución que explicase al menos el 90% de la varianza. El resultado del análisis factorial muestra a los índices PVEPF e intensidad cargando en el primer factor y a los demás índices cargando en los 4 factores restantes (Tabla 4, página 67). El primer factor explica el 46.5% de la varianza, lo que sugiere que el PVEPF y la intensidad podrían considerarse las medidas de diferenciación disponibles de mayor solidez. Los dos índices de integración incluidos en este trabajo correlacionan significativamente con la mayoría de las medidas de diferenciación (Tabla 5, página 67), lo que pone en duda su validez como medidas de integración (supuestamente independientes de la diferenciación). Se argumenta que el principal escollo en el modelo de Landfield sigue siendo la ausencia de una medida de integración cognitiva válida. 2- Relaciones con otras variables psicológicas Se utilizaron los siguientes instrumentos: CORE-OM (Feixas et al, 2012) para los problemas clínicos, NEO-FFI (Costa y McRae, 1999) para la personalidad, CAA (Melero y Cantero, 2008) para el apego, e ISRA (Miguel y Cano, 1986) para la ansiedad. La metodología fue correlacional, empleando la muestra anterior. Los resultados arrojan numerosas correlaciones significativas (débiles-moderadas), que parecen apuntar en una misma dirección: una mayor diferenciación está consistentemente asociada a mayores dificultades (Tablas 9, 10, 11, 12, 15, páginas 69-72). Por ejemplo, mayores PVEPFs están significativamente relacionados con mayor número de dilemas implicativos (-.46), más dificultades clínicas (-.207), menos extroversión (.24), más resolución hostil de conflictos (-.277), menor expresión de sentimientos (-.257), mas autosuficiencia emocional (-.226), mas ansiedad en la vida cotidiana (-.226), y más ansiedad interpersonal (-.235). Es importante señalar que esta consistencia en la dirección de los resultados se ha obtenido utilizando 5 pruebas independientes. Las variables edad, nivel de estudios y género no arrojaron ningún resultado estadísticamente significativo (Tablas 6, 7, 8, página 68). Es plausible que la debilidad de estas correlaciones esté influenciada por los individuos altos en integración. Según el modelo de Landfield, las personas con puntuaciones altas tanto en diferenciación como en integración tendrían en general una mejor salud mental. Debido a que estos individuos no pueden ser identificados y, por tanto, asignados a un grupo diferente, sus resultados más positivos pueden estar arrastrando las correlaciones a la baja. Se sugieren futuras direcciones de investigación en línea con las señaladas por Neimeyer, Anderson y Stockton (2001) en relación al desarrollo de una medida válida de la integración cognitiva.
Landfield (1977) theorised that the psychological construct of cognitive complexity (CC) was made of two independent processes: cognitive differentiation and cognitive integration. Over the years, numerous repertory grid based methods (Kelly, 1955) for measuring these processes have been proposed. This study attempts to explore two aspects of CC. Firstly, to examine the convergent validity of the main differentiation indices, namely PVAFF, Intensity, Bieri, FIC and SQMdif, and the main integration indices, namely Ordination and SQMint; and secondly, to explore the possible relations of each of these measures with demographic and psychological variables, namely age, level of studies, gender, number of implicative dilemmas in the grid (Feixas & Saul, 2004), clinical problems, personality, attachment, and anxiety. 1- Convergent validity The convergent validity of the CC indices above is analysed through correlation and factor analysis using a sample of 100 non-clinical Spanish individuals, who had completed a student-administered semi-structured grid protocol. The extraction method was Principal Components Analysis with Varimax rotation. Due to the exploratory nature of the study, the amount of variance explained by the solution was set to a minimum of 90%. Loadings below .3 were omitted. Results of the factor analysis show PVAFF and Intensity loading in the first factor, while the rest of the indices load on the remaining 4 factors (Table 4, page 67). PVAFF and Intensity account for 46.5% of the variance, which might suggest they are the most solid measures of cognitive differentiation available today. The two integration indices included in this study were found to correlate significantly with most of the differentiation measures (Table 5, page 67), hence bringing their validity as measures of integration (supposedly independent from differentiation) into question. It is argued that the main gap in Landfield´s model seems to be the absence, still today, of a valid integration measure. 2- Relationships to other psychological variables The following measures were used: CORE-OM (Feixas et al, 2012) for clinical problems, NEO-FFI (Costa & McRae, 1999) for personality, CAA (Melero & Cantero, 2008) for attachment, andISRA (Miguel & Cano, 1986) for anxiety. The methodology was correlational, using the sample above. The results show numerous weak to moderate but significant correlations, which seem to point in the same direction: increased cognitive differentiation is consistently associated with more difficulties (Tables 9, 10, 11, 12, 15, pages 69-72). For example, higher PVAFFs are significantly related to more implicative dilemmas (-.46), more clinical difficulties (-.207), less extroversion (.24), more hostile resolution of conflicts (-.277), less expression of feelings (.257), more emotional self-sufficiency (-.226), more daily life anxiety (-.226), and more anxiety in interpersonal relationships (-.235). It is worth highlighting that this consistency in the direction of results has been obtained using 5 independent measures. Age, level of studies and gender do not show any statistically significant results (Tables 6, 7, 8, page 68). It is a theoretical possibility that correlations might have been affected by individuals high in integration. According to Landfield’s model, people high in both differentiation and integration would have better mental health overall. Since these cannot be identified and, thus, allocated to a different group, their more positive scores may be decreasing the strength of the correlations. Future directions of research are suggested in line with those made by Neimeyer, Anderson & Stockton (2001) in regard to the development of a valid integration measure.
Landfield (1977) theorised that the psychological construct of cognitive complexity (CC) was made of two independent processes: cognitive differentiation and cognitive integration. Over the years, numerous repertory grid based methods (Kelly, 1955) for measuring these processes have been proposed. This study attempts to explore two aspects of CC. Firstly, to examine the convergent validity of the main differentiation indices, namely PVAFF, Intensity, Bieri, FIC and SQMdif, and the main integration indices, namely Ordination and SQMint; and secondly, to explore the possible relations of each of these measures with demographic and psychological variables, namely age, level of studies, gender, number of implicative dilemmas in the grid (Feixas & Saul, 2004), clinical problems, personality, attachment, and anxiety. 1- Convergent validity The convergent validity of the CC indices above is analysed through correlation and factor analysis using a sample of 100 non-clinical Spanish individuals, who had completed a student-administered semi-structured grid protocol. The extraction method was Principal Components Analysis with Varimax rotation. Due to the exploratory nature of the study, the amount of variance explained by the solution was set to a minimum of 90%. Loadings below .3 were omitted. Results of the factor analysis show PVAFF and Intensity loading in the first factor, while the rest of the indices load on the remaining 4 factors (Table 4, page 67). PVAFF and Intensity account for 46.5% of the variance, which might suggest they are the most solid measures of cognitive differentiation available today. The two integration indices included in this study were found to correlate significantly with most of the differentiation measures (Table 5, page 67), hence bringing their validity as measures of integration (supposedly independent from differentiation) into question. It is argued that the main gap in Landfield´s model seems to be the absence, still today, of a valid integration measure. 2- Relationships to other psychological variables The following measures were used: CORE-OM (Feixas et al, 2012) for clinical problems, NEO-FFI (Costa & McRae, 1999) for personality, CAA (Melero & Cantero, 2008) for attachment, andISRA (Miguel & Cano, 1986) for anxiety. The methodology was correlational, using the sample above. The results show numerous weak to moderate but significant correlations, which seem to point in the same direction: increased cognitive differentiation is consistently associated with more difficulties (Tables 9, 10, 11, 12, 15, pages 69-72). For example, higher PVAFFs are significantly related to more implicative dilemmas (-.46), more clinical difficulties (-.207), less extroversion (.24), more hostile resolution of conflicts (-.277), less expression of feelings (.257), more emotional self-sufficiency (-.226), more daily life anxiety (-.226), and more anxiety in interpersonal relationships (-.235). It is worth highlighting that this consistency in the direction of results has been obtained using 5 independent measures. Age, level of studies and gender do not show any statistically significant results (Tables 6, 7, 8, page 68). It is a theoretical possibility that correlations might have been affected by individuals high in integration. According to Landfield’s model, people high in both differentiation and integration would have better mental health overall. Since these cannot be identified and, thus, allocated to a different group, their more positive scores may be decreasing the strength of the correlations. Future directions of research are suggested in line with those made by Neimeyer, Anderson & Stockton (2001) in regard to the development of a valid integration measure.
Descripción
Categorías UNESCO
Palabras clave
Citación
Centro
Facultades y escuelas::Facultad de Psicología
Departamento
Psicología Social y de las Organizaciones