Effects of check incheckout on behavioral indices and mathematics generalization

With regard to the CICO intervention phase, an immediate decrease in level and trend mean of 17.9%; range, 12%-25% was observed as compared with the baseline phase.. With regard to the C

Effects of Check-in/checkout on Behavioral Indices and

Mathematics Generalization

Michael D Mong University of Southern Mississippi

Kristin N Johnson Eastern Illinois University Kristi W Mong Mong Psychological Associates

ABSTRACT: Check-in/checkout (CICO) is a behavioral intervention that is used to provide systematic feedback about a student's behavior at the beginning and end of each school day The purpose of the present study was to evaluate the effectiveness of CICO on decreasing problem behaviors and the collateral effects on mathematics performance for 4 at-risk elementary school students A multiple-baseline across-participants design using dyads was used to analyze problem behavior collected through direct classroom observations As ancillary measures, office discipline referrals per week as well as mathematic performance (i.e., digits correct per minute [DCPMj) were collected for each student Treatment integrity and acceptability and social validity were also measured Results indicate a decrease in problem behaviors as well as an increase in DCPM for each participant.

• When children receive office discipline

referrals (ODRs), they can often

simulta-neously exhibit a multitude of issues

includ-ing academic and behavioral problems

These problems rarely exist in isolation, and

in combination they put students in more

dramatic risk of school failure (Mclntosh,

Flannery, Sugai, Braun, & Cochrane, 2008)

Thus, the relationship between academic

performance and problem behaviors

pro-vides concern because of their documented

interaction (Maguin & Loeber, 1995; Roeser

& Eccles, 2000) Students with early behavior

difficulties are at greater risk for developing

academic problems (Fleming, Harachi,

Cortes, Abbott, & Catalano, 2004), and

students with early academic difficulties are

at greater risk for developing problems in

social behavior (Mclntosh, Horner, Chard,

Boland, & Good, 2006; Morrison, Anthony,

Storino, & Dillon, 2001)

Given the documented relationship

be-tween academic and behavior problems,

positive behavior support (PBS) experts

rec-ommend a three-tiered model of behavior

supports to prevent and intervene with

prob-lem behavior (Sugai & Horner, 2001) As

students progress through the tiers, the

inten-sity of the intervention increases, as does the

cost of resources The purpose of the first tier is

to provide primary prevention for all students Those students whose behaviors continue to

be discrepant from their peers are identified for additional support at Tier 2

Approximately 15% to 20% of the popu-lation will benefit from this level of support (Walker & Shinn, 2002) Tier 2 interventions offer at-risk students additional opportunities

to learn expected behaviors that lead to educational success (Lee, Sugai, & Horner, 1999) These services are provided in addition

to the core instruction (National Association of State Directors of Special Education, 2005) Tier 3 interventions are for those students who are highly discrepant from their peers either in behavioral excesses or deficits In addition, these students referred to Tier 3 may be nonresponsive to Tier 2 interventions

For behavior Filter and colleagues (2007) recommended interventions be implemented with students who have two or more ODRs but whose problem behaviors do not pose an immediate danger to self or others These interventions should be de!signed to be quickly accessed, flexible, and to bring about improve-ment (Hawken & Horner, 2003) One inter-vention that may meet these requirements is the check-in/checkout (CICO) program

The CICO program was developed as an

efficient intervention for reducing problem

behavior The CICO program was designed

to increase feedback and positive adult

atten-tion (Crone, Horner, & Hawken, 2004)

Previous research has shown that CICO is

an easy-to-implement behavioral strategy and

is effective at decreasing ODRs (Hawken &

Horner, 2003; March & Horner, 2002),

de-creasing observed problem behaviors

(Fair-banks, Sugai, Guardino, & Lathrop, 2007), and

increasing appropriate behavior (Todd,

Camp-bell, Meyer, & Horner, 2008) The CICO

program has been shown to be a relatively

simple and inexpensive intervention (Hawken,

MacLeod, & Rawlings, 2007) The CICO

program has also been shown to have

rela-tively high levels of teacher and staff

accept-ability Additionally, previous research (Todd

et al., 2008) has shown that typical school

personnel were able to effectively implement

the CICO intervention

If implementing CICO demonstrates

stu-dents engaging in fewer problem behaviors

and spending less time in the school office,

improved academic achievement may likely

follow (Hawken et al., 2007) Indeed, Hawken

and Horner (2003) documented increases in

academic engagement following CICO

imple-mentation by assessing if the student was (a)

looking at the teacher while the teacher was

giving instructions/directions, (b) working with

a peer when instructed to do so, (c) reading

silently or completing a writing assignment, (d)

participating in a teacher-approved activity if

work was completed, or (e) talking about

academic material with the teacher or aide

for at least 7 s Although the primary focus of

the study was on classroom problem

behav-iors, a secondary analysis indicated that the

CICO intervention was associated with

in-creases in mean level of academic

perfor-mance in mathematics for all 4 students

While the previous research is indeed

promising, few studies have examined CICO's

effect on outcome variables such as academic

performance and achievement (Hawken et al.,

2007) The main idea is that although CICO

may or may not have direct effects on academic

achievement, students who spend less time

engaging in inappropriate behaviors may

re-place those behaviors with appropriate

behav-ior Time on task or engagement leads to better

academic performance Furthermore, few

stud-ies have provided measures of CICO effect

size Reporting effect sizes is considered best

practice when presenting empirical research findings in many fields (Wilkinson, 1999) Another limitation of previous CICO research involves the manner in which social validity was assessed To date, few studies have measured the social validity of CICO using a measure with validated psychometric proper-ties

Thus, the purpose of the present study was to (a) extend the literature by examining CICO's effect on classroom problem behaviors as measured by direct behavioral observations and ODRs, (b) determine whether the effects of

a behavioral intervention (CICO) affected stu-dent mathematics performance, and (c) assess the social validity of CICO using a measure with validated psychometric properties

Method

Participants and Setting

The study was conducted in a suburban elementary school located in the southeast United States with approximately 415 students (Grades 3-5), 65% of whom qualified for free

or reduced lunch and 36% of whom were from ethnic minority backgrounds The school had been implementing schoolwide PBS for more than 2 years The Schoolwide Evaluation Tool (Horner, Sugai, Todd, & Lewis-Palmer, 2005) results indicated that the school was imple-menting its schoolwide behavior support plan with 82% treatment integrity

Students were selected for participation in the study if (a) they received five ODRs in a single month, (b) their problem behavior occurred frequently across multiple settings throughout the school day as noted from ODR analysis and teacher interviews, and (c) the results of a functional behavioral assessment (FBA) indicated that the function of the students' problem behavior was hypothesized

to be attention seeking The problem behaviors exhibited by these students are particularly alarming given that the long-term outcomes for students who exhibit early patterns of mal-adaptive behavior are continued poor aca-demic performance, referral for special edu-cation identifiedu-cation and placement, social rejection, low self-esteem, potential for devel-oping a more chronic psychological disorder, and increased school dropout rates (Walker, Ramsey, & Gresham, 2004)

Informed consent to participate was ob-tained for each student from his or her parent/

legal guardian Of the 4 students selected for

intervention, 2 were boys and 2 were girls

Students ranged in age from 8.4 to 9.1 years,

with a mean age of 8.7 years All students were

enrolled in the third grade In terms of race,

there were 2 Caucasian students and 2 African

American students

Lauren Lauren was an 8-year-old

Cauca-sian girl who has never received special

education services, has no previous diagnoses,

and has no history of retention The majority

(87%) of Lauren's ODRs resulted from

inap-propriate behavior in the classroom She was

referred by her classroom teacher for talking

out during classroom instruction, which

ac-counted for 18% of her total ODRs;

noncom-pliance with teacher demands (40%);

disrupt-ing peers durdisrupt-ing independent seat work (29%);

and off-task behavior across settings (13%)

including the classroom and cafeteria Based

on the results of the FBA, Lauren's problem

behaviors were hypothesized to be maintained

primarily by adult attention with peer attention

as a secondary function During the immediate

3 months before the study, Lauren averaged

2.1 (range, 1-3) ODRs per week

Andrew Andrew was an 8-year-old African

American boy who has never received special

education services, has no previous diagnoses,

and has no history of retentions Andrew was

receiving Title I services for reading and math

The majority of Andrew's ODRs (93%) resulted

from inappropriate behavior in the regular

classroom or music classroom during

instruc-tion He was referred by his classroom teacher or

music teacher for noncompliance with teacher

demands (37%), oft'-task behavior (22%), and

talking out (41%) in the general education class

as well as music and physical education classes

Based on the results of the FBA, Andrew's

problem behaviors were hypothesized to be

maintained by adult attention During the

immediate 3 months before the study, Andrew

averaged 4.1 (range, 2-5) ODRs per week

Pam Pam was a 9-year-old African

American girl who has never received special

education services, has no previous diagnoses,

and has no history of retentions The majority

of Pam's ODRs (92%) occurred in the

class-room during instruction and independent seat

work She was referred by her classroom

teacher for refusing to complete assignments

(14%), noncompliance with teacher demands

(54%), talking out in class (24%), and

non-compliance with bus driver demands (8%)

Based on the results of the FBA, Pam's problem

behaviors were hypothesized to be maintained primarily by adult attention with peer attention

as a secondary function During the immediate

3 months before the study, Pam averaged 1.9 (range, 1-3) ODRs per week

Stanley Stanley was an 8-year-old

Cauca-sian boy who has never received special education services, has no previous diagnoses, and has no history of retentions The majority

of Stanley's ODRs (92%) occurred in the classroom during independent seat work He was referred by his classroom teacher for off-task behavior (34%), noncompliance with teacher demands (17%), and talking out in class (41%) as well as off-task behavior in physical education (4%) and in schoolwide specials (4%) Based on the results of his FBA, Stanley's problem behaviors were hypothe-sized to be maintained primarily by adult attention with peer attention as a secondary function During the immediate 3 months before the study, Stanley averaged 2.0 (range, 1-3) ODRs per week

In addition, two guidance counselors and the students' classroom teachers agreed to participate in the study, with the counselors identified as the CICO specialists Prior to the initiation of each phase, the primary investiga-tor trained the teachers and CICO specialists on the required procedures associated with each phase of the study The training consisted of the primary investigator describing the procedures, modeling the procedures, and having the staff member practice the procedures with the primary investigator providing feedback This format was implemented until the individuals were able to implement the procedures inde-pendently The primary investigator was avail-able for any questions or concerns from the teachers and CICO specialists as they arose

FBA Procedures

Prior to initiation of the study, an FBA was conducted for each student The assessment process involved a 20- to 40-min interview conducted by the primary author with each participant's teachers using the Functional Anal-ysis Informant Record for Teachers (FAIR-T; Edwards, 2002) The purpose of the EAIR-T was to determine and help define the problem behaviors, to determine appropriate replace-ment behaviors, to determine potential reinforc-ers for appropriate behavior, and also to help formulate the hypothesized function of the presenting problem behavior To date, studies

have supported the use of the FAIR-T in such a

manner (Doggett, Edwards, Moore, Tingstrom, &

Wilczynski 2001; Dufrene, Doggett, Henington,

& Watson, 2007; Edwards, 2002) Following the

FAIR-T, three 20-min conditional probability

observations were conducted in the student's

natural classroom setting prior to intervention A

FAIR-T hypothesis statement was judged to be

supported if the direct observation data provided

similar information to the antecedent and

consequent events defined in the FAIR-T

hy-pothesis statement As CICO was designed to

increase feedback and positive adult attention

(Crone et al., 2004), it was important to match

the behavioral intervention to the function of the

participant's problem behaviors

Measurement

Direct observation of problem behavior.

Problem behavior was observed 3 days per

week using a 20-min, 10-s partial interval

recording system For each participant,

obser-vations took place during the same academic

class period each day The specific class period

for each student was determined by teachers'

reports of the most problematic time of day

based on the FAIR-T interview Problem

behav-iors included (a) noncompliance, (b) talking

out, and (c) off-task behavior Noncompliance

was defined as failure to complete assigned

instructions or failure to initiate commands

within 5 s Talking out was defined as the

student engaging in vocalizations that were not

preceded by a raised hand and/or were not

initiated by an adult Off-task behavior was

defined as the student oriented away (e.g., face

or body) from the teacher or materials during

instruction or oriented toward irrelevant

per-sons or objects, manipulating materials or

objects inappropriately that are relevant to the

assigned task or activity, or doing other

behaviors that are not related to his or her

assignment or task for a period of 10 s or more

For all students, the observations occurred in

mathematics as it was considered by the

teachers as one of the most problematic times

of the day Direct observation of problem

behavior was calculated by dividing the

num-ber of intervals with observed problem behavior

by the total number of intervals and multiplying

this ratio by 100 to obtain a percentage

Office discipline referrals The ODRs were

collected to compare the rates of problem

behaviors before and during the CICO

program and reported as a weekly measure

As a measure of behavior, ODRs possess sufficient construct validity and adequate concurrent validity with a number of standard-ized measures of individual behavior (as cited

in Irvin, Tobin, Sprague, Sugai, & Vincent, 2004) as well as predictive validity for negative school outcomes, including physical assaults and dropout (Tobin & Sugai, 1999) Office discipline referrals have also been identified as

an effective and efficient measure for decision making in schools (Irvin et al., 2006)

Percentage of daily progress reports points earned Each student's percentage of daily

progress report (DPR) points earned was examined as a measure of appropriate behav-ior Evaluations of the DPR suggest that it possesses internal consistency, temporal sta-bility, and concurrent validity and that its sensitivity allows for detection of treatment effects (Pelham, Fabiano, & Massetti, 2005)

Treatment Integrity

The CICO treatment integrity checklist.

Treatment integrity was measured in a similar manner used by Hawken et al (2007) For each integrity assessment, the first author examined the CICO daily student intervention protocol to determine the degree with which the CICO intervention was implemented as prescribed Treatment integrity was evaluated during 33% of the sessions evenly distributed across all phases of the study based on completion of a checklist during the session Treatment integrity was calculated by the number of items on the checklist completed divided by the total number of items on the checklist and multiplied by 100

Academic Pretreatment Assessment

A Web-based computer program Math Worksheet Generator (interventioncentral.org), was used to generate curriculum-based math-ematics worksheets The program allows the user to design worksheets requiring the use of specific skills State benchmarks were used to determine which skills were representative of each grade level The program was then used to create a worksheet specific to a particular grade level and state benchmark The computer program randomized the order of the (a) problems within a worksheet and (b) factors within each problem Each worksheet listed problems in six rowsof four problems in portrait orientation on a regular 8 'A- by 11 -inch sheet of

white paper Each worksheet contained the 24

problems Prior to beginning the study, the third

author used curriculum-based assessment

(CBA) procedures to identify each student's

instructional level Multiple skill CBA probes

were administered to determine the current

grade-level performance of each of the

stu-dents The student was given one worksheet at

his or her current grade placement in school For

example, a third-grade student was given an

opportunity to complete a third-grade-level

probe Each student was given 60 s to complete

each worksheet The digits correct per minute

(DCPM) on each worksheet was determined by

the number of digits written correctly during a 1

-min Interval, divided by the number of seconds

worked and multiplied by 60 Errors per minute

(EPM) served as a secondary dependent variable

Responses were scored as errors if incorrect

digits were written below the line or if digits

were written in the wrong place or omitted

If performance was determined to be in the

instructional-level range, a worksheet at the

same grade level was administered If the student

performed at the frustrational level (less than 14

DCPM), a worksheet at a lower grade level was

administered These procedures were repeated

until a median instructional level performance

was obtained across three worksheets within the

same grade level to establish baseline

According to Burns, VanDerHeyden, and

Jiban (2006), a student's independent

instruc-tional level was the point in the curriculum at

which he or she could complete math

prob-lems with 14 to 31 digits correct if enrolled in

first through third grade The independent

instructional level was the point in the

curriculum at which a student could complete

math problems obtaining 24 to 49 digits if

enrolled in Grades 4 and higher

During the 8 weeks of CICO intervention,

all students received math instruction once per

day and did not receive any math

interven-tions Furthermore, analysis of the classroom

teachers' weekly lesson plans indicated that

mathematics instruction focused on skills (i.e.,

estimating and rounding, analyzing graphs)

other than mathematics computation Thus,

the instruction in the classroom did not

necessarily directly address the skills on the

mathematic computation probes

Academic Progress Monitoring

To determine student progress, each student

completed three multiple skill grade-level

probes at the predetermined instructional level

to determine the student's median DCPM Problems included addition with two-digit num-bers with regrouping (i.e., 23 -i-18), addition with two three-digit numbers with regrouping (i.e.,

156 -H 379), subtraction with two two-digit numbers with regrouping (i.e., 48 — 19), sub-traction with two three-digit numbers without regrouping (i.e., 275 - 130), multiplication facts

to 10 (i.e., 3 X 5), and division facts to 10 (i.e., 8 -H 2) The multiple skill instructional-level probes were administered once per week The use of the multiple skill probes allowed the evaluation of level and rate of progress changes on instruc-tional-level material

Social Validity

The Behavior Intervention Rating Scale (BIRS; Treuting & Elliott, 1991) is a 24-item scale that employs a 6-point Likert-type format

(1 = strongly disagree to 6 = strongly agree) to

measure teachers' perceptions of treatment acceptability and the perceived efficacy of classroom interventions The BIRS is com-posed of three factors (Acceptability, Effective-ness, and Time of Effect) and a Cronbach's alpha of 97 for the total scale (Elliott & Treuting, 1991) The Acceptability factor is composed of 15 items with a Cronbach's alpha

of 97 The Effectiveness factor consists of seven items with a Cronbach's alpha of 92 The Time of Effect factor is composed of two items with a Cronbach's alpha of 87 (Elliot, 1998) The BIRS was completed by both guidance counselors and the participant's classroom teacher at the end of each week for a total of eight measurements per respon-dent The mean scores for each factor across the CICO intervention phase will be reported The five-item BEP Acceptability Question-naire (Hawken & Horner, 2003) was used to assess the social validity of the intervention with the participants Questions on the BEP Acceptability Questionnaire assessed the ex-tent to which the CICO was perceived to (a) improve problem behavior at school, (b) improve academic performance, (c) be worth the time and effort, (d) be worth recommend-ing to others, and (e) be easy to implement Scores on the BEP Acceptability Questionnaire were recorded on a Likert-type scale (1-6), with higher scores indicating a more favorable impression of the CICO intervention The participants responded to the questionnaire once per week for a total of eight

measure-ments per respondent The mean scores for

each question across the CICO intervention

phase will be reported

Reliability of Measures

Interobserver agreement for problem

be-havior observation data Interobserver

agree-ment (lOA) for problem behavior data was

collected using a second independent observer

for 33% of the sessions evenly distributed across

all phases of the study The lOA was calculated

by adding the number of intervals of agreement

of problem behavior for each session, then

dividing by the total number of observed

intervals for each session, and then multiplying

this ratio by 100 to obtain a percentage

Interobserver agreement of CICO sessions.

As an additional measure of treatment

integri-ty, the third author completed the CICO daily

intervention student protocol during direct

observation of 33% of all the CICO

interven-tion sessions Treatment integrity was

calcu-lated by the number of items on the checklist

completed divided by the total number of

items on the checklist and multiplied by 100

Interscorer agreement The primary author

designed a list of scoring instructions for the

mathematics probes used in the investigation

One scorer and the primary investigator scored

a sample of 15 probes independently The

rules were clarified and revised until there was

at least 90% agreement on a set of 45 sample

probes Scorers were then cleared for scoring

of the probes Approximately 33% of the total

probes were independently scored by the one

scorer and the primary investigator across all

phases of the study

Design and Procedure

The experimental design for this study was

a combined series multiple baseline across

students design The CICO intervention was

first applied to the student with the most stable

baseline in terms of problem behaviors After

an intervention effect was demonstrated and

the subsequent students' baselines remained

stable, the intervention was applied to the

student with the next most stable baseline

(Carr, 2005) To demonstrate empirical control

and to avoid delays to intervention, students

were paired in dyads to create two

multiple-baseline pairs for the current project Two

phases were implemented for this study:

baseline and CICO

Visual analyses for level, trend, and variability were used to determine effects as well as two statistical procedures for effect sizes Effect sizes were calculated using the percentage of nonoverlapping data points all (PND; Olive & Smith, 2005) PND is

calculat-ed by dividing the number of nonoverlapping data points with baseline by the total number

of intervention data points The lowest base-line data point was used to establish the overlap of baseline data points with interven-tion data points for observed behavior and ODRs, whereas the highest baseline data point was used for the percentage of DPR points earned Benchmarks for PND scores have also been established by Scruggs and Mastropieri (2002) Specifically, PND scores below 50% suggest an ineffective intervention effect, scores between 50% and 70% suggest a questionable intervention effect, scores be-tween 70% and 90% suggest an effective intervention effect, and scores greater than 90% suggest a very effective intervention effect Previous researchers (Campbell, 2004; Olive & Smith, 2005) concluded that PND is valid for documenting the effects of interven-tions in single-subject research

Baseline

During baseline, typical schoolwide behav-ior support procedures were in place for all students, including those participating in this study During baseline, direct observation of problem behavior and assessment of academic skills for target students was conducted The ODRs were monitored through collection of the written ODR reports In addition, the students' teachers were given three practice DPRs the week prior to implementation of CICO to ensure that only the target behaviors were rated These practice DPRs served as a baseline measure of the percentage of DPR points earned

Check-in/Checkout

The CICO process involved the following five elements: (a) Students were required to check in with the guidance counselor of their choice before school The counselor provided the student with a DPR form that was carried to class for feedback throughout the day When students checked in, they were asked if they had their DPR from the day before signed by their parents and if they had their materials ready for the school day They received praise

and a lottery ticket for a weekly drawing for

checking in Also during check in, students

were prompted to identify daily goals and

given feedback to encourage success, (b) At

three specified times of the day, students

approached the teacher with the CICO report

card, and the teacher provided the student

with feedback about the student's behavior

by rating either 0 (did not meet expectations),

1 {somewhat met expectations), or 2 (met

expectations) Teachers also provided

imme-diate verbal praise for students who met

behavioral expectations for that time period

and corrective feedback if students did not

meet the expectations, (c) At the end of the

school day, students took the DPR to the

counselor to check out The percentage of

points earned for the day was calculated, and

students received verbal praise and rewards if

they met their daily point goal Students could

choose from among specific rewards

deter-mined by a forced-choice preference

assess-ment including stickers, pencils, time with a

preferred adult, or extra recess time the

following day Similar to the procedures used

by Hawken et al (2007), 80% of the total

points earned was the daily goal for all

students in this study If students did not meet

their daily goal, the counselor provided

information on what to work on for the

following school day (d) Students then took

their DPR home to be signed by a parent/

guardian, (e) The DPR was signed by a parent/

guardian and returned the following morning

Results

Direct Observation of Problem Behavior

Figure I summarizes the results across

participants During baseline, all 4 participants

displayed variable levels of problem

behav-iors, with an overall mean of 32.8% (range,

21-42.7) Upon introduction of the CICO

intervention phase, Lauren and Andrew

dis-played an immediate reduction in level and

trend, whereas Pam and Stanley displayed

gradual decreases in level and trend The

participants were observed to engage in an

overall mean of 21.4% (range, 16.9-29.3)

across the CICO phase

Lauren At baseline, Lauren displayed

prob-lem behaviors in 31.3% (range, 28%-34%),

with an increasing trend and moderate

variabil-ity With regard to the CICO intervention phase,

an immediate decrease in level and trend (mean

of 17.9%; range, 12%-25%) was observed as compared with the baseline phase The PND for Lauren was 100%, suggesting that the CICO intervention was effective at decreasing her observed problem behaviors

Andrew On average, Andrew displayed

problem behaviors in 42.7% (range, 39%-48%) with relatively no trend with moderate variability across the baseline phase With regard to the CICO intervention phase, an immediate decrease in level and trend (mean

of 21.6%; range, 19%-26%) was observed as compared with the baseline phase The PND for Andrew was 100%, suggesting that the CICO intervention was effective at decreasing his observed problem behaviors

Pam At baseline, Pam displayed problem

behaviors in 2 1 % (range, 17%-24%) with increasing trend and moderate variability With regard to the CICO intervention phase,

a gradual decrease in level and trend was observed as compared with the baseline phase Pam averaged 16.9% (range, 10%-25%) of problem behaviors across the phase The PND for Pam was 63%, suggesting that the CICO intervention was somewhat effec-tive at decreasing her observed problem behaviors

Stanley At baseline, Stanley displayed

problem behaviors in 36% (range, 32%-40%) with a gradually increasing trend and moder-ate variability In the CICO intervention phase,

a gradual decrease in level and trend was observed as compared with the baseline phase Stanley averaged 29.3% (range, 25%-6%) of problem behaviors across the phase

The PND for Stanley was 75%, suggesting that

the CICO intervention was effective at de-creasing his observed problem behaviors

Office Discipline Referrals

Figure 2 summarizes the results across

participants During baseline, all 4 participants displayed slight variability with ODRs, with an overall mean of 3.3 (range, 1.7-5.0) ODRs per week Upon introduction of the CICO interven-tion phase, all 4 participants displayed a reduc-tion in level and trend Overall, the participants had a mean of 1.2 (range, 0.63-2.3) ODRs

Lauren At baseline, Lauren had 2.3 ODRs

per week (range, 2-3) with a relatively level trend and slight variability In the CICO intervention phase, a gradual decrease in level and trend was observed as compared with the baseline phase Lauren averaged 1 ODR per

100

80

60

40

20

100

80

60

O 40

X 20

ta

Baseline

60

40

20

0

100

80

60

40

20

0

1 2 3 6 9

CICO Lauren

12 15 18 21 24 27 30 33

1 2 3 6 9 12 15 IE

1 2 3 6

Andrew?

I I I—I—I I I—I—I I—I I I I i_j

21 24 27 30 33 36 39 42

Pam

12 15 18 21 24 27 30 33

-•—>—I—I—I—I—I I I—I—i—I I I iXi t — 1 I I • • • •

Stanley

1 2 3 6 9 12 15 18 21 24 27 30 33 36 39 42

SESSIONS

Figure 1 Percentage oí intervals with problem behavior for Lauren, Andrew, Pam, and Stanley.

week (range, 0-3) The PND for Lauren was

75%, suggesting that the CICO intervention

was effective at decreasing the student's

weekly ODRs

Andrew At baseline, Andrew had 4.3

ODRs per week (range, 4-5) with relatively

no trend and slight variability In the CICO

intervention phase, an immediate decrease in

level and trend was observed as compared

with the baseline phase Andrew averaged 1 ODR per week (range, 0-3) The PND for Andrew was 100%, suggesting that the CICO intervention was effective at decreasing the student's weekly ODRs

Pam At baseline, Pam had 1.7 ODRs per

week (range, 1-2) with an increasing trend and slight variability For the CICO intervention phase, a gradual decrease in level and trend

9

8

7

6

5

-1

3

2

1

O

10

9

8

7

6

5

4

3

2

1

O

10

9

8

7

6

5

4

3

2

1

Û

Lauren

10 11

Pam

Stanley

1 2 3 4 5 6 7 8 9 10 11 12 13 14

•WEEKS

Figure 2 Weekly office discipline referrals (ODRs) for Lauren, Andrew, Pam, and Stanley.

was observed Pam averaged 1 ODR per week

(range, 0-3) The PND for Pam was 63%,

suggesting that the CICO intervention was

somewhat effective at decreasing the student's

weekly ODRs

Stanley At baseline, Stanley had 5.0

ODRs per week (range, 4-6) with relatively

no trend and slight variability For the CICO intervention phase, a gradual decrease in level and trend was observed Stanley averaged 2.3 ODRs per week (range, 0-5) The PND for Stanley was 75%, suggesting that the CICO intervention was effective at decreasing the student's weekly ODRs

Percentage of DPR Points Earned

Lauren During baseline, Lauren earned an

average of 48% (range, 44%-51%) of her

possible DPR points Under CICO, Lauren

earned an average of 79% (range, 66%-96%)

of her possible weekly DPR points The PND

for Lauren was 100%, suggesting that CICO

was effective at increasing the percentage of

daily DPR points earned

Andrew During baseline, Andrew earned

an average of 37% (range, 32%-41%) of his

possible DPR points Under CICO, Andrew

earned an average of 82% (range, 69%-91%)

of his possible weekly DPR points The PND

for Andrew was 100%, suggesting that CICO

was effective at increasing the percentage of

daily DPR points earned

Pam During baseline, Pam earned an

average of 56% (range, 53%-59%) of her

possi-ble DPR points Under CICO, Pam earned an

average of 75% (range, 58%-88%) of her possible

weekly DPR points The PND for Pam was 92%,

suggesting that CICO was effective at increasing

the percentage of daily DPR points earned

Stanley During baseline, Stanley earned

an average of 44% (range, 41%-47%) of his

possible DPR points Under CICO, Stanley

earned an average of 78% (range, 47%-98%)

of his possible weekly DPR points The PND

for Stanley was 92%, suggesting that CICO

was effective at increasing the percentage of

daily DPR points earned

Academic Progress Monitoring

Figure 3 summarizes the DCPM and EPM

obtained for each student on multiple skill

grade-level probes The data were visually

analyzed separately with regard to changes in

level, trend, and variability Students' actual

rate of improvement was also visually

ana-lyzed on grade-level multiple skill probes

Lauren According to the pretreatment

assessment, Lauren was performing at the lower

end of the instructional range for her grade level,

with a median score of 14 DCPM Visual

inspection revealed a level trend with a slight

degreeof variability with a mean of 14.7 DCPM

(range, 14-15) across the baseline phase For the

CICO intervention phase, a gradual increase in

level, trend, and variability was observed with a

mean 16.3 DCPM (range, 14-20) Visual

analysis of errors suggests a gradually

decreas-ing trend and slight variability with a mean of

1.3 EPM (range, 1-2) across baseline Errors in

the CICO intervention phase were observed to decrease in level, trend, and variability with a mean of 0.38 EPM (range, 0-1 )

Andrew According to the pretreatment

assessment using Curriculum Based Measure-ment (CBM), Andrew was performing at the appropriate instructional level for his grade, with

a median score of 15.5 DCPM Visual inspection revealed a gradually increasing trend with some degree of variability with a mean of 15.3 DCPM (range, 14-16) across the baseline phase With regard to the CICO intervention phase, a gradual increase in level, trend, and variability was observed with a mean 17.0 DCPM (range, 16-18) across the phase Visual analysis of errors suggests a relatively level trend and slight variability with a mean of 1.3 EPM (range, 0-2) across baseline Errors in the CICO intervention phase were observed to decrease in level, trend, and variability with a mean of 0.38 EPM (range, 0-2) across the phase

Pam According to the pretreatment

as-sessment using CBM, Pam was performing at the lower end of the instructional range for her grade level, with a median score of 14 DCPM Visual inspection revealed a gradually increas-ing trend with some degree of variability with

a mean of 13.7 DCPM (range, 13-14) across the baseline phase With regard to the CICO intervention phase, a gradual increase in level, trend, and variability was observed with a mean of 14.1 DCPM (range, 11-17) across the phase It may be interesting to note that Pam achieved her lowest performance (11 DCPM) during the second week of the CICO interven-tion This low score coincided with her highest percentage of intervals with observed problem behavior (25%)

Visual analysis of errors suggests a de-creasing trend with slight variability with a mean of 1.7 EPM (range, 1-2) across baseline Errors in the CICO intervention phase were observed to gradually decrease in level and trend with moderate variability and a mean of 0.38 EPM (range, 0-1) across baseline

Stanley According to the pretreatment

assessment using CBM, Stanley was performing

at the appropriate instructional level for his grade, with a median score of 21 DCPM Visual inspection revealed a gradually increasing trend with some degree of variability with a mean of 20.8 DCPM (range, 20-22) across the baseline phase With regard to the CICO intervention phase, a gradual increase in level, trend, and variability was observed with a mean 22.3 DCPM (range, 20-24) across the phase Visual analysis of