BEHAVIORAL INTERVENTION ON SCHOOL PERFORMANCE 1
Effects of Cognitive-Behavioral Intervention on the School Performance of Students with
Emotional or Behavioral Disorders and Anxiety
Stephanie L. Kiley
Coastal Carolina University
EDSP 670
Article Critique
BEHAVIORAL INTERVENTION ON SCHOOL PERFORMANCE 2
Purpose
The purpose of the article “Effects of Cognitive-Behavioral Intervention on the
School Performance of Students with Emotional or Behavioral Disorders and Anxiety” was to
examine the changes in anxiety, maladaptive behavior, and academic engagement as functions of
participation in the cognitive-behavioral anxiety intervention, FRIENDS for Life. I chose this
study because students who experience anxiety report that they have difficulty concentrating and
doing homework and are less likely to complete high school than their nonanxious peers. All
students should have equal learning opportunities, regardless of their disabilities. I am interested
to see how well the effects of the cognitive-behavioral intervention works on students with
emotional or behavioral disorders and anxiety.
Participants and Setting
The study was conducted in a private school for students with emotional or
behavioral needs, grades 1 through 12, in the Southwestern United States. Participants were
recruited from the program’s fourth-through sixth-grade classrooms. Teachers were asked to
nominate students, who they felt exhibited a marked degree of anxiety, resulting in a list of 11
students. These nominations were then narrowed down to students who had exhibited significant
levels of anxiety on a school-administered psychological measure less than 1 year old, such as
the Behavior Assessment System for Children. Ten nominated students met the criteria for
generalized anxiety. Of these students, the first 3 to supply completed parental permission forms
were selected as participants. All participants were male, attending sixth grade. Two were
Caucasian, and one was Hispanic. All participants had attended the school for at least one full
BEHAVIORAL INTERVENTION ON SCHOOL PERFORMANCE 3
year and continued to participate in school-based counseling sessions that had been in effect
prior to the study.
Participant A was 11 years old and classified as EBD. His grade-level functioning
was at third grade in reading and fourth grade in math, and his reported IQ was 107. Participant
B was 11 years old and classified as EBD. His grade level functioning was at fourth to fifth-
grade in reading and third grade in math. His reported IQ was 98. Participant C was 12 years old
and classified as EBD with additional accommodations for a learning disability related to
reading. His grade-level functioning was at a third-grade level in reading and fifth-grade level in
math. The most recent school psychologist report in his file reported an IQ of 91.
Methods
A single-subject, multiple-baseline design across students was used to examine the
effects of intervention on academic engagement and school-appropriate behavior. Single-subject
designs look for systematic variance of dependent variables that occurs with the manipulation of
a chosen independent variable (Horner et al., 2005; Tawny & Gast, 1984), in this case,
systematic variance in academic engagement and behavior occurring with exposure to cognitive
behavioral intervention for anxiety. The third dependent measure, the teacher-reported symptoms
of student anxiety, served as a pre-post measure of student anxiety and was taken prior to
baseline and again after the intervention phase. Intervention for the first participant began once a
stable baseline was achieved. Intervention for subsequent participants was initiated sequentially
after the immediately preceding participant had demonstrated three data points in the desired
direction of change and completed FRIENDS session 5. This postponement to a minimum of
five sessions ensured that data points counted toward change would postdate initial sessions that
BEHAVIORAL INTERVENTION ON SCHOOL PERFORMANCE 4
focused on establishing interventionist-participant rapport rather than introducing the concepts of
the intervention itself. Interobserver agreement (IOA) data were collected by doctoral students
specializing in EBD at a nearby university, during alternating observation sessions, for a total of
55% of sessions during all phases. IOA was calculated by dividing the number of intervals and
multiplying the result by 100, and it averaged 93% overall with a range of 84% to 99%.
To prevent changes in the behaviors of interest due to the presence of the study, rather
than the intervention itself, information regarding the purpose of the study was controlled. The
lead data collector knew when implementation began, but the observers conducting IOA did not.
To verify that sessions followed the topics and sequence of the FRIENDS program,
the authors developed a checklist of the major and minor activities unit completed in each. To
ensure that the submitted checklists corresponded to actual session content, treatment fidelity
