The differential validity of teacher ratings of Inattention/Overactivity and Aggression

The ability of school-based measures of child behavior to predict differentially the IOWA Conners teacher rating scale Inattention/Overactivity (IO) and Aggression (A) factors was evaluated in a sample of 71 school-aged boys. Multiple measures and multiple methods of assessment were utilized to provide a comprehensive assessment of social and academic behavior. These included direct observations of children in the classroom and on the playground, examination of the organization of children's desks, measures of academic performance, peer nominations of popularity and rejection, and sociometric ratings using the Pupil Evaluation Inventory. Despite moderately high correlations between the teacher rating factors (r = .60), considerable evidence was provided for differential validity on measures of academic performance, peer rating measures, and measures of disruptive or inappropriate classroom and playground behavior. These differences between IO and A factors indicated that the factors were logically independent in important ways, supporting prior work validating separate hyperactivity and aggression dimensions.     

A multistudy analysis of socioeconomic status (SES) and the measures and outcomes of cognitive-behavioral treatment with children

The relationship of the child's socioeconomic status (SES) to improvement and generalization from cognitive behavioral-treatment was investigated by comparing high and low SES groups on teacher ratings, performance measures, and judges' ratings of improvement and maintenance of positive change. Fifty-eight teacher referred classroom problem children (grades 2–6) who had served as treated subjects in one of three treatment outcome studies were examined. SES data were collected by administration of a questionnaire to the head of each household. The judges' ratings were performed on clusters of teacher ratings and clusters of performance measures that were present in each study. Results indicated that, despite SES group variation in level of performance, improvement, when it occurred, was evident for both high and low SES groups. In addition, there were significant differences on a number of the performance measures at pretreatment, posttreatment, and follow-up. Initial SES group differences in verbal ability as measured by the Peabody Picture Vocabulary Test (PPVT) appear to have been an important factor in these performance differences.This research was supported by grant number 441 0749 5236 02 from the Graduate School of the University of Minnesota, awarded to Philip C. Kendall. Portions of this research were presented at the Association for the Advancement of Behavior Therapy Convention, San Francisco, December 1979. Portions of the material were prepared while the second author was at the Center for Advanced Study in the Behavioral Sciences and supported by NIMH (No. 5-T32-MH 14581-05) and the MacArthur Foundation. The authors wish to thank Dorothy Raney for her valuable assistance.     

Validation of a role-play measure of children's social skills

A videotape-administered role-play test of children's social skills was developed and its psychometric properties tested. Performance criteria for the test were derived from popular children's ratings of the effectiveness of different role-play responses. The test was administered to 157 fourth- and fifth-grade boys and girls who had been classified as popular, average, neglected, or rejected, on the basis of sociometric testing. The test evidenced good interrater, test-retest, and internal consistency reliabilities. Children's role-play performance correlated significantly with teacher ratings of social competence and with peer-liking ratings. Neglected children performed more poorly on the role-play test than popular children. When differences in intelligence among social status groups were statistically controlled, social status groups did not differ on the role-play test. Results of the discriminant analyses support the conclusion that teacher ratings are better than role-play tests for identifying rejected children, whereas role-play tests and measures of intelligence appear more accurate than teacher ratings for identifying neglected children.     

Utility of teacher behavior ratings for identifying special needs students

The purpose of this study was to explore the efficacy of a teacher behavior rating instrument for identifying special needs students. Using a modified form of theDevereux Elementary School Behavior Rating (DESB)Scale 35 kindergarten through grade 6 regular classroom teachers completed ratings on all of their 876 students. Subsequently, extensive aptitude, academic, social, and behavioral assessment was conducted and those students were identified who were in need of supportive educational programming to function adequately within the regular class setting. Analysis of the teacher behavior ratings indicated a highly significant difference between those students identified for special supportive services and their regular classroom peers on 9 of 11 behavioral factors. The findings lend clear support for the use of classroom teachers' behavior ratings in the identification process.     

The validity of vigilance tasks in differential diagnosis of children referred for attention and learning problems

This study was conducted to evaluate the convergent and discriminant validity of vigilance measures of attention and impulsivity in children. One hundred children referred for evaluation of attention and learning problems were administered a battery of tests including two vigilance tasks, other laboratory measures of inattention and impulsivity, and parent and teacher ratings. It was predicted that vigilance task performance would correlate with teacher ratings and laboratory measures of inattention while correlations with ratings of aggressive behavior were predicted to be lower. Moderate correlations were obtained between the vigilance task scores and other laboratory measures of inattention and impulsivity. However, the vigilance scores did not correlate with teacher or parent ratings of behavior. A factor analysis, conducted to identify underlying dimensions of attention in the battery of tests, revealed strong source effects. The implications for use of vigilance measures in diagnosis are discussed.Portions of this paper were presented at the 1990 Learning Disabilities Association of America International Conference, Anaheim, California.     

The effects of background illumination and stimulant medication on smooth pursuit eye movements of hyperactive children

Problems with visual information processing have been reported in children with the diagnosis of attention deficit disorder with hyperactivity (ADD H), and deficits in oculomotor control have been posited as an important factor in this phenomenon. To assess aspects of oculomotor performance, smooth pursuit eye movements (PEMs) were recorded electrooculographically in 20 ADD H and 20 age-matched control children and computer-analyzed for discrete (velocity arrests) and global (root mean square error) disruptions. The effects of stimulant medication (methylphenidate), together with manipulations designed to influence behavioral (attention) and physiological (cerebellar) processes involved in PEM performance, were examined. The tracking patterns of nonmedicated ADD H children contained significantly more discrete aberrations on baseline conditions. Although no single experimental manipulation significantly improved tracking performance in ADD H children, combining all experimental conditions did normalize PEMs in these subjects. Slight improvements in PEM performance in association with medication correlated positively with does of madication of with medication-related improvments of behavior. In light of these data, arguments are presented in supported of (a) hyporousal as a contributing factor underlying oculomotor difficulties in ADD H children and (b) subcortical involbment in PEM dysfunction.This research was funded in part by grants from The Hospital for Sick Children Foundation and the Ontario Mental Health Foundation (RTP). The authors thank Susan Anthony, Dr. Pamela M. Cooper, Stella Cowley, Dr. Kathy Margittai, and Ralph Nevins for technical assistance, Martin Gillett for computer programming, Dr. Phil Firestone and the staff of the University of Ottawa Child Study Centre, and Dr. B. Lena of the Family and Child Unit, Ottawa General Hospital, for subject referral.     

Hyperactive boys in their classroom: Assessment of teacher and peer perceptions,interactions, and classroom behaviors

Teacher ratings, peer perceptions, peer interactions, and classroom behaviors of 17 hyperactive and 17 active elementary school-age boys, nominated by their teachers, were compared using multivariate analyses and planned comparisons in order to better describe and assess hyperactivity in its most probable setting —the classroom. Hyperactive boys were found to be significantly different from actives on measures from all data sources in that they were perceived and interacted more negatively. Cluster analyses of teacher ratings of 90 hyperactives from a clinical sample and 17 hyperactives from the current sample were used to discriminate among different types of hyperactives. Four types were named anxious, conduct problem, inattentive, and low problem hyperactives. The fact that six conduct problem hyperactives were found to be more disruptive and have higher activity level ratings than six inattentive hyperactives, when observed in their classrooms, points to the need to study and treat hyperactives as heterogeneous groups.This study is based in part on a doctoral dissertation submitted to the Psychology Department, Indiana University, by the first author. Thanks are expressed to the students and teachers of the Monroe County School System who took part in the study. The authors also wish to thank Prof. Robert Sprague of the University of Illinois for providing the Conners Rating Scales on 90 hyperactive children.     

Correspondence of teacher ratings and direct observations of classroom behavior of psychiatric inpatient children

The present investigation examined the correspondence of teacher ratings and direct observations of classroom behavior. Techers, extraclass raters, and observers completed standard rating scales and/or measures of overt classroom behaviors of psychiatric inpatient children (N=32). The study assessed if the correspondence between ratings and direct observations was influenced by who evaluates the child (teachers, raters) and the assessment format (general ratings, discrete behaviors). The results indicated that (1) measures from different assessors correlated in the low to moderate range, (2) data from extraclass raters corresponded more closely with direct observations than with data from teachers, (3) teacher and rater estimates of overt child behavior did not correlate more highly with direct observations than did standard rating scales, and (4) teachers and raters viewed child behavior as more appropriate than direct observations indicated. Measures from teachers, raters, and observers readily distinguished attention deficit disorder children with hyperactivity from their peers. However, teacher evaluations delineated these children more sharply than other assessors.Completion of this paper was facilitated by a Research Scientist Development Award (MH00353) to the first author from the National Institute of Mental Health.     

Comparing Four Methods of Integrating Parent and Teacher Symptom Ratings of Attention-deficit/hyperactivity Disorder (ADHD)

Although parents and teachers are valid informants in the assessment of childhood attention-deficit/hyperactivity disorder (ADHD), there is relatively little systematic research on how these ratings should be optimally combined. We compared four methods of ADHD assessment to determine how well they identified impaired children: (1) parent only, (2) teacher only, (3) parent or teacher (‘or rule’), and (4) parent and teacher (‘and rule’). We obtained parent and teacher ratings of ADHD from the Disruptive Behavior Disorder Rating Scale on 232 5- to 10-year-old children (69% male; 47% Caucasian) with (n = 121) and without (n = 111) ADHD. We used receiver operating characteristic curves (ROC) and seemingly unrelated regression analyses (SUR) to evaluate how accurately each method identified categorically- and dimensionally-defined measures of functional impairment. Parent ratings of ADHD optimally identified globally impaired children based on categorical and dimensional measures. However, teacher ratings of ADHD most accurately identified children who were negatively regarded by peers using categorical, but not dimensional, measures. No ADHD assessment method effectively identified children with academic difficulties. Although multiple informants are valuable in the assessment of ADHD, no single method was consistently superior in identifying impaired children across domains. We consider alternative assessment strategies in ADHD as well as other potential factors that may contribute to modest agreement among informants.     

Continuities in maternal reports and child behaviors over time in hyperactive and comparison groups

Maternal reports, observations of nursery and elementary school behavior, and teacher ratings of problems were available for hyperactive and control children who had participated in a longitudinal study. This paper examines consistencies in maternal reports and child behaviors over time, and their relationship to teacher ratings in elementary school. Maternal reports of infant sleep difficulties were related to maternal ratings of hyperactivity at 41/2 and 61/2 years. Maternal ratings of activity at 41/2 were predictive of 61/2-year ratings of hyperactivity and conduct problems. In addition, behavior in a research nursery at 41/2 predicted teacher ratings of problems and classroom behavior in grade two. Hyperactive preschoolers who left the table most during structured activities were more often out-of-seat and off-task in school. Controls who were more aggressive in the nursery were more disruptive in the classroom. These data indicate continuities in both maternal reports and child behaviors.Collection of follow-up data were supported by Grant No. MA-4505 from the Medical Research Council of Canada to Dr. Campbell and were collected when she was at McGill University and the Montreal Children's Hospital. Portions of these data were presented at the American Psychological Association, Washington, D.C., September 1976.     

The interaction of cognitive styles and classroom environment in determining first-graders' behavior

This study investigates the importance of two “person” variables, reflection impulsivity and divergent thinking, and one “situation” variable, the classroom climate of first-grade classrooms, for the classroom adaptation of first-graders. Three types of classrooms differing in classroom climate were selected: open, intermediate, and traditional. Subjects were the children (n = 189) in these classrooms. Classroom adaptation variables included self concept, school attendance, and teacher ratings of classroom behavior, overall academic rank, and personal preference. Both main effect and interactional relationships between “person” and “situation” variables and classroom adaptation were hypothesized. Classroom environment was found to affect degree of classroom disturbance manifested by the children with students in open classrooms exhibiting most and students in the intermediate classrooms exhibiting least. Boys in the open classrooms had higher achievement anxiety and lower teacher preference ratings than boys in traditional or intermediate classrooms. Reflective children obtained higher ratings as self-reliant learners and, for boys, reflection was related to low classroom disturbance and high achievement anxiety. Interactional relationships between classroom environment and reflection-impulsivity were also found for boys. Divergent thinking was not related to classroom adjustment for boys. Girls high in divergent thinking were higher in school involvement and achievement anxiety and more preferred by teachers.     

Multimethod Psychoeducational Intervention for Preschool Children with Disruptive Behavior: Two-Year Post-Treatment Follow-up

This paper describes the 2-year post-treatment follow-up of preschool children identified as having high levels of disruptive behavior at kindergarten entry. They were assigned to four treatment conditions: A no-treatment group, parent-training only, treatment classroom only, and the combination of parent training with the treatment classroom. Interventions lasted the entire kindergarten academic year. Initial post-treatment results reported previously indicated no effects for the parent-training program but some efficacy for the classroom intervention program. For this report, the disruptive behavior (DB) children were subdivided into those who did (n = 74) and did not (n = 77) receive the treatment classroom. Two-year post-treatment follow-up results indicated no differences between the classroom treated and untreated DB groups. These groups also failed to differ in the percentage of children using available treatments across the follow-up period. The DB children in both groups had significantly more symptoms of ADHD and ODD than a community control group (N = 47) at follow-up. They also received higher ratings of externalizing problems on the parent Child Behavior Checklist, more severe ratings of behavior problems at home, and ratings of more pervasive behavior problems at school, and had poorer academic skills. Results suggested that early intervention classrooms for DB children may not produce enduring effects once treatment is withdrawn, and that better approaches are needed for identifying those DB children at greatest risk for later maladjustment.     

Interrelationships between temperament characteristics and first-grade teacher-student interactions

The relationships between temperament characteristics and first-grade teacherstudent interaction were investigated. It was hypothesized that the temperament characteristics of Adaptability and Approach/Withdrawal would be potent predictors of teacher-student interactions since they had been found previously to be strong predictors of children's emotional and behavioral adjustment to school. The teachers of 105 first-grade children completed temperament questionnaires during the early part of the academic year, and their interactions with the children were observed soon thereafter. The teacher temperament ratings were factor-analyzed, and regression models were generated to predict classroom interactions. The study's hypothesis was supported, and three separate temperament dimensions were found to be predictive of classroom interactions. Suggestions are made for future extensions of the findings.     

Agreement of parent and teacher rating scales with comprehensive clinical assessments of attention deficit disorder with hyperactivity

Forty-one clinic-referred children were given the DSM-III diagnosis of attention deficit disorder with hyperactivity (ADD/H) based on structured diagnostic interviews of the child and mother and on two teacher and two parent rating scales. The degree of agreement between diagnostic classification of ADD/H based solely on criterion scores on each rating scale and classification based on the comprehensive clinical assessments that included use of the same rating scaleswas assessed by comparing the ratings of the 41 children with ADD/H to those of 47 clinic-referred children given other diagnoses or no diagnosis. The accuracy of classification was highly similar across scales. Increasing the cutoff score increased the acing true positives. Using the point of intersection of these two curves to choose the optimal cutoff scores resulted in 70–75% accurate classification and 25–30% misclassification of children given clinical diagnoses of ADD/H. This rate of misclassification suggests that rating scales alone should not be used to classify ADD/H for most clinical and research purposes, especially because the method of comparison used in this study favored finding a high degree of correspondence.     

Social perspective-taking and teacher ratings of peer interaction in hyperactive boys

Hyperactive and matched control boys at two age levels were compared on teacher ratings of peer interaction and three measures of social perspective-taking. The teachers' ratings of peer interaction discriminated between the hyperactive children and their controls. In addition, a significant interaction effect indicated that teachers rated the older hyperactive group as more deviant with peers than both their younger hyperactive counterparts and their age-matched controls. In a structured laboratory setting, the hyperactive children did not differ from control children on the three measures of social perspective-taking. The perspective-taking measures did, however, show differences across age groups. Implications for treatment are examined.The research reported in this paper is based on master's thesis submitted to the Graduate Faculty of Arts and Sciences of the University of Pittsburgh by the first author. The authors are indebted to Dr. C. Keith Conners for his assistance in recruiting hyperactive subjects and to Drs. Alex Siegel and Carl Barenboim for their critical evaluations regarding this investigation. Appreciation is expressed to the staff, teachers, parents, and students of the Forest Grove Elementary School for their cooperation.     

Parent reports of child behavior problems: Bias in participation

This paper compares the characteristics of 274 children whose parents agreed to provide information about their children's functioning in research with children whose parents did not. Parents had provided permission for child participation in school-approved research. Measures included self-, peer, teacher, and parent ratings. Parents of minority children were less likely to participate than parents of nonminority children. Among nonminorities, children of parent participants were viewed as more socially skilled and liked by their peers; teachers rated them as having less attention problems, less depression, and better academic skills than children of nonparticipating parents. Among minorities, no differences emerged. Implications for research involving the use of parent ratings are discussed.Portions of this paper were presented at the annual meeting of the Society for Research in Child and Adolescent Psychopathology, Sarasota, Florida, February 1992. The authors appreciate the efforts of Shari Wasserstein in assisting with the organization of the data. The authors also appreciate the helpful comments of two anonymous reviewers.     

Psychostimulant effects on academic and behavioral measures for ADHD junior high school students in a lecture format classroom

This study evaluated the effects of two doses of psychostimulants on nine ADHD students' performance in a classroom designed to be analogous to a junior high school American history class conducted during a summer treatment program. Double-bline, placebo-controlled assessment revealed significant and clinically important effects of psychostimulants on quiz and test performance, observations of attention and behavior during lectures, and teacher ratings, as well as accuracy on assignments completed during study hall. Analysis of individual differences revealed a 66% positive response rate. The implications for use of psychostimulants with ADHD young adolescents are discussed.This research was conducted while the second author was supported in part by grants from the National Institute on Alcohol Abuse and Alcoholism (AA06267) and the National Institute on Drug Abuse (DA 05605). This study was conducted during the 1989 Summer Treatment Program (STP) conducted by the Attention Deficit Disorder Program at the Western Psychiatric Institute and Clinic. We thank WPIC for its support, as well as the ADD and STP program staffs for their helpful cooperation, particularly Kelly Davis, Karen Greenslade, Vincent Grudberg, Gray Vallano, and Mary Vodde-Hamilton.     

The utility of tracking child behavior

To assess the impact of observation training and tracking of child behavior, each of 20 elementary school teachers participating in a behavior modification practicum was assigned to either an experimental or a control group. For each teacher, a disruptive, distractible child was identified as appropriate for behavioral intervention. All teachers were taught identical intervention strategies, but the two groups differed in the degree to which teachers were initially trained and subsequently required to observe and record discrete child behaviors. Multiple measures were used to assess the influence of monitoring child behaviors. These included observed child behavior, observed teacher behavior, teacher ratings, and correspondence between teacher perception and child behavior. Results showed that tracking the behavior of children had little or no effect on any measure.Support for this study was provided by the Molson Foundation and the Grant Foundation. Computing assistance was furnished by Health Sciences Computing Facility, UCLA. The author wishes to thank Marc Wilchesky, John Corson, Dick Jones, and Gerald Patterson for assisting in various aspects of this study.     

Diagnostic importance of combined parent and teacher ratings on the Revised Behavior Problem Checklist

The Revised Behavior Problem Checklist (RBPC) was completed by parents and teachers for 105 child psychiatry outpatient boys aged 6 to 12 years. In addition, DSM-III diagnoses were determined independent of checklist results, and the children were divided into major diagnostic groups of externalizing, internalizing, and mixed disorders. Combined parent and teacher ratings proved more effective than separate ratings in distinguishing the externalizing and mixed groups from the internalizing group on the RBPC externalizing factors. The overall classification of individual boys with discriminant function analysis increased from 72% to 83% when the ratings of both parents, rather than one parent, were combined with teacher ratings. Enhancing diagnostic accuracy of type of psychiatric disorder by using combined parent and teacher behavior checklist ratings is discussed.     

A multitrait-multimethod analysis of the preschool behavior questionnaire

The Preschool Behavior Questionnaire is a teacher rating instrument yielding a total score reflecting overall level of adjustment and three subscale scores labeled Hostile/Aggressive, Anxious/Fearful, and Hyperactive/Distractible. This study assessed the convergent and discriminant validity of this teacher rating measure against two alternative measures within a multitrait-multimethod analysis. The results supported the construct validity of the Hostile/Aggressive and Anxious/Fearful scores as well as the overall score. There was, however, no support for the validity of the Hyperactive/Distractible score.This research was supported, in part, by a grant from the Pickering Foundation. The authors would like to thank Maria Bhaneja for her help in the collection and coding of data. Thanks are also extended to the teachers and children who participated in the study.