See discussions, stats, and author profiles for this publication at: [Link]

net/publication/51254228

Check-in/check-out: A systematic evaluation and component

analysis

Article  in  Journal of Applied Behavior Analysis · June 2011

DOI: 10.1901/jaba.2011.44-315 · Source: PubMed

CITATIONS READS

55 136

2 authors, including:

Cynthia M Anderson
May Institute
63 PUBLICATIONS   1,360 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Cynthia M Anderson on 28 May 2014.

The user has requested enhancement of the downloaded file.

JOURNAL OF APPLIED BEHAVIOR ANALYSIS 2011, 44, 315–326 NUMBER 2 (SUMMER 2011)

CHECK-IN/CHECK-OUT: A SYSTEMATIC EVALUATION AND

COMPONENT ANALYSIS
AMY CAMPBELL AND CYNTHIA M. ANDERSON
UNIVERSITY OF OREGON

Tier 2 interventions are implemented similarly across students and thus serve as an efficient and
cost-effective method of behavior support in school settings. Check-in/check-out is a Tier 2
intervention with documented effectiveness (e.g., Hawken & Horner, 2003; Todd, Campbell,
Meyer, & Horner, 2008). Key features of the intervention include brief morning and afternoon
meetings with the intervention coordinator, use of a point card on which the teacher monitors
student behavior, and teacher feedback at predetermined times. The present study sought to add
to the literature by examining the relative contributions of the teacher-feedback components of
check-in/check-out via the use of a component analysis. Working with 4 children in a general
education setting, we first evaluated the effectiveness of the procedure using reversal designs.
Next, we systematically removed teacher-feedback components to assess effects on problem
behavior and academic engagement. For 3 of 4 participants, we were able to remove all teacher-
feedback sessions and the point card; for the 4th participant, we removed only 2 of 3 teacher-
feedback sessions due to time constraints.
Key words: classrooms, conditioned reinforcement, feedback, schools, token economy
________________________________________

Schools in the United States are increasingly preintervention functional assessment to guide
implementing three-tiered prevention-oriented development of an individualized multicompo-
models to guide social behavior interventions nent intervention. Although a substantive
(Greenwood, Kratochwill, & Clements, 2008; literature base supports Tier 1 (e.g., Barrett,
Gresham, 2004). Tier 1 supports are imple- Bradshaw, & Lewis-Palmer, 2008; Horner,
mented with all students to prevent the Sugai, & Anderson, 2010; Horner et al.,
development of social behavior problems. 2009; Putnam, Luiselli, & Sunderland, 2002)
Interventions at this level consist of explicit and Tier 3 (e.g., Carr et al., 1999; Ervin et al.,
instruction in prosocial behavior and often 2001; March & Horner, 2002; Sasso, Conroy,
include consequences for appropriate and Peck-Stichter, & Fox, 2001), the literature base
inappropriate behavior. Tier 2 interventions for Tier 2 interventions is small (Anderson &
are implemented with students who do not Borgmeier, 2010).
respond to the Tier 1 support and supplement A promising Tier 2 intervention is check-in/
the primary intervention. Tier 2 interventions check-out (CICO, also known as the behavior
are implemented similarly across students who education plan; Crone, Horner, & Hawken,
receive the intervention. Finally, Tier 3 supports 2003). CICO is an extension of home–school
are for students with severe problem behavior notes (Davies & McLaughlin, 1989; Dougherty
that has not responded to Tier 1 or 2 & Dougherty, 1977; Schumaker, Hovell, &
interventions. Tier 3 interventions involve a Sherman, 1977) designed for implementation
within a three-tiered framework. Thus, the
This work was completed in partial fulfillment of the intervention is implemented similarly across
requirements for a PhD in psychology by the first author,
who is now at Grand Valley State University. students (i.e., it is manualized). CICO is
Address correspondence to Cynthia M. Anderson, designed for students who exhibit nondanger-
Department of Special Education and Clinical Sciences, ous problem behavior during academic routines
5261 University of Oregon, Eugene, Oregon 97403 (e-
mail: canders@[Link]). and is designed to supplement the Tier 1
doi: 10.1901/jaba.2011.44-315 intervention by (a) providing more frequent

315
316 AMY CAMPBELL and CYNTHIA M. ANDERSON

instruction regarding expected behavior, (b) with the exception of McIntosh et al., 2009)
increasing structured contact between students and none included a reversal of the independent
and adults in the school, (c) providing a formal variable to assess functional control.
mechanism for students to receive feedback on Another important research direction is to
their behavior, and (d) increasing opportunities explore the relative effects of various compo-
for reinforcement contingent on expected nents of CICO. The intervention consists of
behavior. Together, the various components of several components, including morning and
CICO result in increased interaction between afternoon meetings with the intervention coor-
students and adults. Research suggests that dinator, scheduled feedback from teachers, a
CICO is most effective for students whose point card, a token economy, and a home–
behavior is maintained by adult attention rather school note. It is not clear whether all
than students whose behavior is maintained by components of CICO are necessary for success.
peer attention or avoidance of academic tasks If CICO is effective with fewer components, the
(Campbell & Anderson, 2008; Fairbanks, intervention could be simplified. The present
Sugai, Guardino, & Lathrop, 2007). study expands the literature on CICO by
To date, CICO has been evaluated in eight replicating findings of previous studies using a
experimental or quasiexperimental studies. Of more rigorous experimental design and assess-
these studies, the majority have focused on ing the relative contribution of the teacher-
changes in office discipline referrals, an indirect feedback portion of CICO.
measure of behavior change (Filter et al., 2007;
Hawken, MacLeod, & Rawlings, 2007; March METHOD
& Horner, 2002; McIntosh, Campbell, Carter, Setting and Participants
& Dickey, 2009). In the most rigorous The study took place in a suburban elemen-
examination to date, McIntosh et al. (2009) tary school (K–5) located in the Pacific
used a multivariate analysis of variance to Northwest. During the 2006–2007 school year,
examine effects of CICO on office discipline 72% of the students qualified for free or
referrals and teacher reports on a norm- reduced-price lunch services. The school had
referenced rating scale. They found that, for been implementing schoolwide positive behav-
students whose problem behavior was suggested ior support (SWPBS) for approximately 5 years.
by a teacher interview to be maintained by In this school, the Tier 1 component of SWPBS
attention, CICO resulted in statistically signif- was fully in place (as measured by the School-
icant differences in teacher ratings and office wide Evaluation Tool; Sugai, Lewis-Palmer,
discipline referrals from baseline to treatment. Todd, & Horner, 2001) and consisted of (a)
Studies using direct observation to assess behavioral expectations (be safe, be respectful,
effects of CICO have shown that the interven- be responsible) that had been operationally
tion reduces the frequency of problem behavior defined for specific settings and explicitly taught
during academic routines (Fairbanks et al., to all students, (b) a schoolwide token economy
2007; Hawken & Horner, 2003; March & for prosocial behavior, and (c) a continuum of
Horner, 2002; Todd, Campbell, Meyer, & consequences for rule violations (e.g., reteach-
Horner, 2008) and increases academic engage- ing, office referral). The school had been
ment (Hawken & Horner, 2003; March & implementing CICO as a Tier 2 intervention
Horner, 2002). Although these results are for 2 years. Students exposed to CICO
promising, more research is needed, because (including participants in this study) continued
only four of the existing studies used adequate to participate in all aspects of the Tier 1
experimental designs (all used single subjects, intervention. Students were selected to partic-
 EVALUATION OF CICO 317

ipate in the study if (a) they had from two to category of specific learning disability in
five office discipline referrals and the school’s reading. He received small-group reading
SWPBS team believed that CICO would be a instruction from the special education teacher,
good fit, or (b) they were nominated for CICO but all other instruction was delivered in the
by a teacher. general education setting. He had been referred
Kyle was a second-grade Caucasian student to the study by his teacher because of disruption
who qualified for special education services and noncompliance, including talking to peers
under the category of specific learning disability when the expectation was to be quiet, making
in reading. In addition, he had been diagnosed inappropriate noises (tapping pencil on desk,
with attention deficit hyperactivity disorder humming), and refusing to complete indepen-
(combined type) and was taking methylpheni- dent work assignments.
date (levels remained constant throughout the
study). He received 50% of his reading Design and Procedure
instruction in a general education setting, with A functional assessment was conducted first.
the remaining time spent in small-group Next, a reversal design was used to assess effects
instruction in a resource room setting. He had of CICO on problem behavior and academic
been referred to the study by his classroom engagement. The component analysis of CICO
teacher due to disruptive behavior (talking to began when a participant had earned 80% or
peers during large-group instruction, making more of possible points for 15 consecutive days
noises, being out of seat). and, for at least 5 days, an 80% or greater
reduction was observed in problem behavior
Mike was a fifth-grade Caucasian student
relative to the last three baseline points.
who received all instruction in a general
education setting. Although he did not qualify Functional Assessment
for special education, he received additional
The functional assessment consisted of a
small-group reading instruction from an assis- teacher interview and observations. The inter-
tant (in his large-group reading class). He had view was conducted using a revised version of
been referred by his teacher due to disruptive the Functional Assessment Checklist for Teach-
behavior (talking to peers during instruction, ers and Staff (Anderson & Borgmeier, 2007).
being out of seat) and failure to complete After the interview, three 10-min observations
assignments. were conducted in the context (e.g., oral
Nick was a fifth-grade Latino student who reading) in which problem behavior was
qualified for special education services under the reported to occur most frequently. During
category of speech and language impairment. He observations, data were collected using 5-s
had been diagnosed with attention deficit partial-interval recording. Data were collected
hyperactivity disorder (combined type) and was on problem behavior, adult attention (verbal or
taking amphetamine and dextroamphetamine physical interaction), task avoidance (removal of
(dosages remained constant throughout the task or lack of engagement), and peer attention
study). He received all academic instruction in (verbal or physical interaction).
the general education setting. He had been
referred to the study by his classroom teacher due Baseline and Training in CICO
to disruptive and off-task classroom behavior, During baseline, teachers were instructed to
including being out of seat and talking to peers conduct their class and manage student behav-
when independent work was expected. ior as they usually did. Because the Tier 1
Paul was a fifth-grade Caucasian student. He schoolwide intervention was in place through-
qualified for special education under the out the study, participants may have received
318 AMY CAMPBELL and CYNTHIA M. ANDERSON

formal acknowledgment for meeting school the specific times for morning, noon, and
expectations and office referrals for violations afternoon feedback sessions based on natural
of school rules. transitions during the day (e.g., before reading
Prior to starting the CICO program, each started, after recess, before lunch, after math),
participant attended two 30-min training and feedback was based on student behavior
sessions with the CICO coordinator and the during the time period since the last feedback
first author. These sessions focused on (a) session. For example, the afternoon feedback
teaching the child the routines of the CICO session focused on behavior that had occurred
program, (b) providing examples and non- since the noon feedback session rather than the
examples of desired behaviors, and (c) practice entire day. During each feedback session, the
checking in and checking out in a specific teacher awarded participants up to three points
location. Students were also told how they for emitting behaviors congruent with each of
could earn their weekly points and how their the schoolwide expectations (i.e., be safe, be
goals were set. respectful, and be responsible). Points were
Check-in/Check-out assigned using a 3-point scale (1 5 student had
a hard time, 2 5 student did okay, 3 5 student
Check-in/check-out consisted of a morning
did great). Therefore, students could earn up to
meeting with the CICO coordinator, periodic
9 points per feedback session, for a total of
feedback meetings with the participant’s teach-
45 points each day. Each student’s goal was to
er, an afternoon meeting with the coordinator,
and home feedback. The CICO coordinator, earn at least 80% of possible points each day
who was a special education teacher employed (this was the standard goal for all students in the
by the school, was responsible for introducing school). Study participants were told the
students to CICO, reviewing the process with minimum number of points they needed to
teachers when a student started the interven- earn each day (to reach 80% of points), because
tion, entering data into the schoolwide data the total points possible varied across phases. In
system, and meeting with the school’s SWPBS this school, students on CICO earned tangible
team on alternate weeks to review CICO data and intangible rewards each week based on the
and to monitor progress of all participants. In number of points accumulated during prior
addition, the coordinator conducted brief weeks. The point value for items ranged from
morning check-in and afternoon check-out relatively small (e.g., 10 points for a pencil,
with all participants. 20 points for a notepad) to large (e.g., 100
Each morning before school started, students points for 5-min additional recess, 300 points
had a brief (approximately 2 min) check-in with for lunch with a favorite adult such as the
the coordinator at the school. Students received principal). Because the number of points
their daily point card, turned in the previous possible for study participants varied, points
day’s parent report form, and engaged in a toward rewards were based on the percentage of
short, positive interaction with the CICO possible points earned each day (10 points for
coordinator. The CICO daily point card was earning 100% of the possible points, 9 points
a piece of cardstock paper (4 in. by 5 in.) that for 90%, and 8 points for 80%).
showed five opportunities for the student to Teachers were not given instructions on how
receive feedback and earn points for appropriate to respond to problem behavior or to specific
behavior: check-in, morning, noon, afternoon, instances of prosocial behavior; the intervention
and check-out. Points for morning check-in and focused entirely on use of the point card to
afternoon check-out were awarded by the provide feedback at set points and the awarding
CICO coordinator. The teachers determined of points contingent on meeting school expec-
 EVALUATION OF CICO 319

tations. All teachers attended a training session coordinator at the beginning and end of each
on implementation of CICO at the start of the day, and points were earned only for checking
school year, and the coordinator met individ- in and out; point delivery was not contingent
ually with teachers and reviewed the interven- on behavior during the school day. In the final
tion whenever a teacher’s student began CICO. phase of the analysis, when the point card was
At the end of the day (typically 5 to 10 min removed, students earned 100% of points
prior to dismissal) the student walked to the simply by attending the morning and afternoon
specified check-out location (e.g., the counsel- meetings with the coordinator.
or’s office) and gave the daily report card to the
coordinator. The coordinator recorded the Response Measurement and
number of points earned, completed a parent Interobserver Agreement
report, and provided feedback to the student Data were collected on problem behavior,
regarding the student’s points. If the student academic engagement, and the percentage of
earned 80% or more of the possible points, the points earned. Problem behavior included
coordinator provided praise. If the student disruption, out of seat or location, noncompli-
earned less than 80% of points, the coordinator ance, and verbal or physical interaction.
provided neutral feedback (e.g., ‘‘What can you Disruption was defined as making inappropriate
do to meet the goal tomorrow?’’). noises, talking out of turn, and using objects in
Feedback was provided to the parent via a a manner for which they were not designed
daily home report that indicated whether or not (e.g., banging the desk, slamming a book). Out
the student earned 80% or more of points. of seat or location was defined as the student’s
Space was provided for comments from the buttocks not in contact with the seat when the
coordinator, and parents were asked to sign the expectation was to be seated or not in the
form. Parents were encouraged to provide location he was supposed to be in. Noncompli-
incentives at home when their child had a good ance was defined as verbally refusing to follow
day and to refrain from delivering punitive an adult direction or simply not completing the
consequences if their child did not earn 80% or request within 10 s, and negative verbal or
more of points; however, no data were collected physical interactions were defined as hitting,
to determine whether this occurred. kicking, pinching, throwing objects at a person,
or making a derogatory comment about or
Component Analysis towards another individual. Data were collected
Participants continued to meet with the on academic engagement, because all teachers
CICO coordinator at morning check-in and stated that problem behaviors were interfering
afternoon check-out. To assess the relative with the child’s engagement during instruction;
contribution of teacher feedback, we systemat- decisions about phase changes, however, were
ically removed feedback meetings. First, noon based on levels of problem behavior.
feedback was removed; students continued to Academic engagement was defined as follow-
receive feedback once in the morning and again ing teacher requests within 10 s, having eyes
in the afternoon. Next, morning feedback was oriented toward teacher or relevant materials for
removed; students now received feedback only the task, and working on in-class assignments.
in the afternoon. Finally, the afternoon feed- Data were collected using pen and paper across
back session was removed. This necessarily 15-min observations using a 5-s partial-interval
involved removal of the point card in the system. Observations were conducted 3 to
classroom setting because students no longer 5 days per week during academic activities that
checked in with teachers to review the point were suggested by the functional assessment to
card. Students continued to meet with the be when problem behavior most often occurred.
320 AMY CAMPBELL and CYNTHIA M. ANDERSON

These activities were large-group reading for (c) the staff member provided a prompt for the
Kyle, independent math work for Mike and student to be successful that day, (d) the student
Nick, and any independent work for Paul. turned in the home report, (e) the student
The school’s CICO coordinator tracked approached the teacher to receive feedback, (f)
points earned by all students on the program. the teacher assigned points to the student, (g)
The coordinator calculated the percentage of the teacher provided verbal feedback regarding
possible points earned by a student each day by the student’s behavior, (h) the student checked
dividing the number of points earned by the out with the adult, (i) the student presented the
number of points possible. In addition, the completed card to the adult, (j) the staff
percentage of days in which the student met his member added up and recorded total points,
goal was calculated by dividing the numbers of (k) the staff member provided verbal feedback
days the goal was met by the total number of regarding the student’s behavior, and (l) the
days. It should be noted that we asked teachers staff member completed the parent report and
to continue to complete the point card (but not handed it to the student.
to review it with students) during all phases of Interobserver agreement was assessed for
the component analysis so we could assess the 25% of fidelity observations. Agreement was
percentage of points awarded throughout the calculated by dividing the total number of items
study. that were scored the same by the total number
Interobserver agreement was assessed by of total items and multiplying by 100%.
having two independent observers simulta- Agreement averaged 97% (range, 91% to
neously collect data during 32% of observations 100%). Across participants, fidelity was high,
across all four participants. Agreement was with an average of 97% (range, 83% to 100%).
calculated by dividing the number of intervals
in which both observers agreed a response did
Social Validity and Contextual Fit
or did not occur by the number of total
intervals and multiplying by 100%. For Teachers’ perceptions of student problem
problem behavior, agreement coefficients were behavior and contextual fit were assessed once
97% (range, 91% to 100%) for Kyle, 95% or twice per week throughout the study using a
(range, 89% to 100%) for Mike, 91% (range, 2-item questionnaire. The first question asked
90% to 100%) for Nick, and 93% (range, 89% teachers to rate the target student’s problem
to 100%) for Paul. For academic engagement, behavior compared to other students in the class
agreement coefficients were 94% (range, 89% using a 5-point Likert scale (1 5 much better
to 97%) for Kyle, 90% (range, 89% to 95%) than other students, 5 5 much worse than other
for Mike, 95% (range, 90% to 100%) for Nick, students). The second item asked teachers to rate
and 90% (range, 89% to 93%) for Paul. the amount of effort put into managing student
behavior compared to other students in the class
Fidelity of Implementation (1 5 little or no effort, 5 5 tremendous effort).
Fidelity of implementation was measured on Contextual fit was assessed during the initial
27% of the days that the students participated CICO phase and at the completion of the study
in CICO. Experimenters observed the morning via the teacher-completed Contextual Fit Ques-
check-in and afternoon check-out and class- tionnaire (Horner, Salantine, & Albin, 2003)
room feedback sessions. The observers com- This questionnaire assessed the teachers’ per-
pleted a 12-item checklist that rated the ceptions of the ease of implementation, amount
presence of the following key features of CICO: of effort needed to implement the intervention,
(a) The student checked in with the adult, (b) and whether the effects of the intervention were
the staff member provided the daily point card, worth the effort.
 EVALUATION OF CICO 321

Figure 1. Conditional probability of a consequence given problem behavior in the functional assessment.

Results of the functional assessment observa- to assess whether CICO could be faded
tions are depicted in Figure 1. We calculated completely, because the academic year ended
the conditional probability that a given conse- for summer vacation.
quence occurred during the same or subsequent Mike exhibited problem behavior during an
5-s interval of the problem behavior by dividing average of 28% of intervals during baseline.
the number of intervals that problem behavior Implementation of CICO resulted in an
was followed by a given consequence in the immediate decrease in problem behavior (M
same or next interval divided by the total 5 11%). A brief return to baseline was
number of intervals scored with problem implemented to assess functional control;
behavior. For all participants, a greater propor- however, due to the intensity of his behavior
tion of problem behavior was followed by adult and teacher request, the reversal was limited to
attention than by escape from task or peer two data points. After the return to CICO,
attention. This suggested that the behavior was problem behavior again occurred in relatively
maintained, at least in part, by adult attention. few intervals. The noon and morning feedback
Intervals scored with problem behavior are sessions were removed successfully. Slight
depicted in Figure 2 for all participants. increases in problem behavior were observed
Intervals scored with academic engagement are when Mike no longer received teacher feedback.
shown in Figure 3. In all figures, phases of the Thus, his teacher was concerned that Mike
component analysis are labeled by the number would ‘‘revert to his old ways’’ and requested
of feedback sessions students had with teachers. that the point card and the afternoon feedback
Kyle emitted problem behavior during an be reinstituted.
average of 37% of intervals during baseline. Problem behavior for Nick occurred during
After implementation of CICO, an immediate an average of 28% of intervals during baseline.
reduction of problem behavior was observed, Problem behavior decreased immediately after
with problem behavior occurring in an average implementation of CICO (M 5 8%). Follow-
of 12% of intervals. After a brief return to ing a brief return to baseline (shortened due to
baseline to assess functional control, CICO was teacher request), CICO was reinstated, and
reinstated, resulting in an immediate decrease in problem behavior immediately decreased and
problem behavior. Both the noon and morning remained low as feedback sessions were re-
feedback sessions were removed successfully, moved sequentially. Nick’s teacher requested
such that Kyle received teacher feedback only in that the point card and afternoon feedback
the afternoon. Unfortunately, we were not able session be reinstituted because she was con-
322 AMY CAMPBELL and CYNTHIA M. ANDERSON

Figure 2. Percentage of 5-s intervals scored with problem behavior across conditions for Kyle, Mike, Nick, and Paul.
The numbers in the phase lines delineate the number of teacher-feedback sessions each participant attended.
 EVALUATION OF CICO 323

Figure 3. Percentage of 5-s intervals scored with academic engagement across conditions for Kyle, Mike, Nick, and
Paul. The numbers in the phase lines delineate the number of teacher-feedback sessions each participant attended.
324 AMY CAMPBELL and CYNTHIA M. ANDERSON

cerned that he ‘‘couldn’t succeed’’ without it. low during the first and second fading phases.
Problem behavior continued to occur infre- During the third fading phase, the average
quently thereafter. teacher rating increased a bit (M 5 2.7) but still
Problem behavior for Paul was variable remained well below ratings provided during
during baseline, occurring in 23% of intervals baseline. The teachers’ perceived effort of
on average. Although some variability remained managing student behavior also decreased when
after introduction of CICO, an overall reduc- CICO was implemented. Teachers’ mean rating
tion in problem behavior was observed (M 5 of effort was 3.7 (1 5 little or no effort, 5 5
16%). Following a short return to baseline, tremendous effort) during baseline, 2.1 during
CICO was again associated with a decrease in CICO, and 1.6 during the first two fading
problem behavior (M 5 8%), and the noon, phases. However, mean ratings of effort re-
morning, and afternoon feedback sessions were turned to 3.7 during the third fading phase.
removed successfully. As with the other partic- Teachers rated the contextual fit of CICO as
ipants, Paul’s teacher worried that his problem high during the initial implementation of
behavior would increase if the feedback CICO (M 5 98%) as well as at the completion
remained withdrawn; thus the afternoon feed- of the study (100% for all participants).
back session plus point card were reinstituted.
For all participants, implementation of DISCUSSION
CICO resulted in increases in academic en- The present study contributes to the literature
gagement (Figure 3), which was maintained as supporting CICO as an effective secondary
the teacher removed feedback sessions system- intervention in the context of a three-tier
atically. During CICO, the average percentage prevention system, SWPBS. As in previous
of points earned was 84% for Kyle, 82% for studies, CICO was shown to result in significant
Mike, 90% for Nick, and 86% for Paul. When reductions in problem behavior (e.g., Fairbanks
the noon and morning feedback sessions were et al., 2007; Hawken & Horner, 2003; March &
removed, participants continued to earn a high Horner, 2002; Todd et al., 2008). Further, gains
percentage of points (above 84% for each in academic engagement were observed when
participant). A slight reduction in points earned CICO was implemented. In this and other
was observed when the final feedback session studies, CICO was implemented by school staff
was removed (Ms 5 79% for Mike, 83% for over a 2-year period, suggesting that teachers and
Nick, and 72% for Paul). Because participants staff viewed the intervention to be feasible and
continued to earn points for checking in and useful, a finding supported by the social validity
out, this degradation in points awarded oc- data collected in this study.
curred in the classroom, suggesting that teachers One goal of this study was to assess the
may have noted some deterioration in student relative contribution of teacher feedback on the
behavior. effect of CICO via a component analysis.
During baseline, the average teacher rating of During CICO, all participants received feed-
participant problem behavior compared to back from teachers three times per day and met
other students in the class was 3.5 (1 5 much briefly with the coordinator each morning and
better than other students, 5 5 much worse than afternoon. All teacher feedback sessions were
other students). When CICO was implemented, removed for three participants; for the remain-
the rating decreased to an average of 2.1, ing participant, only two of the three feedback
indicating that the participants’ behavior was sessions were removed due to time constraints.
perceived to be similar to or better than typical Unfortunately, the extent to which problem
peers. The average rating continued to remain behavior would have remained low in the
 EVALUATION OF CICO 325

absence of the teacher-feedback session and teacher (giving the teacher the card when
point card is unknown because teachers re- entering the class, receiving periodic feedback)
quested that they be permitted to reinstate the or meetings with the coordinator act as
afternoon feedback session and point card. abolishing operations, reducing the effectiveness
These findings, although preliminary, suggest of teacher attention as a reinforcer for problem
several directions for future research, including behavior. Several studies have documented that
a replication of the current study to determine presession delivery of attention reduces atten-
whether the effects of the intervention will be tion-maintained problem behavior (e.g., McCo-
maintained over time without teacher-feedback mas, Thompson, & Johnson, 2003; McGinnis,
sessions. In addition, experimenters might assess Houchins-Juarez, McDaniel, & Kennedy,
whether full implementation of CICO is 2010). The findings of McComas et al. are
necessary prior to removal of feedback sessions. especially relevant to the current study, because
An additional direction is a more explicit presession delivery of attention resulted in fewer
exploration of the relation between adult atten- instances of problem behavior for participants
tion and CICO. Previous research (March & whose behavior was maintained by adult
Horner, 2002; McIntosh et al., 2009) has shown attention but not for participants whose
differential outcomes for CICO based on the problem behavior was escape maintained.
function of problem behavior. In these studies, It also is possible that the most powerful
CICO was effective for students whose behavior component of CICO (and one not manipulated
was maintained by adult attention but ineffective in this study) was the morning check-in and
for students whose problem behavior was afternoon check-out with the intervention
maintained by escape or avoidance of academic coordinator. Perhaps this positive adult contact
tasks. The results of the present study are in school set the occasion for more prosocial
consistent with previous findings, in that CICO behavior. Future research that elucidates the
was effective for three participants whose prob- mechanisms by which CICO is effective should
lem behavior was suggested to be maintained by allow educators to manipulate those features to
adult attention. However, because an experimen- increase its effectiveness for students who might
tal functional analysis was not conducted, we not otherwise respond. In addition, it might be
cannot rule out the possibility that problem possible to simplify CICO by removing
behavior was maintained by other consequences components that are shown to be unnecessary.
or that attention delivery was simply temporally Another direction for exploration is the
correlated with problem behavior. extent to which a schoolwide intervention such
That CICO may be more effective for as CICO would be effective in schools that are
students whose behavior is sensitive to adult not implementing a prevention-oriented inter-
attention is interesting, because the precise vention. The school in this study had been
mechanism for this effect is not clear. In CICO, implementing SWPBS for some time. As a
teachers are simply instructed to complete the result, all teachers were familiar with the use of
point card based on student behavior and to explicit instruction and feedback for social
provide periodic feedback about performance. behavior, and the school administrator actively
No explicit instructions are provided regarding supported the use of CICO (e.g., by attending
how teachers should respond to problem meetings in which CICO data were presented
behavior. It thus seems unlikely that teachers and meeting with teachers whose implementa-
consistently withhold attention following prob- tion of CICO was poor). The extent to which
lem behavior when CICO is in place. It is high-fidelity implementation would be achieved
possible that the scheduled interactions with the in a school without such a history is unknown.
326 AMY CAMPBELL and CYNTHIA M. ANDERSON

REFERENCES Hawken, L. S., MacLeod, S. K., & Rawlings, L. (2007).

Effects of the behavior education program (BEP) on
Anderson, C. M., & Borgmeier, C. (2007). Efficient office discipline referrals of elementary school students.
functional behavior assessment: the functional assessment Journal of Positive Behavior Interventions, 9, 94–101.
checklist for teachers and staff (FACTS). Eugene, OR: Horner, R., Salantine, S., & Albin, R. (2003). Self
Educational and Community Supports. Available assessment of contextual fit in schools. Eugene, OR:
from [Link] Educational and Community Supports.
Anderson, C. M., & Borgmeier, C. (2010). Tier II Horner, R. H., Sugai, G., & Anderson, C. M. (2010).
interventions within the framework of school wide Examining the evidence-base for school-wide positive
positive behavior support: Essential features for behavior support. Focus on Exceptional Children, 42, 1–14.
design, implementation, and maintenance. Behavior Horner, R. H., Sugai, G., Smolkowski, K., Eber, L.,
Analysis in Practice, 3, 33–45. Nakasato, J., Todd, A. W., et al. (2009). A
Barrett, S., Bradshaw, C., & Lewis-Palmer, T. (2008). randomized, wait-list controlled effectiveness trial
Maryland state-wide PBIS initiative. Journal of assessing school-wide positive behavior support in
Positive Behavior Interventions, 10, 105–114. elementary schools. Journal of Positive Behavior
Campbell, A., & Anderson, C. M. (2008). Enhancing Interventions, 11, 133–144.
effects of check-in/check-out with function-based March, R. E., & Horner, R. H. (2002). Feasibility of
support. Behavioral Disorders, 33, 233–245. contributions of functional behavioral assessment in
Carr, E. G., Horner, R. H., Turnbull, A. P., Marquis, J. schools. Journal of Emotional and Behavioral Disor-
G., McLaughlin, D. M., McAtee, M. L., et al. (1999). ders, 10, 158–171.
Positive behavior support for people with developmental McComas, J. J., Thompson, A., & Johnson, L. (2003).
disabilities: A research synthesis. Washington DC: The effects of presession attention on problem
American Association on Mental Retardation. behavior maintained by different reinforcers. Journal
Crone, D. A., Horner, R. H., & Hawken, L. S. (2003). of Applied Behavior Analysis, 36, 297–307.
Responding to problem behavior in schools: The behavior McGinnis, M. A., Houchins-Juarez, N., McDaniel, J. L.,
education program. New York: Guilford. & Kennedy, C. H. (2010). Abolishing and establish-
Davies, D. E., & McLaughlin, T. F. (1989). Effects of a ing operation analyses of social attention as positive
daily report card on disruptive behavior in primary reinforcement for problem behavior. Journal of
students. B.C. Journal of Special Education, 13, Applied Behavior Analysis, 43, 119–123.
173–181. McIntosh, K., Campbell, A., Carter, D., & Dickey, C.
Dougherty, E. H., & Dougherty, A. (1977). The daily (2009). Differential effects of a tier two behavior
report card: A simplified and flexible package for intervention based on function of problem behavior.
classroom behavior management. Psychology in the Journal of Positive Behavior Interventions, 11, 82–93.
Schools, 14, 191–195. Putnam, R. F., Luiselli, J. K., & Sunderland, M. (2002).
Ervin, R., Miller, P., Bertsch, K., Piper, A., Ehrhardt, K., Longitudinal evaluation of behavior support inter-
& Poling, A. (2001). A descriptive analysis and vention in a public middle school. Journal of Positive
critique of the empirical literature on school-based Behavior Interventions, 4, 182–188.
functional assessment. School Psychology Review, 30, Sasso, G., Conroy, M., Peck-Stichter, J., & Fox, J. (2001).
193–210. Slowing down the bandwagon: The misapplication of
Fairbanks, S., Sugai, S., Guardino, D., & Lathrop, M. functional assessment for students with emotional or
(2007). Response to intervention: An evaluation of a behavioral disorders. Behavioral Disorders, 26,
classroom system of behavior support for second 282–296.
grade students. Exceptional Children, 73, 288–310. Schumaker, J. B., Hovell, M. F., & Sherman, J. A. (1977).
Filter, K., Benedict, E., McKenna, M., Horner, R. H., An analysis of daily report cards and parent-managed
Todd, A., & Watson, J. (2007). Check-in/check-out: privileges in the improvement of adolescents’ class-
A post-hoc evaluation of an efficient, secondary-level room performance. Journal of Applied Behavior
targeted intervention for reducing problem behaviors Analysis, 10, 449–464.
in schools. Education and Treatment of Children, 30, Sugai, G., Lewis-Palmer, T., Todd, A. W., & Horner, R. H.
69–84. (2001). School-wide evaluation tool. Eugene, OR: Educa-
Greenwood, C. R., Kratochwill, T. R., & Clements, M. tional and Community Supports, University of Oregon.
(Eds.). (2008). Schoolwide prevention models: Lessons Todd, A., Campbell, A., Meyer, G., & Horner, R. (2008).
learned in elementary schools. New York: Guilford. Evaluation of a targeted group intervention in
Gresham, F. M. (2004). Current status and future elementary students: The check-in/check-out pro-
directions of school-based behavioral interventions. gram. Journal of Positive Behavior Interventions, 10,
School Psychology Review, 33, 326–343. 46–55.
Hawken, L. S., & Horner, R. H. (2003). Evaluation of a
targeted intervention within a schoolwide system of Received March 12, 2010
behavior support. Journal of Behavioral Education, 12, Final acceptance July 20, 2010
225–240. Action Editor, Jennifer Zarcone