Joint Factor Analysis of the Children's Depression Inventory and the Revised Children's Manifest Anxiety Scale

Abstract

Used a joint factor analysis with the Children's Depression Inventory (CDI; Kovacs, 1980/81, 1992) and Revised Children's Manifest Anxiety Scale (RCMAS; C. R. Reynolds & Richmond, 1978, 1985) to identify items that uniquely measured depression and anxiety. Data from 750 youngsters in Grades 4 through 7 were analyzed using principal-axis factoring with an oblique rotation. Salient factors were identified using guidelines provide by Gorsuch (1997). Item overlap and the large negative affectivity component across instruments were evident. Items that overlapped or had nonsalient loadings were eliminated. The sample was randomly split into 2 groups of 375 and analyses were repeated. Results indicated that a unique 9-item depression factor composed largely of items representing a negative view of oneself existed. In addition, a unique 7-item anxiety factor emerged that consisted of items reflecting worry. The validity of these abbreviated scales was explored using a separate sample of 131 students in Grades 4 through 9. The abbreviated scales were correlated with scales of positive and negative affect consistent with predictions. Findings suggest exploring alternative scoring strategies for the CDI and RCMAS to eliminate problems associated with overlapping items.

Síntomas de depresión, ansiedad y estrés post-traumático en adolescentes siete meses después del terremoto del 27 de febrero de 2010 en Chile

“…Para este estudio se utilizaron las siguientes escalas: a) El Inventario de Depresión de Niños (Kovacs, 1981;Stark & Laurent, 2001). Es una escala de auto-reporte de 27 ítems con tres alternativas de respuesta (0 = "no me ocurre", 1 = "me ocurre poco" y 2 = "me ocurre mucho"), destinada a medir la presencia de síntomas de depresión en niños y adolescentes, a quienes se les pide escoger una de las alternativas en cada ítem en términos de las "últimas dos semanas".…”

“…Según el manual del inventario (Kovacs, 1992), en base a una población estandarizada (μ = 10.5; σ = 7.3), el puntaje de corte para establecer quién tiene una "Alta" o "Baja" presencia de síntomas se estableció como una desviación estándar por sobre la media. Por lo tanto, el puntaje de corte utilizado en la presente investigación fue de 18. b) La Escala Revisada de Ansiedad Manifiesta para Niños (Reynolds & Richmond, 1978;Stark & Laurent, 2001). Es una escala de auto-reporte de 37 ítems de respuesta dicotómica (SÍ o NO) destinada a medir la presencia de síntomas de ansiedad en niños y adolescentes, a quienes se les pide determinar si cada una de las afirmaciones lo describe o no.…”

An Evaluation of the Applicability of the Tripartite Constructs to Social Anxiety in Adolescents

“…Several investigations have supported two-factor models of PA and NA as distinct constructs in school children (Cole et al, 1997;Crook, Beaver, & Bell, 1998;Stark & Laurent, 2001) and adolescents (Huebner & Dew, 1995;Inderbitzen & Hope, 1995;Lonigan, Hooe, David, & Kistner, 1999;Valentiner, Gutierrez, & Blacker, 2002). More recent research has supported the utility of all three constructs in children and adolescents in clinical (Joiner, Catanzaro, & Laurent, 1996;Joiner & Lonigan, 2000;Lonigan, Carey, & Finch, 1994) and nonclinical (Cannon & Weems, 2006;Olino, Klein, Lewinsohn, Rohde, & Seeley, 2007;Phillips, Lonigan, Driscoll, & Hooe, 2002) settings.…”

“…The shared variance between these two symptom clusters has been estimated to be as high as 72% (Cole, Truglio, & Peeke, 1997) and controlling for item overlap between measures has resulted in only a minimal decrease in shared variance (Cole et al, 1997;Stark & Laurent, 2001). In a review of the literature, Brady and Kendall (1992) found a 16% comorbidity rate for anxiety and depressive disorders in community, non-treatment-seeking youth samples and rates ranging from 28% to 62% in clinical samples.…”

Treatment Moderation and Secondary Outcomes: Results from a Randomized Clinical Trial

“…Nevertheless, psychopathology researchers have paid far less attention to the development of strong measures of emotional distress (i.e., anxiety and depression) and externalizing problem behaviors than the development of measures of psychopathology with nonoverlapping items (e.g., Carter, Silverman, Allen, & Ham, 2008;Stark & Laurent, 2001). Research using measures of psychopathology with nonoverlapping items is needed to understand the relations between anxiety, depressive, and externalizing problems in ways that further informs classification efforts.…”