Stability of the adjustment scales for children and adolescents

Abstract

Investigation of the short-term (90-day) stability of the Adjustment Scales for Children and Adolescents is reported for 124 randomly selected children in grades ranging from kindergarten to grade 12. Significant test-retest stability coefficients were obtained and mean differences across the retest interval did not exceed .8 raw score points. The Solitary Aggressive-Impulsive, Diffident, and Lethargic/Hypoactive syndromes and the global Underactivity scale showed significant raw score and T score changes across the retest interval, but the effect strengths were small. Syndromic Profile Classifications and Discriminant Classifications were also significantly consistent across the retest interval. Results were similar to those obtained in other stability studies of teacher report behavior rating scales. School psychologists prefer objective assessment methods that facilitate links between assessment and intervention (Reschly & Ysseldyke, 1995), rather than inferential methods, in assessing psychopathology and problem behavior. Consequently, standardized behavior rating scales and checklists have achieved great popularity among school and clinical psychologists (Merrell, 1994a).Among school psychologists, behavior rating scales are the most frequently used instruments in assessing emotional and behavioral difficulties of youths (Stinnett, Havey, & Oehler-Stinnett, 1994). Behavior rating scales are "one of the most efficient, sound, and effective ways . . . to identify a referred student's behavioral strengths and weaknesses . . ." (Knoff, 1995, p. 857) and their use has also been designated a "best practice" in the assessment of emotional and behavioral disorders (McConaughy & Ritter, 1995).In addition to their more objective method, behavior rating scales allow relatively unobtrusive evaluations of student behaviors in natural social settings such as schools, classrooms, and homes. Within the classroom and other school settings, teachers are natural observers and informants since they have the comparative experience of observing many students across time and varied social contexts. As such, they seem to take a normative perspective in rating difficulties in children. Consequently, teachers have often been considered among the most accurate adult raters of child behavior (Kamphaus & Frick, 1996). Behavior rating scales are also cost-effective methods for assessing different behaviors across different environments and raters.Although behavior rating scales have many positive qualities, there are potential threats to their validity such as rater bias (i.e., halo effect, leniency error), rater competency, relevant contact, and rater agreement. Rater agreement may not necessarily be a problem as it is possible that behaviors may vary with respect to different environments and the scale may simply measure those differences. School psychologists are generally aware of classroom differences and the impact on behaviors but need to keep this issue in mind in the assessment process. Many behavior rating scales have also bee...

Interrater agreement for discriminant classifications for the adjustment scales for children and adolescents

“…Diagnostic agreement of ASCA discriminant classifications also was consistent with kappas (.54 to .59) reported for the DSM-IV field trials for disruptive behavior disorders (Lahey et al, 1994). The present level of agreement for discriminant classifications was higher than that obtained in a 90-day test-retest stability study of the ASCA ( ϭ .35; Canivez et al, 2001).…”

“…Test-retest reliabilities over a 30-schoolday interval ranged from .66 to .91 for the six core syndromes (N ϭ 40), and equaled .75 for the Overactivity scale and .79 for the Underactivity scale. Canivez et al (2001) replicated the stability of the ASCA T scores over a 90-day interval as well as supporting stability of the ASCA syndrome profiles and discriminant classifications. Watkins and Canivez (1997) replicated the interrater agreement for ASCA T scores while Canivez and Watkins (2001) found significant interrater agreement for ASCA syndrome profiles.…”

Situational specificity makes the difference in assessment of youth behavior disorders

“…Research has provided substantial support for the structural integrity (McDermott, 1993), reliability (i.e., internal consistency, stability and interrater agreement : Canivez, Perry, & Weller, 2001;Schaefer, Watkins, & Canivez, 2001;Watkins & Canivez, 1997), convergent validity (Canivez & Rains, 2002) and validity generalization (Canivez, & Bordenkircher, 2002;McDermott, 1994), including structural invariance and reliability generalization to preadolescents, adolescents, boys, girls, and children from ethnic minority populations 1993;McDermott, Watkins, Sichel, Weber, Keenan, Holland, & Leigh, 1995).…”

Adjustment scales for children and adolescents and Native American Indians: Factorial validity generalization for Ojibwe youths

“…Extensive evidence for ASCA score reliability and validity is presented in the ASCA manual (McDermott, 1994) and in independent studies. Internal-consistency estimates (Canivez, 2004;McDermott, 1993McDermott, , 1994, stability estimates (Canivez, Perry, & Weller, 2001;McDermott, 1993McDermott, , 1994, and interrater-agreement estimates Canivez, Watkins, & Schaefer, 2002;McDermott, 1993McDermott, , 1994Watkins & Canivez, 1997) have supported the reliability of ASCA scores.…”