Minnesota State University, Mankato Minnesota State University, Mankato
Cornerstone: A Collection of Scholarly Cornerstone: A Collection of Scholarly
and Creative Works for Minnesota and Creative Works for Minnesota
State University, Mankato State University, Mankato
All Graduate Theses, Dissertations, and Other
Capstone Projects
Graduate Theses, Dissertations, and Other
Capstone Projects
2014
A Comparison of the Effectiveness of a Token Economy System, a A Comparison of the Effectiveness of a Token Economy System, a
Response Cost Condition, and a Combination Condition in Response Cost Condition, and a Combination Condition in
Reducing Problem Behaviors and Increasing Student Academic Reducing Problem Behaviors and Increasing Student Academic
Engagement and Performance in Two First Grade Classrooms Engagement and Performance in Two First Grade Classrooms
Britta Leigh Fiksdal
Minnesota State University - Mankato
Follow this and additional works at: https://cornerstone.lib.mnsu.edu/etds
Part of the School Psychology Commons
Recommended Citation Recommended Citation
Fiksdal, B. L. (2014). A Comparison of the Effectiveness of a Token Economy System, a Response Cost
Condition, and a Combination Condition in Reducing Problem Behaviors and Increasing Student
Academic Engagement and Performance in Two First Grade Classrooms [Master’s thesis, Minnesota
State University, Mankato]. Cornerstone: A Collection of Scholarly and Creative Works for Minnesota State
University, Mankato. https://cornerstone.lib.mnsu.edu/etds/343/
This Dissertation is brought to you for free and open access by the Graduate Theses, Dissertations, and Other
Capstone Projects at Cornerstone: A Collection of Scholarly and Creative Works for Minnesota State University,
Mankato. It has been accepted for inclusion in All Graduate Theses, Dissertations, and Other Capstone Projects by
an authorized administrator of Cornerstone: A Collection of Scholarly and Creative Works for Minnesota State
University, Mankato.
A Comparison of the Effectiveness of a Token Economy System, a Response
Cost Condition, and a Combination Condition in Reducing Problem Behaviors and
Increasing Student Academic Engagement and Performance in Two First Grade
Classrooms.
Britta L. Fiksdal
A Dissertation Submitted in Partial Fulfillment of the
Requirements for the Degree of
Doctor of Psychology
School Psychology
Minnesota State University, Mankato
Mankato, Minnesota
May 2014
ii
A Comparison of the Effectiveness of a Token Economy System, a Response Cost
Condition, and a Combination Condition in Reducing Problem Behaviors and Increasing
Student Academic Engagement and Performance in Two First Grade Classrooms.
Britta L. Fiksdal
This dissertation has been examined and approved by the following members of the
dissertation committee.
_____________________________________
Daniel Houlihan, Ph.D., Advisor
_____________________________________
Kathy Bertsch, Ph.D., Committee Member
_____________________________________
Kevin J. Filter, Ph.D., Committee Member
_____________________________________
Teresa Wallace, Ph.D., Committee Member
__________________
Date
iii
Copyright
Copyright © Britta L. Fiksdal 2014
All Rights Reserved
iv
Dedication
This dissertation is dedicated to my incredibly supportive and understanding fiancé,
Brett, and my family who have been there for me no matter what throughout my doctoral
training. They have provided me with both emotional and financial stability to complete
this journey and shown me that I can make my dreams come true. Their confidence in me
is priceless and has meant the world to me.
I also want to thank my advisor, Dr. Houlihan, for his endless support throughout my
training. Several hours have been dedicated to this project and it could not have been
completed without his expertise and guidance. I have enjoyed working with him and am
honored to have had him as my mentor.
v
Table of Contents
Copyright…………………………………………………………………………………iii
Dedication………………………………………………………………………………...iv
List of Figures………………………………………………………………………...….vii
Abstract of the Dissertation……………………………………………………………....ix
Chapter I Introduction and Literature Review…………………………………………….1
Token Economies………………………………………………………………….1
Response Cost……………………………………………………………………..6
Combination of Token Economies and Response Cost……………………….......8
Florida Pilot Study…………………………………………………………….....15
Purpose of the Current Study………………………………………………….....17
Chapter II Methods………………………………………………………………………19
Participants ………………………………………………………………………19
Dependent Measures……………………………………………………………..20
Independent Measures…………………………………………………………...21
Procedure………………………………………………………………………...22
Pre-Baseline…………………………………………………………………..22
Baseline…………………………………………………………………….....22
Token Economy………………………………………………………………22
Response Cost………………………………………………………………...23
Combination…………………………………………………………………..24
Token Exchange………………………………………………………………24
Data Analysis…………………………………………………………………25
vi
Inter-Rater Reliability………………………………………………………...26
Treatment Integrity…………………………………………………………...26
Chapter III Results…………………………………………………………………….....27
Academic Engagement…………………………………………………………..27
Disruptive Behavior……………………………………………………………...32
Academic Performance...………………………………………………………...37
Student Preference…………………………………………………………….....42
Teacher Preference…………………………………………………………….....43
Treatment Fidelity………………………………………………………………..47
Inter-Rater Reliability……………………………………………………………47
Chapter IV Discussion…………………………………………………………………...48
Appendix A…………………………………………………………………………………
Consent and Assent Forms…………………………………………………….....53
Appendix B…………………………………………………………………………………
Observation Forms and Teacher Materials………………………………………71
Appendix C…………………………………………………………………………………
Data Collection Forms…………………………………………………………...79
vii
List of Figures
Figure 1: Figure 1. Percent of intervals students were academically engaged in C.S.’
classroom during the Token Economy, Response Cost, Combination, and Baseline
conditions across the three phases.
Figure 2: The average percentage of intervals students were academically engaged by
condition across the three phases for C.S.’ classroom.
Figure 3: Percent of intervals students were academically engaged in S.M.’s classroom
during the Token Economy, Response Cost, Combination, and Baseline conditions across
the three phases.
Figure 4: The average percentage of intervals students were academically engaged by
condition across the three phases for S.M.’s classroom.
Figure 5: Rate of disruptive student behaviors per minute during the Token Economy,
Response Cost, Combination, and Baseline conditions for C.S.’ classroom across the
three phases.
Figure 6: Average rate of disruptive student behaviors per minute by condition across the
three conditions for C.S.’ classroom.
Figure 7: Rate of disruptive student behaviors per minute during the Token Economy,
Response Cost, Combination, and Baseline conditions for S.M.’s classroom across the
three phases.
Figure 8: Average rate of disruptive student behaviors per minute by condition across the
three conditions for S.M.’s classroom.
Figure 9: Academic performance as measured by average class points earned on a three
point quiz given to students daily during the Token Economy, Response Cost,
Combination, and Baseline conditions across the three phases for C.S.’ classroom.
Figure 10: Average academic performance as measured by average points earned on daily
three point quiz by condition across the three phases in C.S.’ classroom.
viii
Figure 11: Academic performance as measured by average class points earned on a three
point quiz given to students daily during the Token Economy, Response Cost,
Combination, and Baseline conditions across the three phases for S.M.’s classroom.
Figure 12: Average academic performance as measured by average points earned on daily
three point quiz by condition across the three phases in S.M.’s classroom.
ix
Abstract of the Dissertation
A Comparison of the Effectiveness of a Token Economy System, a Response Cost
Condition, and a Combination Condition in Reducing Problem Behaviors and Increasing
Student Academic Engagement and Performance in Two First Grade Classrooms.
By
Britta L. Fiksdal
Doctor of Psychology in School Psychology
Graduate School of Psychology
Minnesota State University, Mankato, 2014
Carlos Panahon, Ph.D., Chair
Previous research has shown that token economy systems and response cost procedures
are effective in reducing disruptive behaviors in classrooms and increasing academic
engagement. Few studies have compared the effectiveness of combining these two
classroom management techniques, examined academic performance, and directly
observed academic engaged time. The current study compared the effectiveness of four
conditions: baseline, response cost procedure, token economy system, and a combination
condition among two, first grade classrooms in a small town in central Wisconsin using
direct observation and permanent product of a three question quiz. Behaviors assessed
included problem behaviors in the classroom, academic engaged time, academic
performance, and student and teacher preference. An alternating treatments design was
utilized in which one of the four conditions were employed each day during the math
lesson in a randomized predetermined order.
1
CHAPTER I
INTRODUCTION AND LITERATURE REVIEW
Token economies have a long history of changing behaviors among humans.
Lancaster started the trend with the use of tickets within large classrooms in the early
1800’s followed by the use of cherries and cakes in the early 1950’s to teach Latin and
Greek to children (Lancaster, 1805; Skinner, 1966). One of the first therapeutic
applications of a token economy was delivered by Avendano Carderera in 1959 who gave
a ticket to children for good behavior (Rodriguez, Montesinos, & Preciado, 2005). Staats
and colleagues applied a token economy system to a student with reading problems in the
late 1950’s. These studies indicate that token economies have been used for quite some
time to modify behavior. However, despite the research indicating its effectiveness, token
economies are not currently used as often as they could be in schools (Matson, &
Boisjoli, 2009).
Token Economies
Token economy systems are when participants earn tokens contingent on certain
behaviors which are then exchanged for predetermined backup reinforcers at a later point
in time. The main element to a token economy system is that the tokens are delivered
contingent on a specific behavior and linked to meaningful reinforcer(s) (Kazdin, 1977;
Wolery, Bailey, & Sugai, 1988). Ayllon and Azrin wrote a book in 1968 titled, The
Token Economy, which emphasized the effectiveness of using token economies with
children with developmental delay as well as various problematic behaviors. The book
also discussed the effectiveness of implementing a token economy system with typically
2
developing children. The majority of research evaluating the effectiveness of token
economies was conducted between the 1960s-1980s however; until recently the number
of studies between then and now has been minimal (Matson & Boisjoli, 2009).
A token economy system, when applied correctly, shares many of the same
features of other behavior modification interventions (Hall, 1979). They typically consist
of a list of instructions for the individuals involved, including: the target behavior(s) that
will and/or will not be reinforced, a method to ensure the token is contingent on behavior
which allows the token to become a reinforcing stimulus, and a set of rules that explain
how, when, and under what conditions the tokens can be exchanged for the backup
reinforcers (O’Leary, & Drabman, 1971). In a token economy, points or tokens are
delivered contingent on a target behavior over a specified period of time. After a certain
time interval has passed, the students can exchange the number of tokens they have for
backup reinforcers. The size of the backup reinforcer should be in relation to the number
of tokens the individual earned and are exchanging. There are many advantages to token
economy systems such as; bridging the gap between a target response and the backup
reinforcer, maintaining performance over an extended period of time until the backup
reinforcer can be delivered, and allowing behavior to be reinforced at any time. Token
economies are also less likely to be affected by satiation and can provide a visual
reminder of the progress or lack of progress the student has made regarding their
behavior (Kazdin, & Bootzin, 1972).
Within a token economy, the token is a stimulus that signals the delivery of a
backup reinforcer at a later point in time. The token can be any object that can be easily
3
delivered, easily kept, and easily exchanged. Some advantages of using a tangible
reinforcer include: tokens are portable, no maximum exists, the number of tokens can
represent the amount of reinforcement, they are durable and can be used continuously,
devices can be used to automatically deliver tokens contingent on behavior, the physical
characteristics of the token can be standardized or personalized, and can be made to be
indestructible (Kazdin & Bootzin, 1972). In some cases, natural reinforcement of teacher
praise and attention will not be effective in changing classroom behavior, in these
situations, token economies are often times found to be effective. Token economies are
most effective when there are multiple backup reinforcers as opposed to one reinforcer.
By having a large variety of backup reinforcers to choose from, it is less likely students
will become satiated and the chances of each student finding at least one item that
functions as a reinforcer is increased (O’Leary, & Drabman, 1971).
The first step in a token economy is to identify a target behavior and developing
an operational definition. After the target behavior is identified, tokens must be made,
backup reinforcers need to be gathered, and rules regarding delivery and exchanging of
tokens must be developed. Next, the tokens must be established as secondary reinforcers
for the backup reinforcers. When establishing the tokens as reinforcing, it is important to
go through a practice with students in which they are told and shown how to earn tokens,
the rules for exchanging, and then allow them to exchange the tokens for a reinforcer so
they have access to the contingency (Kazdin, & Bootzin, 1972).
Research studies have found that, in part because of the flexibility of the different
features, token economies have been effective in reducing problem behavior and
4
increasing positive behavior in a variety of subjects under a variety of different
conditions and with multiple behaviors (Kazdin, 1982; O’Leary, & Drabman, 1971). A
review of the literature conducted by Matson and Boisjoli (2009), found that token
economies have been used successfully for different behaviors such as remaining in seat,
increasing attention, increasing appropriate verbalizations and social skills, and
increasing self-help skills, decreasing inappropriate call-outs in class, decreasing
aggressive behaviors, decreasing disruptive behaviors within class, increasing academic
behaviors such as completing homework assignments, increasing test performance,
increasing academic engaged time and academic performance, and increasing academic
accuracy. Token economies have been successful for multiple subjects as well including;
children with developmental delay, cognitive deficits, autism, ADHD, emotional and
behavioral problems, conduct disorders and typically developing children. They have also
been used with adults with psychiatric diagnosis, legal offenders, employees, and
teachers. Token economies have been administered by multiple individuals such as
parents, teachers, school psychologists, employers, doctors, nurses, and clinical
psychologists.
While there are a multitude of studies published proving the effectiveness of
token economy systems at reducing problem behaviors and increasing positive behaviors
in classroom and school settings, systematic evaluations of the experimental literature to
validate the use in schools have not been completed recently. Maggin and colleagues
conducted such a literature review to evaluate the quality of research designs used to
determine whether or not token economy systems are, in fact, an evidence based
5
intervention for behavior management in both classroom and school settings (Maggin,
Chafouleas, Goddard, & Johnson, 2011). The study used four questions to guide their
review of the literature which included looking at What Works Clearinghouse (WWC)
standards, student characteristics and intervention features, statistical summaries of
treatment effects, and methodological strengths and weaknesses. After their initial search,
they started with a total of 834 articles to be screened for retrieval, 118 articles made it
past the initial inclusion screening, 36 articles were considered potentially relevant, and
24 studies were included in the synthesis. The reasons for exclusions included: ineligible
intervention, ineligible dependent variable, ineligible population, ineligible design,
irretrievable data, and not an intervention study. Of the 24 studies, there were a total of
90 cases of which 67 of them used students as the unit of analysis and 23 used the
classroom as the analysis. Additionally, of the 25 studies, only four different single
subject designs were used: reversal (n = 15), AB (n = 5), multiple baseline (n = 3), and
ABA (n = 1). Overall, according to WWC standards, there is currently insufficient
support for token economies as an evidence based classroom management strategy
mainly due to methodological rigor of the current studies. However, when you take into
account all of the studies, token economy systems were found to be effective as both a
classroom management system and individual behavior intervention program for three of
the four effect sizes calculated through significant improvement in student functioning as
a result of introducing token economy systems in the classroom.
6
Response Cost
Response cost is a punishment procedure that has been used in school settings to
effectively change behavior (McGoey, & DuPaul, 2000). Originally, response cost
referred to changing the work required to emit a behavior, in other words changing the
cost of the behavior to affect the rate of that behavior (Weiner, 1962). The use of
response cost in the school settings is somewhat different from the original definition. In
a school setting, the incorporation of a response cost includes taking away tokens or
points contingent on problem behaviors. Typically these tokens are given
noncontingently at the beginning of a time interval, lesson, or session and the student gets
to keep them as long as they do not engage in any of the problem behaviors. At the end of
the time period, they can exchange the tokens they have left for backup reinforcers.
Response cost is a punishment based system whereas a token economy is reinforcement
based (Kazdin, 1972; Pace, & Forman, 1982).
Response cost is not the same as extinction or time-out. Extinction is the
withdrawal of reinforcers maintaining an undesirable behavior and time-out is removing
the student from a reinforcing environment contingent on undesirable behavior. Response
cost is not extinction because you are not removing the functional reinforcer for problem
behavior, instead you are withdrawing secondary reinforcers contingent on undesirable
behavior. Response cost is not time-out because you are not removing the student from
the environment. Instead, they stay in the classroom and lose a token contingent on
problem behavior. Response cost is different from a token economy in that you do not
receive tokens contingent on desirable behavior, instead, the tokens are given
7
noncontingently in the beginning of a time interval and are taken away contingent on
problem behavior. Whatever amount of tokens the student has left is then exchanged for
backup reinforcers at a specified time (Kazdin, 1972).
A review of the literature shows response cost procedures have been found
effective for multiple individuals within classroom settings such as developmental delay,
cognitive delay, emotional and behavioral problems, and students with academic
difficulties. A variety of behaviors have also been changed drastically with response cost
procedures including out of seat behavior, calling out in class, off task behavior,
disruptive behavior, academic performance, smoking, and weight loss. Most studies that
have examined the recovery of suppressed behaviors through response cost have found
that the behavior does not recover when the contingency is withdrawn (Kazdin, 1972).
Typically, punishment procedures have been associated with side effects such as
escape and avoidance behaviors along with emotional consequences. Previous studies
have indicated that escape behaviors are not associated with response cost like it is with
the delivery of aversive stimuli. Also, there have been no negative emotional
consequences reported with response cost procedures (Litenberg, 1965). It is likely that
these negative side effects are not associated with response cost because the removal of a
positive reinforcer (token) is not as aversive in magnitude compared to the delivery of an
aversive stimulus (Schmauk, 1970). Current research has not focused on emotional
consequences as much as reducing problem behaviors.
8
Combination of Token Economies and Response Cost
Some researchers and classroom teachers have combined token economy systems
with response cost procedures. In these classroom management techniques the individual
is able to earn points contingent on desirable behavior and can also lose points contingent
on undesirable behavior. At the end of the interval, the tokens they have left can be
exchanged for backup reinforcers (Weiner, 1962). One advantage to this combination
procedure includes the ability to provide tokens contingent on a behavior that is
completely unrelated to the behavior in which the tokens are taken away. Response cost
used to be used quite frequently within token economy systems, however, a more recent
review of the literature suggests that it is not used as often in school settings as it used to
be and neither are token economy systems (Matson, & Boisjoli, 2009).
There have been multiple research studies on the effectiveness of token
economies, response cost procedures, and combination procedures in reducing problem
behaviors among human subjects. The first token economy system to be used in a larger
classroom setting was in 1967 by O’Leary and Becker. The classroom consisted of 17
students all of whom had emotional and behavioral disturbances. After the introduction
of the token economy, disruptive behavior decreased significantly from 76% of intervals
during baseline to 10% during intervention. O’Leary and colleagues (1969) also
conducted a token economy system with seven students in a second grade classroom all
of whom exhibited disruptive behaviors. The implementation of the token economy
system reduced disruptive behaviors significantly compared to baseline rates. A token
economy system was used to reduce violent behavioral outbursts and loud noise among
9
psychiatric patients (Winkler, 1970). A review of token economy systems within
classroom behavior found that many behaviors were successfully increased such as being
quiet, hanging up coats, sitting at their desk, academically engaged, completing a task,
following instructions, and facing the front of the class and teacher (Kazdin, & Bootzin,
1972). More recently, Kahng and colleagues (2003) provided tokens contingent on eating
certain amounts of food and eating novel food for a four year old girl diagnosed with
pervasive developmental disorder with food refusal.
Siegel and colleagues (1969) used a response cost procedure to reduce speech
disfluencies among normal-speaking college ages students. Four females and one male
participated in the study at the University of Minnesota. Results indicated the procedure
was very effective at suppressing disfluencies during spontaneous speech. They used
money as the backup reinforcers for the points they earned throughout the speech. A
response cost procedure was used to reduce problematic behaviors among delinquent
soldiers (Winkler, 1970). Response cost procedures have also been used to reduce
aggressive statements, tardiness, and specific word usage among three delinquent boys
(Phillips, 1968). Phillips and colleagues (1971) studied the effectiveness of a response
cost procedure on delinquent youths in Achievement Place. Results indicated that point
loss contingent on problem behavior produced significant increases in desirable
(incompatible) behaviors such as promptness, completing quizzes, saving money, and
keeping a clean bedroom. A study conducted by Pace and Forman (1982) found that a
response cost procedure was effective for 55 second graders enrolled in a Title -1
program in a low socioeconomic status neighborhood school. Results indicated the fines
10
associated with the response cost procedure was effective in reducing disruptive
behaviors such as out of seat, inappropriate vocalization, being noisy, touching other
people’s property, and aggression.
More recently, in 2004, Conyers and colleagues compared the effectiveness of a
response cost condition with differential reinforcement of other behavior on reducing
disruptive behaviors among 25 students in a preschool classroom. Disruptive behavior
decreased from 64% of intervals to 5% for the last six sessions of the response cost
procedure. Initially, differential reinforcement of other behavior resulted in a more drastic
decrease of disruptive behavior but over time disruptive behaviors increased to 27% of
intervals. Therefore, the response cost condition maintained lower rates of disruptive
behavior more effectively than differential reinforcement of other behavior.
McLaughlin and Malaby (1972) compared the effectiveness of a token economy
system and a response cost condition with a classroom containing 25-27 fifth and sixth
grade students. In the Point Loss phase the teacher removed points contingent on problem
behaviors. In the Quiet Behavior Point phase students earned points contingent on
desirable behaviors that were incompatible with an ineffective learning environment.
Results indicated that both were effective in reducing problem behavior and increasing
desirable behavior. McGoey and DuPaul, (2000) compared the effectiveness of a token
economy system and response cost procedure in reducing inappropriate social behaviors,
off task behavior, following rules, and tantrumming among four preschool students
diagnosed with ADHD. Results showed little difference between the two interventions in
11
the ability to change behavior. Both the response cost and token economy conditions
resulted in a decrease in problematic behaviors compared to baseline rates.
While a number of studies have focused on reducing problem behaviors among
individuals, a number of studies have focused on changing academic behaviors such as
studying, staying on task, completing homework assignments, and completing tests as
well. A review of the literature shows that response cost and token economy systems
have been effective for a wide variety of school age populations such as developmental
and cognitive delay in summer school programs and hospital settings, teenage students
and elementary students, a small child with Phenylketonuria (PKU), and children with
autism and social skills deficits (Matson, & Boisjoli, 2009).
In 1965, Birnbrauer, Wolf, Kidder, and Tague found that a token economy system
resulted in higher levels of accuracy on homework and increased rates of studying overall
for 15 children diagnosed with cognitive delay compared to baseline in which no token
economy was employed. A study conducted in 1968 compared noncontingent
reinforcement to contingent reinforcement using a token economy system. The study
showed that noncontingent reinforcement was not as effective in changing and increasing
study behavior among the 12 preschool children with above average intelligence
compared to the token economy system (Bushell, Wrobel, & Michaelis, 1968). Wolf and
colleagues found that a token economy system was effective in increasing report card
grades and regular classroom assignments along with language, reading, and arithmetic
performance when a token economy system was employed for students in a remedial
education elementary classroom (Wolf, Guiles, & Hall, 1968). Walker and colleagues
12
conducted a study that assessed a token economy system on task-oriented behavior for
six children all with average or above average functioning but were described by their
teachers as being disruptive and hyperactive. Results showed the percentage of on-task
intervals increased from an average of 39% of intervals during baseline to an average of
90% of intervals during the token economy condition (Walker, Mattson, & Buckley,
1969).
Panek (1970) compared a response cost condition to a token economy system for
learning word associations among patients diagnosed with schizophrenia. Subjects
included 32 male patients between the ages of 30-77 years old who have lived in the
hospital between 2-38 years. Patients were split into two groups; response cost group or
token economy groups. Results showed an increase in word association for both groups
of patients. Therefore, this study showed no difference between reward based and
punishment based programs for learning word associations and neither condition resulted
in generalization to new words. Broden, Hall, Dunlap, and Clark (1970) first
implemented a token economy condition and then a combination response cost and token
economy condition while assessing study rates among seventh and eighth grade students
who were all behind their peers academically by at least one year. In the token economy
condition, the researchers noticed an increase in study behavior during the token
economy system in which students earned one extra minute of lunch contingent on
appropriate study behaviors. While study behavior increased from 29% of intervals
during baseline to 74% of intervals during the token economy condition, the researchers
were disappointed to see the study behaviors did not generalize to times of the day in
13
which the token economy was not implemented. Therefore, the researchers implemented
a combination condition in which students were able to earn the points contingent on
appropriate study behaviors but could also lose those points contingent on problem
behaviors. After the introduction of this condition, appropriate study behavior increased
to an average of 80% of intervals throughout the entire day indicating greater
generalization for the combination condition.
A study conducted in 1972 by Kaufman and O’Leary showed it was possible to
increase reading behavior and task engagement using both a response cost condition and
a token economy condition among 16 students living in a psychiatric hospital. Their
study indicated that while both were effective in increasing reading and task engagement
skills, there was not a significant difference between the two conditions.
Iwata and Bailey (1974) compared the effectiveness of a token economy and
response cost condition on student’s academic and social behaviors among 15 students in
a special education classroom. The students were divided into two different groups and
each group experienced both the token economy and response cost conditions throughout
the study. Results showed that the average number of problem completed by students
during the token economy and response cost conditions showed a slight increase
compared to baseline rates, accuracy remained similar throughout the entire study, and
off task behavior reduced significantly for both conditions compared to baseline. This
study also assessed whether or not students preferred one condition over the other.
Therefore, the last phase of the study involved the students being allowed to pick if they
participated in a response cost condition or a token economy condition. Results showed
14
there was no significant pattern of preference between the two conditions. In fact, four
students consistently chose the token economy system, five students consistently chose
the response cost condition, and the remaining six students switched back and forth
between the two conditions.
Other studies have assessed teacher, parent, and student preference for response
conditions and token economy systems. In general, techniques that focus on increasing
positive behaviors have been rated higher and as more acceptable compared to techniques
that have focused on reducing negative behaviors with the exception of response cost
techniques (Frentz, & Kelley, 1986). Little and Kelley (1989) assessed treatment
acceptability for five different parenting techniques: response cost, rewards for good
behavior, timeout with spanking, spanking alone, and timeout alone. Results indicated
parents rated response cost as the most acceptable and it was rated high on the Parent’s
Consumer Satisfaction Questionnaire. Reynolds and Kelley (1997) assessed treatment
acceptability of a response cost procedure using the Intervention Rating Profile – 15 and
a 6-point Likert scale. Results showed prior to treatment the teachers had rated response
cost as a favorable classroom management technique. After the teacher’s employed a
response cost procedure in their classroom, their ratings increased and teachers rated it as
a highly acceptable treatment. A study conducted in 1998 examined the treatment
acceptability among mothers who have children who exhibit disruptive and problem
behaviors. The techniques assessed included: differential attention, over-correction,
positive reinforcement, response cost, spanking, and time-out. The conclusion of the
study showed mothers rated positive reinforcement the highest followed by response cost
15
and time out. Differential attention, overcorrection, and spanking were rated the lowest
by mothers (Jones, Eyberg, Adams, & Boggs, 1998). A similar study conducted in 2007
showed similar results in which response cost, token economy, and time out were rated as
the most acceptable and overcorrection, ignoring, and differential attention were rated
lower (Pemberton, & Borrego, 2007). McGoey and DuPaul (2000) noted that teachers
found the response cost condition to be more acceptable and chose to implement it within
their classroom during the choice condition.
Florida Pilot Study
A pilot study was conducted in the spring of 2012. The school was located in a
small suburb of Orlando Florida in a low socioeconomic neighborhood. The majority of
the students were bilingual speaking English and Spanish. The majority of the parents
spoke Spanish and only two parents had received an education higher than High School.
The teacher had been referred for the study due to behavior problems in her classroom.
The teacher, Mrs. C, was a certified fourth grade teacher who had five years of teaching
experience. The subjects in the study included 22 children in a fourth grade general
education classroom. The classroom included 14 boys and 8 girls with an average age of
9 years old. Three students were not included in the study, one student was on an IEP that
included an individualized token economy system that the team did not want changed, a
second student’s parents did not provide consent for their child to participate, and a third
student entered the classroom half way through the study. Therefore, a total of 19
students participated in the study. Using methods identical to the current study, the data
showed contrasting results to the earlier studies conducted.
16
In the pilot study, data showed an increase in academic engagement during the
Combination and Token Economy conditions and lower levels of academic engagement
during baseline and Response Cost conditions. Lower rates of disruptive/problematic
behavior was observed during the Combination and Token Economy conditions
compared to the baseline and Response Cost conditions. With regards to Academic
Understanding, the highest understanding occurred during the Combination condition.
Data from the student survey shows students favorite conditions were the Token
Economy and Combination conditions, their least favorite conditions were baseline and
Response Cost, the condition that made it easiest to learn was the Combination condition,
the condition in which learning was rated as the hardest were the Response Cost and
baseline conditions, and students preferred to continue the Combination condition in the
future. Data from the teacher survey shows Ms. C preferred the Token Economy and
Combination conditions, she reported students appeared more anxious and problem
behaviors were higher during baseline and Response Cost, and she will be administering
Token Economy or the Combination strategy in the future.
Some limitations to the pilot study included the small sample size of only one
class with one teacher who was in charge of teaching the class and delivering tokens
during each condition. Therefore, the strategies were not administered as rigorously as if
multiple adults were in the classroom. Lastly, the classroom was video recorded making
observing students difficult at times if they walked out of view of the camera. Future
research is important to clarify which classroom management strategy is the most
beneficial for increasing student academic engagement and performance as well as the
17
most efficient for reducing problematic behavior in the classroom. Most of the studies
showing Response Cost as being just as effective if not more effective were conducted in
the 1970s and 1980s. The demographics of our students have changed since then,
therefore creating a need to validate current classroom management strategies commonly
used among teachers.
Purpose of the Current Study
The purpose of the present study is to update the literature as well as compare the
effectiveness of three classroom management strategies; response cost, token economy,
and combination of response cost and token economy. The study will assess the ability
of the three conditions to reduce problem behaviors, increase academic engaged time,
increase academic performance, and identify teacher and student preference.
Additionally, the study was designed to enhance the research showing support for token
economy systems as an empirically based behavior management intervention for both
individuals and classrooms in the school setting and to meet the methodological features
of single-case design studies set for by Kratochwil and colleagues (2002; 2010). These
five features include: 1. Operational definitions if all variables and settings, 2.
Replication of effects, 3. Collection of treatment integrity data, 4. Collection of
interobserver agreement/reliability data, and 5. Collection of social validity data. It is
hypothesized that both the Token Economy and Combination conditions will result in
higher academic engagement rates and lower problem behavior rates compared to
Response Cost and Baseline conditions. Additionally, it is hypothesized that students will
have higher academic performance in the Token Economy and Combination conditions
18
compared to the Response Cost and Baseline conditions. Furthermore, it is hypothesized
that both students and teachers will report preferring Token Economy and Combination
conditions over Response Cost and Baseline conditions.
19
CHAPTER II
METHODS
Participants
The subjects in the current study included two first grade classrooms in an
elementary school located in a small town in West-Central Wisconsin. The teachers were
referred for the study due to behavior problems in the classroom that required the teacher
to stop their instruction or class activity at least three times on average per lesson. In the
first classroom, the teacher was a veteran with more than 20 years experience at the
elementary level (teacher C.S.). This classroom had 16 students total during Math class
with one student receiving special education support during instruction for emotional and
behavioral needs. In the second classroom, the teacher was a newer teacher with less than
three years experience at the elementary level (teacher S.M.). This classroom had 14
students total during Math class with one student receiving special education support as
needed for emotional and behavioral needs and two students receiving academic support
through English as a Second Language instruction as needed during independent
seatwork.
Students in C.S. classroom were between 80-95 months old with an average age
of 86.3 months old. Ninety-four percent of the students were Caucasian and 1% of the
students were Hispanic. Students in S.M. classroom were between 80-92 months old with
an average age of 85.71 months old. Sixty-nine percent of the students were Caucasian,
30% were African American, and 1% was Asian.
20
Dependent Measures
Data on problem behaviors were collected via frequency counts from directly
observing the classroom each day. Problem behavior was defined as exhibiting any
behaviors or audible vocalizations that were disruptive, interfered with learning, or
impeded instructional delivery. Specific examples include fidgeting, drawing on self,
talking out, and disruptive interaction with peers that interfered with learning, leaving the
assigned instructional area, and making audible vocalizations not related to the
instructional task such as singing, humming, or talking back.
Data on academic engaged time was collected via momentary time sampling with
15 second intervals from directly observing the classroom each day. Observers recorded
each student in a systematic order for a total of 25 minutes. Academic engagement was
defined as the student looking at materials, raising hand, working on tasks that the teacher
specified, and/or engaged in communication with peers or teacher that is relevant to the
task at hand.
Data on academic performance was collected through analyzing the permanent
product of a three question quiz each student completed at the end of each math lesson.
The quiz included either multiple choice or true/false questions covering the material
from the current lesson and was developed by the teacher prior to the start of the lesson.
Student and teacher preference was assessed at the end of the study by asking
both the teachers and the students questions about the different conditions. Students were
individually interviewed and asked which was their favorite condition and why, least
favorite condition and why, and which condition they would like their teacher to
21
implement next week. Teachers were sent an email with questions asking them which
procedure they liked administering best and why, which procedure they liked
administering least and why, if they noticed their students behaving better or
academically engaged more during any of the conditions, if they noticed their students
misbehaving more or academically engaged less during any of the conditions, what they
liked about the different strategies, what they did not like about the different strategies,
what were some advantages to the different strategies you used, what were some of the
disadvantages to the different strategies you used, if you could make any changes what
would they be, which one would you be most likely to do in the future and why, which
one would you be least likely to do in the future and why.
Independent Measures
The independent measures of the current study were the different classroom
management strategies employed by the teacher. An alternating treatments design was
used throughout two phases. During the first phase, the teacher alternated between all
four conditions; the baseline condition, response cost condition, token economy
condition, and combination condition each day throughout the week for four weeks. The
order was randomly assigned to control for any history and sequence effects. During the
second phase, the teacher alternated between the two conditions found to be the most
effective at reducing problem behaviors and increasing academic engagement in their
classroom during the first phase. The order of the two conditions was randomly assigned
to control for any history and sequence effects. The study concluded with each teacher
22
employing the classroom management strategy that was the most effective at reducing
problem behaviors and increasing academic performance in their classroom.
Procedure
Pre-Baseline. Prior to data collection, the researcher gained IRB approval for the
study and located two teachers interested in participating in the study. The researcher and
teachers discussed what they considered to be academically engaged as well as
operationally defined the problem behaviors they had witnessed in their classrooms.
Along with IRB approval and teacher approval, the researcher also obtained consent from
the superintendent, school principal, and each student’s parents.
Baseline. During baseline, the teacher started the lesson by giving the students the
following instructions, “During today’s math lesson, you will not be given any tokens nor
will you be able to lose any tokens. I still want you all to be on your best behavior.” The
teacher then taught the lesson as normal, without delivering any type of tangible
reinforcement contingent on behavior. At the end of the lesson, the teacher transitioned
the kids to the next activity since there were no tokens for students to exchange.
Token Economy. During this condition, the teacher started the lesson by giving
the students the following instructions, “During today’s math lesson, you will have the
opportunity to earn a token for “good” behavior. When you earn a token, I will place it
in your name slot on the wall or name card located at your desk depending on where we
are in the classroom. Your tokens cannot be taken away right now, you can only earn
them for good behavior. At the end of the lesson you can exchange your tokens for either
a hand stamp or piece of candy from the reward box.” The teacher then started the lesson
23
as normal and delivered tokens to students contingent upon desirable behavior. When
delivering a token, the teacher briefly stated what behavior the student was earning the
token for (e.g., “I like the way you are reading quietly in your seat.”). The teacher
continued to deliver tokens throughout the math lesson for the day. At the end of the
lesson, the teacher allowed students to exchange their tokens. The magnitude and size of
the reinforcer was determined by the number of tokens the student had earned to
exchange.
Response Cost. During this condition, the teacher started out the lesson by giving
the students the following instructions, “During today’s math lesson, each of you will be
given five tokens in your name slot on the wall or name card located on your desk
depending on where we are in the classroom. Each time you misbehave, I will come and
take a token away. You cannot earn tokens back today; you can only keep them if you do
not engage in any problem behaviors and follow classroom expectations and rules. At the
end of the lesson you can exchange whatever tokens you have left for hand stamps or
candy in the reward box.” The teacher then gave each student five tokens and started the
lesson as normal. Throughout the lesson, anytime a student engaged in
problem/disruptive behavior (as identified in the problem behavior definition list) the
teacher went over to the student and quietly took away a token from their name slot or
card and told the student why the token was being taken away (e.g. “I do not like the way
you are twirling your book, instead you should be reading chapter 4.”). The teacher
continued to take away tokens throughout the math lesson contingent on problem
behaviors. At the end of the lesson, the teacher allowed students to exchange whatever
24
tokens they had left. The magnitude and size of the reinforcer was determined by the
number of tokens the student had left to exchange.
Combination Condition. During this condition, the teacher started out the lesson
by giving the students the following instructions, “During today’s math lesson, you will
have the opportunity to earn tokens for “good” behavior. When you earn a token, I will
place it in your name slot on the wall or name card located on your desk depending
where we are in the classroom. Your tokens can be taken away if you engage in any
problem behaviors. So throughout math today, you can earn tokens for good behavior
AND you can get your tokens taken away for bad behavior. At the end of the lesson you
can exchange however many tokens you have for hand stamps and/or pieces of candy in
the reward box.” The teacher then started the lesson as normal and delivered a to students
contingent on good behavior with a brief, quiet description of what behavior the student
was earning the token for (e.g. “I like the way you are reading quietly in your seat.”) and
took away tokens contingent on inappropriate behavior with a brief, quiet description of
what behavior the student was getting the token taken away for (e.g., “I do not like the
way you are singing and looking around instead of reading your book.”). The teacher
continued to deliver and take away tokens throughout the math lesson. At the end of the
math lesson, the teacher allowed students to exchange their tokens. The magnitude and
size of the reinforcer were determined by the number of tokens the student had left to
exchange.
Token Exchange. The tokens students earned were exchanged at the end of each
math lesson. Students were not able to keep the tokens or save them across sessions to
25
control for the effects of saving tokens for larger reinforcers at a later time. Students who
had 5 or more tokens at the end of math class exchanged them for three items, students
who had 3 or 4 tokens exchanged them for 2 items, students who had 1 or 2 tokens
exchanged them for 1 item, and students who did not have any tokens at the end of the
math lesson were unable to receive any items. Items consisted of a hand stamp or a piece
of candy from the reward box. Students had the option of two different hand stamps that
were switched out on a weekly basis. Adults throughout the school are aware that
students earn stamps on their hand for positive/good behavior and often ask students
about the stamps and provide positive social praise. Additionally, students are
encouraged to go home and tell their parents what they did to earn a hand stamp. Students
had the option of many different pieces of candy from small suckers to soft pieces of
candy such as Starbursts to hard pieces of candy such as Jolly Ranchers. Students were
not able to eat the candy in school, they had to put the candy in their mailbox to go home
with them at the end of the day. If a student had 5 or more tokens at the end of math
class, they could pick three items; therefore, they could pick three hand stamps, or three
pieces of candy, or two hand stamps and one pieces of candy, or two pieces of candy and
one hand stamp. Choice of backup reinforcers was a focus in the study to reduce satiation
and maintain the reinforcing efficacy of the tokens.
Data Analysis. The data collected for this study was visually analyzed on a
weekly basis to check for student performance and effectiveness of the different
management strategies. Visual analysis was also used to determine which two strategies
were most effective at reducing problem behaviors and increasing academic engagement
26
in each classroom for implementation during phase two and which was the most effective
for implementation during the final phase.
Inter-Rater Reliability. Inter-rater reliability was collected and analyzed on 33%
of sessions. Inter-rater reliability was collected on problem behavior occurrences in the
classroom, academic engagement among students in the classroom, academic
performance scoring of the three question quiz, and treatment fidelity.
Treatment Integrity. Treatment integrity was assessed by recording the number
of steps in each classroom management strategy the teacher successfully carried out such
as correctly reading the instructions to the class at the beginning of each math lesson
dependent on the condition for the day, delivering/removing tokens contingent on student
behavior dependent on the condition for the day, allowing students to exchange tokens at
the appropriate rate at the end of each math lesson, and ensuring each student took a
teacher developed 3 question objective quiz at the end of each math lesson. The number
of steps the teacher missed was subtracted from the number of total steps that should
have been completed and divided to determine the percentage of steps accurately
completed. This was conducted for 100% of the math lessons and analyzed daily. If the
teacher reached the minimum criterion of 100% of steps successfully completed she was
given positive verbal attention and praise, if the teacher missed one or more of the steps
she was sent an email with the steps she missed and a meeting would have been
scheduled to further discuss the steps missed and conduct a booster training session.
27
CHAPTER III
RESULTS
Academic Engagement
Results from the study show both Token Economy and Combination conditions
were more effective at increasing student engagement during math for both classrooms.
As can be seen from figure 1, student academic engagement during Math lessons for
classroom C.S. was higher during both the Token Economy and Combination conditions
compared to the Response Cost and Baseline conditions during Phase 1. Students were
academically engaged between 75% and 80% of the time with an average of 77% of the
time academically engaged during baseline, 94% - 95% of the time with an average of
95% of the time academically engaged during Token Economy, 82% - 85% of the time
with an average of 85% of the time academically engaged during Response Cost, and
90% - 94% of the time with an average of 93% of the time academically engaged during
the Combination condition throughout Phase 1 (see figure 2). Results from the second
phase of the study showed Token Economy was slightly better at increasing student
engagement during math compared to the Combination condition for C.S. classroom.
Students were academically engaged between 94% - 97% of the time with an average of
95% of the time during Token Economy compared to 90% - 92% of the time with an
average of 91% of the time academically engaged during the Combination conditions.
The final phase of the study consisted of only the Token Economy condition in which
students were academically engaged between 95% - 98% of the time with an average of
96% of the time (see figures 1 & 2).
28
Figure 1. Percent of intervals students were academically engaged in C.S.’ classroom
during the Token Economy, Response Cost, Combination, and Baseline conditions across
the three phases.
.
0
20
40
60
80
100
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29
Percentage of Intervals Academically
Engaged
Session
Academic Engagement - Classroom CS
Baseline
Token Economy
Response Cost
Combination
Phase
Phase Two
Phase Three
0
20
40
60
80
100
Phase 1 Phase 2 Phase 3
Percentage of Intervals Academically
Engaged
Average Academic Engagement by
Condition and Phase - C.S.
Baseline
Token Economy
Response Cost
Combination
29
Figure 2. The average percentage of intervals students were academically engaged by
condition across the three phases for C.S.’ classroom.
As can be seen from figure 3, student academic engagement during Math lessons
for classroom S.M. was higher during both Token Economy and Combination conditions
compared to the Response Cost and Baseline conditions during phase 1. Students were
academically engaged between 73% - 80% during Baseline with an average of 76%
academically engaged, between 89% - 94% during Token Economy with an average of
93% academically engaged, between 79% - 84% during Response Cost with an average
of 81% academically engaged, and between 90% - 94% during the Combination
condition with an average of 91% academically engaged during phase 1 (see figures 1 &
2). Results from the second phase of the study showed Token Economy was slightly
better at increasing student engagement during math compared to the Combination
condition for S.M. classroom. Students were academically engaged between 94% - 96%
of the time during Token Economy with an average of 95% of the time and between 88%
- 91% of the time during the Combination condition with an average of 90% of the time
academically engaged. The final phase of the study consisted of only the Token Economy
condition in which students were academically engaged between 93% - 97% of the time
with an average of 96% of the time (see figures 1 & 2).
30
Figure 3. Percent of intervals students were academically engaged in S.M.’s classroom
during the Token Economy, Response Cost, Combination, and Baseline conditions across
the three phases.
Figure 4. The average percentage of intervals students were academically engaged by
condition across the three phases for S.M.’s classroom.
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Percentage of Intervals Academically Engaged
Session
Academic Engagement - Classroom SM
Baseline
Token Economy
Response Cost
Combination
Phase One
Phase Two
Phase Three
0
20
40
60
80
100
Phase 1 Phase 2 Phase 3
Percentage of Intervals Academically
Engaged
Average Academic Engagement by
Condition and Phase - S.M.
Baseline
Token Economy
Response Cost
Combination
31
Table 1
Percentage of Non-Overlapping Data Points Among Conditions for Academic
Engagement by Classroom
Academic
Engagement
Condition(s)
Compared to:
Effect Size
Descriptor
C.S.
Token Economy,
Combination,
Response Cost
Baseline
100%
Highly
Effective
Combination, Token
Economy
Response Cost
100%
Highly
Effective
Response Cost
Baseline
100%
Highly
Effective
Token Economy Combination
Phase 1 =
25%
Phase 2 =
100%
Phase 1 =
Ineffective
Phase 2 =
Highly
Effective
S.M.
Token Economy,
Combination,
Response Cost
Baseline
100%
Highly
Effective
Combination, Token
Economy
Response Cost
100%
Highly
Effective
Response Cost
Baseline
100%
Highly
Effective
Token Economy
Combination
Phase 1 =
0%
Phase 2 =
100%
Phase 1 =
Ineffective
Phase 2 =
Highly
Effective
Table 1 shows that all three conditions, Token Economy, Combination, and Response
Cost were highly effective compared to the Baseline condition at increasing academic
32
engagement in both classrooms. Additionally, Token Economy and Combination
conditions were Highly Effective compared to the Response Cost condition at increasing
academic engagement in both classrooms. Lastly, Token Economy was not more
effective compared to the Combination condition at increasing academic engagement
during the first phase for either classroom, however, during the second phase, Token
Economy was highly effective at increasing academic engagement compared to the
Combination condition.
Disruptive Behavior
Results from the study show both Token Economy and Combination conditions
were more effective at decreasing the rate of disruptive behaviors during math for both
classrooms compared to the Response Cost and Baseline conditions during phase 1. As
can be seen from figures 5 and 6, the rate of disruptive behavior for C.S. classroom
ranged from .92 – 1.2 behaviors per minute during the baseline condition with an average
of 1.04 behaviors per minute. During the Token Economy condition, disruptive behavior
ranged from .16 - .32 behaviors per minute with an average of .22 behaviors per minute.
During the Response Cost condition, disruptive behavior ranged from .52 - .88 behaviors
per minute with an average of .72 behaviors per minute. During the Combination
condition, disruptive behavior ranged from .12 - .4 behaviors per minute with an average
of .15 behaviors per minute. Results from the second phase of the study show that Token
Economy was slightly better at decreasing the rate of disruptive behavior for C.S.
classroom compared to the Combination condition. According to figures 5 and 6,
disruptive behavior ranged from .12 - .2 behaviors per minute during Token Economy
33
with an average of .15 behaviors per minute compared to .24 - .32 disruptive behaviors
per minute during Combination with an average of .28 behaviors per minute. The final
phase of the study consisted of the Token Economy condition for C.S. classroom.
Disruptive behaviors ranged from .12 - .13 behaviors per minute with an average of .13
behaviors per minute.
Figure 5. Rate of disruptive student behaviors per minute during the Token Economy,
Response Cost, Combination, and Baseline conditions for C.S.’ classroom across the
three phases.
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Disruptive Behaviors per Minute
Session
Disruptive Behaviors
- Classroom CS
Baseline
Token Economy
Response Cost
Combination
Phase One
Phase Two
Phase Three
34
Figure 6. Average rate of disruptive student behaviors per minute by condition across the
three conditions for C.S.’ classroom.
As can be seen from figures 7 and 8, the rate of disruptive behavior for S.M.
classroom ranged from 1 – 1.24 behaviors per minute during baseline with an average of
1.12 behaviors per minute during phase 1. During the Token Economy condition,
disruptive behavior ranged from .16 - .28 behaviors per minute with an average of .21
behaviors per minute. During the Response Cost condition, disruptive behavior ranged
from .6 – 1.08 behaviors per minute with an average of .8 behaviors per minute and
during the Combination condition, disruptive behavior ranged from .24 - .52 behaviors
per minute with an average of .35 behaviors per minute. During the second phase of the
study, results showed that Token Economy was slightly better at reducing the rate of
disruptive behavior among students during math compared to the Combination condition.
According to figures 7 and 8, the rate of disruptive behaviors ranged from .16 - .24
behaviors per minute during Token Economy with an average of .19 behaviors per
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Phase 1 Phase 2 Phase 3
Disruptive Behaviors per Minute
Average Disruptive Behaviors by
Condition and Phase
- C.S.
Baseline
Token Economy
Response Cost
Combination
35
minute while the rate of disruptive behaviors ranged from .27 - .36 behaviors per minute
for Combination with an average of .3 behaviors per minute. The final phase of the study
consisted of the Token Economy condition for S.M. classroom. Disruptive behaviors
ranged from .12 - .26 behaviors per minute with an average of .16 behaviors per minute.
Figure 7. Rate of disruptive student behaviors per minute during the Token Economy,
Response Cost, Combination, and Baseline conditions for S.M.’s classroom across the
three phases.
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
24 25 26 27 28
29 30
Disruptive Behaviors per Minute
Session
Disruptive Behaviors
- Classroom SM
Baseline
Token Economy
Response Cost
Combination
Phase One
Phase Two
Phase Three
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Phase 1 Phase 2 Phase 3
Disruptive Behaviors per Minute
Average Disruptive Behaviors by
Condition and Phase - S.M.
Baseline
Token Economy
Response Cost
Combination
36
Figure 8. Average rate of disruptive student behaviors per minute by condition across the
three conditions for S.M.’s classroom.
Table 2
Percentage of Non-Overlapping Data Points Among Conditions for Disruptive Behavior
by Classroom
Disruptive Behavior
Condition(s)
Compared to:
Effect Size
Descriptor
C.S.
Token Economy,
Combination,
Response Cost
Baseline
100%
Highly
Effective
Combination, Token
Economy
Response Cost
100%
Highly
Effective
Response Cost
Baseline
100%
Highly
Effective
Token Economy Combination
Phase 1 =
0%
Phase 2 =
100%
Phase 1 =
Ineffective
Phase 2 =
Highly
Effective
S.M.
Token Economy,
Combination,
Response Cost
Baseline
94%
Highly
Effective
Combination, Token
Economy
Response Cost
100%
Highly
Effective
Response Cost
Baseline
89%
Moderately
Effective
Token Economy Combination
Phase 1 =
88%
Phase 2 =
100%
Phase 1 =
Moderately
Effective
Phase 2 =
Highly
Effective
37
Table 2 shows that all three conditions, Token Economy, Combination, and Response
Cost were highly effective compared to the Baseline condition at decreasing problem
behaviors in both classrooms. Token Economy and Combination conditions were Highly
Effective compared to the Response Cost condition at decreasing problem behaviors in
both classrooms. Response Cost condition was highly effective at reducing problem
behaviors compared to the Baseline condition in C.S. classroom but was only moderately
effective in S.M. classroom. Lastly, when looking at Token Economy compared to the
Combination condition, there was no difference during phase 1 in C.S. classroom,
however during phase 2 Token Economy was highly effective. For S.M. classroom,
during phase 1 Token Economy was moderately effective and during phase 2, Token
Economy was highly effective compared to the Combination condition.
Academic Performance
Results from the study show that, on average, Baseline and Token Economy
conditions were slightly better at increasing Academic Performance compared to
Response Cost and the Combination conditions for C.S. classroom (see figure 10).
During phase 1, academic performance ranged from 1.78 points to 2.91 points during
Baseline with an average of 2.37 points. During Token Economy, academic performance
ranged from 1.88 points to 2.81 points with an average of 2.35 points. During Response
Cost, academic performance ranged from 1.93 points to 2.62 points with an average of
2.22 points and during the Combination condition, academic performance ranged from
1.81 points to 2.27 points with an average of 1.99 points. During the second phase of the
study, Token Economy resulted in slightly higher Academic Performance scores
38
compared to the Combination condition with a range of 2.27 to 2.31 points (average = 2.3
points) compared to a range of 1.75 to 2.53 points (average = 2.1 points). During the final
phase of the study, only the Token Economy condition was delivered and academic
performance ranged from 1.84 – 2.8 points with an average of 2.32 points (see figures 9
& 10).
Figure 9. Academic performance as measured by average class points earned on a three
point quiz given to students daily during the Token Economy, Response Cost,
Combination, and Baseline conditions across the three phases for C.S.’ classroom.
0
0.25
0.5
0.75
1
1.25
1.5
1.75
2
2.25
2.5
2.75
3
1 2 3 4 5 6 7 8 9 101112131415161718192021222324252627282930
Average Points Earned by Class
Session
Academic Performance - C.S.
Baseline
Token Economy
Response Cost
Combination
Phase One
Phase Two
Phase Three
39
Figure 10. Average academic performance as measured by average points earned on daily
three point quiz by condition across the three phases in C.S.’ classroom.
Results from the study show that, on average, Token Economy and Response Cost
conditions were slightly better at increasing academic performance compared to the
Baseline and Combination conditions during phase 1 for S.M. classroom (see figure 12).
During phase 1, academic performance ranged from 2.2 points to 2.82 points during
Baseline with an average of 2.43 points. During Token Economy, academic performance
ranged from 2.36 points to 2. 85 points with an average of 2.59 points. During the
Response Cost condition, academic performance ranged from 2.28 points to 2.93 points
with an average of 2. 6 points and during the Combination condition academic
performance ranged from 2.07 points to 2.39 points with an average of 2.29 points.
During the second phase of the study, Token Economy resulted in slightly higher
academic performance scores compared to the Combination condition with a range of
2.31 – 2.69 points (average = 2.59 points) compared to a range of 2.31 – 2.85 points
0
0.5
1
1.5
2
2.5
3
Phase 1 Phase 2 Phase 3
Average Points Earned
Average Academic Performance by Condition
and Phase
- C.S.
Baseline
Token Economy
Response Cost
Combination
40
(average = 2.51 points). During the final phase of the study, only the Token Economy
condition was delivered and academic performance ranged from 2.58 – 2.91 points with
an average of 2.78 points (see figures 11 & 12).
Figure 11. Academic performance as measured by average class points earned on a three
point quiz given to students daily during the Token Economy, Response Cost,
Combination, and Baseline conditions across the three phases for S.M.’s classroom.
.
0
0.25
0.5
0.75
1
1.25
1.5
1.75
2
2.25
2.5
2.75
3
1
2 3 4 5 6 7 8 9 10111213
14151617
181920212223242526272829
30
Average Points Earned by Class
Session
Academic Performance
- S.M.
Baseline
Token Economy
Response Cost
Combination
Phase One
Phase Two
Phase Three
0
0.5
1
1.5
2
2.5
3
Phase 1 Phase 2 Phase 3
Average Points Earned
Average Academic Performance by Condition
and Phase - S.M.
Baseline
Token Economy
Response Cost
Combination
41
Figure 12. Average academic performance as measured by average points earned on daily
three point quiz by condition across the three phases in S.M.’s classroom.
Table 3
Percentage of Non-Overlapping Data Points among Conditions for Academic
Performance by Classroom
Academic
Performance
Condition(s)
Compared to:
Effect Size
Descriptor
C.S.
Token Economy,
Combination,
Response Cost
Baseline
0%
Ineffective
Combination, Token
Economy
Response Cost
17%
Ineffective
Response Cost
Baseline
0%
Ineffective
Token Economy Combination
Phase 1 =
25%
Phase 2 =
0%
Phase 1 =
Ineffective
Phase 2 =
Ineffective
S.M.
Token Economy,
Combination,
Response Cost
Baseline
12%
Ineffective
Combination, Token
Economy
Response Cost
0%
Ineffective
Response Cost
Baseline
11%
Ineffective
Token Economy Combination
Phase 1 =
25%
Phase 2 =
0%
Phase 1 =
Ineffective
Phase 2 =
Ineffective
42
Table 3 shows that none of the conditions, Token Economy, Combination, Response Cost
or Baseline were moderately or highly effective at increasing academic performance in
either classroom.
Student Preference
Results from the individual student interviews for C.S. classroom showed that
students favorite condition was Token Economy (n = 11) followed by Combination (n =
3), Response Cost (n = 1), and Baseline (n = 0). Students least favorite condition was
Response Cost (n = 8) followed by Baseline (n = 5), Token Economy (n = 1), and
Combination (n = 1). The condition most students wanted their teacher to implement in
the future was Combination (n = 8) followed by Token Economy (n = 6), Baseline (n =
1), and Response Cost (n = 0). One student was absent the day of the individual
interviews.
Results from the individual student interviews for S.M. classroom showed that
students favorite conditions were Combination (n = 6) and Token Economy (n = 6)
followed by Response Cost (n = 0), and Baseline (n = 0). Students least favorite condition
was Baseline (n = 7) followed by Response Cost (n = 4), Token Economy (n = 1), and
Combination (n = 0). One student was not sure which condition was their least favorite
and did not provide an answer to this question. The condition most students wanted their
teacher to implement in the future was Combination (n = 6) followed by Token Economy
(n = 5), Response Cost (n = 1), and Baseline (n = 0). One student was absent the day of
the individual interviews.
43
Teacher Preference
Results from the individual teacher interviews found that, overall, both teachers
found advantages of the classroom management strategies on student behavior compared
to the baseline condition. Teacher C.S. reported Response Cost was the easiest condition
to administer because she was able to set it up ahead of time and only take away tokens
when students were not meeting her expectations, therefore, it took the least amount of
time away from delivering instruction. Additionally, her students do not sit in one spot
throughout the lesson, and so less interaction with the tokens made the lesson run more
smoothly. C.S. reported the Combination condition was the hardest to administer because
of the amount of time it took to carry out. With her students moving throughout the
classroom during the lesson, she had to wander around the room to where the tokens were
located to remove/deliver tokens contingent on behavior for each student which took
more time and resulted in the lesson running less smoothly. With that said, C.S. reported
she felt her students were less engaged during the Response Cost condition compared to
the Token Economy and Combination conditions. She believed this was because once the
tokens were gone, her students knew there were no more opportunities to earn them back
even if they fixed the problem. When asked which condition she will most likely
implement in the future, C.S. reported she would likely implement the Combination
condition because it aligns closely with her behavior management philosophy. She
reported she is least likely to implement the Response Cost condition in the future
because the students have nothing to work for once they lose their tokens.
44
Teacher S.M. reported she liked administering the Combination Condition the
most because she felt like she had more control of her class. They were able to earn
tokens for appropriate behavior but she could also take away tokens for inappropriate
behavior. S.M. felt this encouraged her students to promote positive behavior during the
entire lesson. S.M. reported Response Cost as the condition she liked administering the
least because she did not like the idea that once her students had all of their tokens taken
away they had nothing to work for. She felt she had students who viewed losing all of
their tokens as being, “out of the game,” and no longer needed to try or engage in good
behavior. S.M. noticed her students behaved better and were more academically engaged
during all three of the classroom management strategies compared to the Baseline
condition but she felt of the three (Response Cost, Token Economy, Combination), her
students were more disruptive and less academically engaged during the Response Cost
condition. When asked which condition S.M. is most likely to implement in the future,
she reported that her students loved the Token Economy condition and also seemed to
like the Combination condition which was her favorite. She reported she is currently
implementing the Combination Condition during Math class and plans to use all three,
but primarily the Combination and Token Economy conditions, throughout her teaching
lessons as she feels they will be helpful. S.M. reported she is least likely to implement the
Response Cost condition in the future because she wants to encourage positive, respectful
behavior in her classroom. With Response Cost, she felt she was focusing mainly on
negative behavior and not able to recognize the positive behaviors as much as she wanted
to.
45
Throughout the individual teacher interviews, both teachers mentioned
advantages and disadvantages to participating in the study as well as changes to the
strategies they would consider in the future. C.S. reported the strategies were similar to
strategies she had employed in the past with her students, although she had never actually
used tokens. One of the advantages of the study then was the experience with tokens. One
of the disadvantages to the study was getting used to handing out and taking away the
tokens. It made her ability to teach at her regular pace much harder, particularly at first.
This issue was compounded by the fact that her students move around from their desk to
multiple different floor areas throughout the classroom during instructional time making
a central location for tokens to be housed impossible. Implementation would have been
easier if it would have been throughout the day instead of concentrated into the Math
lesson. C.S. reported another advantage was the opportunity to try the different classroom
management strategies in her classroom, particularly because some of them aligned so
nicely with her classroom philosophy. It gave her new strategies that were similar to her
previous strategies but with a new twist. C.S. reported the tokens were much easier to
manage when the students were at their desks working compared to on the floor because
the tokens could be housed on their name trays. C.S. reported she would also change the
number of tokens required for exchange because she went through quite a bit of candy.
S.M. reported a number of student, classroom, and teacher advantages to the
study. She felt her students were more engaged and eager to learn because a new strategy
was used each day, her students were listening more and therefore learning, her students
enjoyment of math class increased, and students practiced following the rules of being
46
respectful, responsible, and safe. S.M. reported she had a more positive, working
classroom environment and a more manageable classroom. She also reported fewer
classroom behavior disruptions overall, increase in academic engagement, and the ability
as a teacher to encourage positive behaviors. S.M. reported student, classroom, and
teacher disadvantages as well. She worried some of her students might have felt
anxiety/pressure about whether or not they were going to lose a token. She also worried
that after time, her students would have gotten tired of candy and stamps as a reward and
lose motivation and that handing out the prizes at the end of math time took time.
Additionally, at times, she lost focus of the lesson, particularly in the beginning when she
was getting used to each of the different conditions because she was focusing so much on
whether or not she should be delivering or taking away a token from a student. S.M.
reported she liked being able to use the different strategies each day to evaluate which
worked best for her students and she particularly liked how two of the three focused on
positive behaviors. She felt that many of her current classroom management tools
focused on poor behaviors rather than positive behaviors so she was happy to add two
more tools to her toolbox that focus on reinforcing positive behaviors. Like C.S., S.M.
reported at times she found it hard to hand out/take away the tokens due to the students
moving around the room and having a central location for the tokens to be housed was
not possible. One change she would think about would be to sometimes have students
start with five tokens and sometimes have them start with zero tokens for the
Combination condition.
47
Treatment Fidelity
Treatment fidelity was collected for 100% of the sessions on the instructions
given to the class, tokens handed out according to the correct schedule/condition, the
exchange of tokens at the end of math class, and the administration of the three point
quiz. Both teachers had 100% fidelity for all four treatment components across all
sessions throughout the three phases.
Inter-rater Reliability
Inter-rater reliability was collected for 33% of the sessions for each of the
classrooms. The mean total agreement for academic engagement for C.S. classroom for
all sessions was 98.3%, ranging from 95% - 100%. The total agreement for disruptive
behavior for C.S. classroom for all sessions was 100% along with the total agreement for
treatment fidelity and the grading of the three point quizzes. The mean total agreement
for academic engagement for S.M. classroom for all sessions was 98.2%, ranging from
93% - 100%. The total agreement for disruptive behavior for S.M. classroom for all
sessions was 100% along with the total agreement for treatment fidelity and the grading
of the three point quizzes.
48
CHAPTER IV
DISCUSSION
Overall, the results from this study are similar to the results of the pilot study
conducted in Florida with both Token Economy and the Combination conditions being
the most effective at increasing academic engagement and decreasing disruptive behavior
in the classroom compared to Response Cost and Baseline conditions. Additionally,
students’ favorite conditions were Token Economy and Combination conditions and their
least favorite conditions were Response Cost and Baseline. In both studies, the students
wanted their teacher to implement the Combination condition in the future over the other
three conditions.
The findings of this study are interesting because previous studies have found
little to no significant difference in the ability of these different classroom management
strategies to reduce problem behaviors (Kaufman & O’Leary, 1972; McGoey & DuPaul,
2000; McLaughlin & Malaby, 1972) while the current study found a difference between
the conditions. Both Token Economy and Combination conditions were more effective at
reducing disruptive classroom behaviors compared to the punishment based Response
Cost condition. The results of the current study also showed a difference between the
conditions for academic engagement. While some previous studies have shown little to
no difference between the punishment based Response Cost conditions and the reward
based Token Economy strategies (Kaufman & O’Leary; Panek, 1970) at increasing
academic engagement, the current study showed that the reward based condition, Token
49
Economy, and Combination conditions were more effective at increasing academic
engagement in students during math class.
Additionally, previous studies have shown support for Response Cost to be the
preferred or “accepted” classroom management strategy among educators and parents
(Frentz & Kelley, 1986; Little & Kelley, 1989), however, the current study shows support
for the Token Economy and Combination conditions. While C.S. reported the Response
Cost condition was the easiest to implement, she stated it was also the condition in which
her students were the least academically engaged. Additionally, she reported in the
future, she is most likely to implement the Combination condition and least likely to
implement the Response Cost condition. S.M. had a similar viewpoint although she said
the Combination condition was her favorite condition and Response cost was her least
favorite to administer. Like C.S., S.M. reported her students were the least academically
engaged during the Response Cost condition and not only is she more likely to implement
both the Combination and Token Economy conditions in the future, she is currently
implementing the Combination condition during Math.
The students in the current study clearly preferred the Token Economy condition
and Combination condition over the Response Cost and Baseline conditions. Some of the
reasons why they preferred Token Economy included, “You can’t lose them, you can
only earn them,” “I like earning them and not getting them taken away,” and “You can’t
get them taken away,” “I like earning them.” Some of the reasons stated why they
preferred the Combination condition included, “You start with 5 and you can lose and
earn so you do not have to be so good,” “because you earn and lose and you start with
50
some,” “I just like it so much,” “You can get them taken away but you can earn them
again,” “Because you start with 5 and you can earn more,” and “Because when you make
a mistake and you get them taken away you can earn it back again if you do ok.” The
student who said Response Cost was their favorite condition said it was because, “You do
not have to earn any, you just keep the ones you have or get them taken away.” Some of
the reasons students said Response Cost was their least favorite condition included; “You
can’t earn them back once you lose them,” “I don’t like getting them taken away,” “The
teacher takes mine away,” “Because mine always get taken away,” “They can get taken
and you can’t get them back,” and “Can only get them taken away but you can’t earn
them back no matter how hard you try.” Some of the reasons students said Baseline was
their least favorite condition included; “Because I like candy and you don’t get candy,”
“No tokens,” “You can’t earn tokens,” “You can’t earn candy or stamps,” “I like tokens
and want them,” “Because we start with none and end with none,” “Because we don’t get
any candy and no tokens,” and “Because we have no tokens.” The student who said
Token Economy was their least favorite said it was because, “You have to be like an
‘angel’ to get a token.” The other student who said Token Economy was their least
favorite and the student who said Combination was their least favorite was unable to
provide an answer for why. Unlike the pilot study, the researchers were unable to ask
students which condition(s) made it easier/harder for them to learn and why due to the
age of the students and their limited ability to self-report.
The current study sought to address the methodological weaknesses of previous
studies assessing Token Economy systems as evidence based classroom management
51
strategies for reducing problem behaviors among students. This study addressed the
methodological weaknesses by: collecting and analyzing treatment fidelity, social
validity, and interobserver agreement data. Additionally, the study met the requirements
for a sound methodological design such as including a baseline, having a minimum of
three data points per condition and phase, and operationally defining the independent and
dependent variables (Kratochwill et al., 2002; 2010).
Although the pilot study found a difference between the conditions for academic
performance using the ‘Paws up for Understanding’ self-assessment, the current study did
not find a clear distinction between the conditions. Each time a new math topic or skill
was introduced, student performance on the three point quiz decreased and gradually
increased throughout the unit until it was complete and the next topic or skill was
introduced again. Therefore, one limitation to the current study was that history effect
may not have been controlled for as tightly as planned for measuring academic
performance with the three point quiz.
A second limitation to the study was the limited settings assessed. The conditions
were only assessed during the math lesson each day for each teacher and only for first
grade. A larger, more diverse sample size and longer session length would be beneficial
for future studies, particularly to see if the trend of student and teacher preference for
Token Economy and Combination conditions continue and to determine if these two
conditions are more effective at increasing academic engagement and reducing disruptive
behavior in the classrooms compared to the Response Cost and Baseline conditions.
52
A third limitation to the study was the administration of the tokens within each
condition. The teacher was responsible for delivering/removing tokens contingent on
student behavior along with teaching her class. Therefore, at times, the tokens were not
administered as rigorously as they could have been if someone were in the classroom
with the sole responsibility of watching student behavior and manipulating tokens
contingent on student behavior.
53
APPENDIX A:
CONSENT AND ASSENT FORMS:
1. Teacher Consent Form
2. Principal Consent Form
3. Parent/Guardian Consent Form
4. Student Assent Form (Ages 6-9)
5. Student Assent Form (Ages 10-13)
6. Agency Consent Form
7. Agency Letter of Intent
54
Teacher Consent Form
Informed Consent
We are interested in conducting a research project with teachers in your school district.
At this time, it is our prospect to train teachers to use three different classroom
management procedures; response cost, token economy, and a combination of response
cost and token economy. Data on the instances of problem behaviors as well as academic
performance will be recorded and analyzed through visual analysis of the data. The
teachers we are requesting to work with are those working in elementary schools teaching
children in general education classrooms.
This proposed research project is in association with the School Psychology Doctoral
Program of Minnesota State University, Mankato and has passed the human subject’s
research review. Participation in this study is voluntary, and refusal or withdrawal from
the study, at any time, will involve no penalty to participants in the study nor will it
impact any current or future relationship with Minnesota State University Mankato. Dan
Houlihan, Ph.D. is supervising this research, which is being proposed by Britta Fiksdal,
M.A., School Psychology Graduate Student. If there are any questions, feel free to call
Dr. Houlihan at (507)389-6308 or Britta Fiksdal at (507) 450-1828. The IRB case
number for this study is: 538290-1
The following is a description of the research we are requesting to perform (See Teacher
Manual below for additional details):
It is our desire to obtain the consent of teacher(s) within your school district in order to
obtain data from their classrooms during four different classroom management
conditions; response cost, token economy, a combination of token economy and response
cost, and a control baseline condition. Data on the instances of problem behaviors
exhibited as well as academic performance will be collected and compared to see if there
is a difference in effectiveness between the conditions. Research has shown that response
cost is just as effective in reducing problem behaviors as token economies and token
economies that incorporate response cost systems. The specific purpose of our research is
to compare the effectiveness of academic performance between the four conditions along
with problem behaviors. Each teacher will be collecting their own data on problem
behaviors and academic performance for their class on a separate data sheet and reporting
to the researchers on a daily basis. In light of this research, it is our prospect that this
information aid researchers in proposing future areas of research and may prove valuable
in the training of teachers and classroom management.
At no time will we record any behavior of the children in the classroom or an individual
who has not consented to our observation. In addition, no identifying information
55
regarding the students will be taken and any identifying information about the teachers
will be kept in our secured research lab in a locked cabinet. Students will be assigned a
number id by the teachers; therefore, data given to the researchers will not have any
identifying information.
Each of the teachers who agree to participate in our study will be asked to give formal
consent to the observation of them in their classroom. With agency and teacher consent,
the study should take approximately two months.
It is anticipated that the teachers and students of the classroom(s) involved will not
experience any harmful affects whatsoever from participating in this study. In this
respect, in no way would any information gained from the observation be used in a
judgmental manner toward the teacher(s) or be shared with the public in a judgmental
manner. Additionally, data will be analyzed by Dr. Houlihan and students associated with
our research team in the School Psychology Doctoral Program at Minnesota State
University, Mankato who will be trained in proper methods of informed consent and
confidentiality. All consent forms and data will be stored in a locked cabinet in a locked
closet in the School Psychology clinic on campus under the care of Dr. Daniel Houlihan.
We intend to complete this study this winter and anticipate working with your school. We
appreciate your time in considering working with us on this endeavor. Again, please feel
free to contact Dr. Houlihan at (507)389-6308 if you have any questions. If you have
questions regarding the rights and treatment of human subjects participating in research
studies, you can contact the Dean of Graduate Studies, Dr. Barry Ries at (507) 389-2321.
We would greatly appreciate you working with us and request that you complete the
below form giving permission for schools in your district to participate in our research.
Sincerely,
Dan Houlihan, Ph.D. Britta Fiksdal, M.A.
Professor and Director of School Psychology Doctoral Graduate Student
56
------------------------------------------------------------------------------------------------------------
---------
I have read the above description of the research study on Teacher/Student Interaction to
be conducted by Dan Houlihan, Ph.D. in conjunction with students from the School
Psychology Doctoral Program at Minnesota State University, Mankato.
I give consent for researchers associated with this study to naturalistically observe me in
my classroom.
Teacher’s Name (Print): ________________________________Title:
_____________________
School: __________________________________Grade/Classroom:
______________________
Level of Education Licensure:_________________________________
Teacher’s
Signature_________________________________________________Date___________
______
Participant Code #:_____________
57
Principal Consent Form
Dear ___________________, Date___________________
We are interested in conducting a research project with teachers in your school district.
At this time, it is our prospect to train teachers to use three different classroom
management procedures; response cost, token economy, and a combination of response
cost and token economy. Data on the instances of problem behaviors as well as academic
performance will be recorded and analyzed through visual analysis of the data. The
teachers we are requesting to work with are those working in elementary schools teaching
children in general education classrooms.
This proposed research project is in association with the School Psychology Doctoral
Program of Minnesota State University, Mankato and has passed the human subject’s
research review. Participation in this study is voluntary, and refusal or withdrawal from
the study, at any time, will involve no penalty to participants in the study nor will it
impact any current or future relationship with Minnesota State University Mankato. Dan
Houlihan, Ph.D. is supervising this research, which is being proposed by Britta Fiksdal,
M.A., School Psychology Graduate Student. If there are any questions, feel free to call
Dr. Houlihan at (507)389-6308 or Britta Fiksdal at (507) 450-1828. The IRB case
number for this study is: 538290-1
The following is a description of the research we are requesting to perform:
It is our desire to obtain the consent of teacher(s) within your school district in order to
obtain data from their classrooms during four different classroom management
conditions; response cost, token economy, a combination of token economy and response
cost, and a control baseline condition. Data on the instances of problem behaviors
exhibited as well as academic performance will be collected and compared to see if there
is a difference in effectiveness between the conditions. Research has shown that response
cost is just as effective in reducing problem behaviors as token economies and token
economies that incorporate response cost systems. The specific purpose of our research is
to compare the effectiveness of academic performance between the four conditions along
with problem behaviors. Each teacher will be collecting their own data on problem
behaviors and academic performance for their class on a separate data sheet and reporting
to the researchers on a daily basis. In light of this research, it is our prospect that this
information aid researchers in proposing future areas of research and may prove valuable
in the training of teachers and classroom management.
58
At no time will we record any behavior of the children in the classroom or an individual
who has not consented to our observation. In addition, no identifying information
regarding the students will be taken and any identifying information about the teachers
will be kept in our secured research lab in a locked cabinet. Students will be assigned a
number id by the teachers; therefore, data given to the researchers will not have any
identifying information.
Each of the teachers who agree to participate in our study will be asked to give formal
consent to the observation of them in their classroom. With agency and teacher consent,
the study should take approximately two months.
It is anticipated that the teachers and students of the classroom(s) involved will not
experience any harmful affects whatsoever from participating in this study. In this
respect, in no way would any information gained from the observation be used in a
judgmental manner toward the teacher(s) or be shared with the public in a judgmental
manner. Additionally, data will be analyzed by Dr. Houlihan and students associated with
our research team in the School Psychology Doctoral Program at Minnesota State
University, Mankato who will be trained in proper methods of informed consent and
confidentiality. All consent forms and data will be stored in a locked cabinet in a locked
closet in the School Psychology clinic on campus under the care of Dr. Daniel Houlihan.
We intend to complete this study this winter and anticipate working with your school. We
appreciate your time in considering working with us on this endeavor. Again, please feel
free to contact Dr. Houlihan at (507)389-6308 if you have any questions. If you have
questions regarding the rights and treatment of human subjects participating in research
studies, you can contact the Dean of Graduate Studies, Dr. Barry Ries at (507) 389-2321.
We would greatly appreciate you working with us and request that you complete the
below form giving permission for schools in your district to participate in our research.
Sincerely,
Dan Houlihan, Ph.D. Britta Fiksdal, M.A.
Professor and Director of School Psychology Doctoral
Graduate Student
59
------------------------------------------------------------------------------------------------------------
---------
I have read the above description of the research study on Teacher Requests to be
conducted by Dan Houlihan, Ph.D. in conjunction with students from the School
Psychology Doctoral Program at Minnesota State University, Mankato. In addition, I
understand that participation in this study is voluntary.
I give permission for _____________________________________ School to participate
in this research study.
In addition, I give permission for the following classrooms to be contacted regarding this
study:
________________________________________________________________________
______
________________________________________________________________________
______
________________________________________________________________________
______
Principal’s
Name ______________________________________________ Date
_____________________
60
Parental Consent Form for Participation in Research
The purpose of this letter is to give consent for your child to participate in the
research titled, “Comparing the Effectiveness of Reducing Problem Behavior and
Increasing Academic performance between a Response Cost, Token Economy, and a
Combination Condition” conducted by Dan Houlihan, Ph.D. (Professor of School
Psychology) and Britta Fiksdal, M.A. (School Psychology Doctoral Graduate Student) in
association with Minnesota State University, Mankato. Participation is entirely voluntary.
At this time, it is our prospect to train teachers to use three different classroom
management procedures (token economy, response cost, and combination). Data on the
instances of problem behaviors as well as academic performance will be recorded and
analyzed through visual analysis of the data, however, all student information will be
kept confidential and no identifying information will be shared about your student
(students will be given id numbers by the teachers prior to submitting any data). Previous
research shows that each of these three classroom management strategies is effective in
reducing problem behavior; however, little research has focused on the effects on student
academic performance. This study has been approved through the University’s
Institutional Review Board as well as approved by your child’s school district, principal
(name of principal here), and classroom teacher (name of classroom teacher here).
During the token economy condition, your student will receive tokens for
exhibiting appropriate or “good” behavior. During the response cost condition, your
student will receive a certain number of tokens at the beginning of the class. They will
lose one token if they exhibit inappropriate or “problem” behavior. During the
combination condition, your student will start with a certain number of tokens. Your
student will receive tokens for exhibiting appropriate or “good” behavior and will lose
tokens if they exhibit inappropriate or “problem” behavior. At the end of each class, your
student can trade their tokens in for a reward. The amount or type of reward will depend
on the amount of tokens your student has earned. During the baseline condition, students
will not have the opportunity to lose or earn tokens. Throughout each condition, the
61
teacher will be delivering instruction as he/she normally would. Your student will not
miss out on any educational instruction by participating in this study nor will it negatively
impact their learning environment.
The results of this participation will be confidential and will not be released in any
individually identifiable form. Additionally, data will be analyzed by Dr. Houlihan and
Britta Fiksdal, who are trained in proper methods of informed consent and
confidentiality. Data collection procedures will take place over a period of seven to eight
weeks. Per district requirements, only those children with signed parent consent forms
will be included in the data collection.
This proposed research project is in association with the School Psychology
Doctoral Program of Minnesota State University, Mankato and has passed the human
subject’s research review. Participation in this study is voluntary, and refusal or
withdrawal from the study, at any time, will involve no penalty to participants in the
study nor will it impact any current or future relationship with Minnesota State
University Mankato. Dan Houlihan, Ph.D. is supervising this research, which is being
proposed by Britta Fiksdal, M.A., School Psychology Graduate Student. If there are any
questions, feel free to call Dr. Houlihan at (507)389-6308 or Britta Fiksdal at (507) 450-
1828. The IRB case number for this study is: 538290-1
We intend to complete this study this winter and we appreciate your consent for
your child to participate in this endeavor. Feel free to contact Britta Fiksdal (507) 450-
1828 or Dr. Houlihan (507) 389-6308 if you have any questions.
We greatly appreciate you working with us and request that you sign the below
form giving permission for your child to participate in this research study.
62
Sincerely,
Dan Houlihan, Ph.D. Britta Fiksdal, M.A.
Professor and Director of School Psychology Doctoral Graduate
Student
------------------------------------------------------------------------------------------------------------
---------
I have read the above description of the research study on Classroom Management to be
conducted by Dan Houlihan, Ph.D. in conjunction with students from the School
Psychology Doctoral Program at Minnesota State University, Mankato.
I give consent for my child to participate in this research project associated with
Classroom Management.
Parent Name (Print): ________________________ Child Name
(Print):__________________
Parent Signature: _______________________ Date:_______________________
63
Student Assent (Ages 6-9)
Student’s Name _____________________________________
My name is _______________________ and I am from Minnesota State University,
Mankato. I will be working with your teacher on different ways to teach your class. Your
teacher may be handing out tokens that can be exchanged for prizes during class. I would
like you to be on your best behavior and listen to your teacher like you normally would.
During class, when the teacher hands out tokens, you can save them and exchange them
for prizes at the end of the class or day. Some days you will be given tokens throughout
class, other days you will be given tokens at the beginning of class, and some days there
will not be any tokens. Your teacher will explain to you each day whether or not you can
earn tokens. You cannot save them overnight; they must be exchanged before going
home from school each day. Your teacher will be watching to see how students behave in
your classroom and how well they are paying attention. You will not be taken out of your
classroom at any time nor will you be asked to do anything different from the rest of your
class.
After about two months of working with your classroom, I will ask you some questions
about the tokens. Your answers will only be used by researchers to learn more about
children like you. The teachers and other students in your school will never know your
answers to the questions. Your answers will be written down and put with other
children’s answers. Your answers will help other teachers learn about children and the
best way to teach them.
Your parent or guardian and teacher have said that it is okay for you to participate. Your
teacher will be handing out tokens most days for approximately two months. If you
decide that you do not want to receive tokens or you do not wish to answer the questions
at the end just tell your teacher at any time. This is not a test and there are no wrong
answers or behaviors. You will receive prizes when you earn and exchange tokens. These
prizes will be determined by your teacher but could include items such as a pencil, piece
of candy, getting to line up first for lunch, etc.
64
Tell your teacher or parents, if you are worried or unhappy about anything that happens
during class the next two months.
__________________________________________ _________________
Signature Date
65
Student Assent (Ages 10-13)
Student’s Name _____________________________________
My name is _______________________ and I am from Minnesota State University,
Mankato. You are being asked to be part of a research project that will help adults
understand how to best teach elementary students. I will be working with your teacher on
different ways to teach your class. Your teacher may be handing out tokens that can be
exchanged for prizes during class. I would like you to be on your best behavior and listen
to your teacher like you normally would.
During class, when the teacher hands out tokens, you can save them and exchange them
for prizes at the end of the class or day. Some days you will be given tokens throughout
class, other days you will be given tokens at the beginning of class, and some days there
will not be any tokens. Your teacher will explain to you each day whether or not you can
earn tokens. You cannot save them overnight; they must be exchanged before going
home from school each day. Your teacher will be watching to see how students behave in
your classroom and how well they are paying attention. You will not be taken out of your
classroom at any time nor will you be asked to do anything different from the rest of your
class.
After about two months of working with your classroom, I will ask you some questions
about the tokens. Your answers will only be used by researchers to learn more about
children like you. The teachers and other students in your school will never know your
answers to the questions. Your answers will be written down and put with other
children’s answers. Your answers will help other teachers learn about children and the
best way to teach them.
Your parent or guardian and teacher have said that it is okay for you to participate. Your
teacher will be handing out tokens most days for approximately two months. If you
decide that you do not want to receive tokens or you do not wish to answer the questions
at the end just tell your teacher at any time. This is not a test and there are no wrong
answers or behaviors. You will receive prizes when you earn and exchange tokens. These
prizes will be determined by your teacher but could include items such as a pencil, piece
of candy, getting to line up first for lunch, etc.
66
Tell your teacher or parents, if you are worried or unhappy about anything that happens
during class the next two months.
__________________________________________ _________________
Signature Date
67
Agency Consent Form
Dear ___________________, Date___________________
We are interested in conducting a research project with teachers in your school district.
At this time, it is our prospect to train teachers to use three different classroom
management procedures; response cost, token economy, and a combination of response
cost and token economy. Data on the instances of problem behaviors as well as academic
performance will be recorded and analyzed through visual analysis of the data. The
teachers we are requesting to work with are those working in elementary schools teaching
children in general education classrooms.
This proposed research project is in association with the School Psychology Doctoral
Program of Minnesota State University, Mankato and has passed the human subject’s
research review. Participation in this study is voluntary, and refusal or withdrawal from
the study, at any time, will involve no penalty to participants in the study nor will it
impact any current or future relationship with Minnesota State University Mankato. Dan
Houlihan, Ph.D. is supervising this research, which is being proposed by Britta Fiksdal,
M.A., School Psychology Graduate Student. If there are any questions, feel free to call
Dr. Houlihan at (507)389-6308 or Britta Fiksdal at (507) 450-1828. The IRB case
number for this study is: 538290-1
The following is a description of the research we are requesting to perform:
It is our desire to obtain the consent of teacher(s) within your school district in order to
obtain data from their classrooms during four different classroom management
conditions; response cost, token economy, a combination of token economy and response
cost, and a control baseline condition. Data on the instances of problem behaviors
exhibited as well as academic performance will be collected and compared to see if there
is a difference in effectiveness between the conditions. Research has shown that response
cost is just as effective in reducing problem behaviors as token economies and token
economies that incorporate response cost systems. The specific purpose of our research is
to compare the effectiveness of academic performance between the four conditions along
with problem behaviors. Each teacher will be collecting their own data on problem
behaviors and academic performance for their class on a separate data sheet and reporting
to the researchers on a daily basis. In light of this research, it is our prospect that this
information aid researchers in proposing future areas of research and may prove valuable
in the training of teachers and classroom management.
At no time will we record any behavior of the children in the classroom or an individual
who has not consented to our observation. In addition, no identifying information
regarding the students will be taken and any identifying information about the teachers
will be kept in our secured research lab in a locked cabinet. Students will be assigned a
68
number id by the teachers; therefore, data given to the researchers will not have any
identifying information.
Each of the teachers who agree to participate in our study will be asked to give formal
consent to the observation of them in their classroom. With agency and teacher consent,
the study should take approximately two months.
It is anticipated that the teachers and students of the classroom(s) involved will not
experience any harmful affects whatsoever from participating in this study. In this
respect, in no way would any information gained from the observation be used in a
judgmental manner toward the teacher(s) or be shared with the public in a judgmental
manner. Additionally, data will be analyzed by Dr. Houlihan and students associated with
our research team in the School Psychology Doctoral Program at Minnesota State
University, Mankato who will be trained in proper methods of informed consent and
confidentiality. All consent forms and data will be stored in a locked cabinet in a locked
closet in the School Psychology clinic on campus under the care of Dr. Daniel Houlihan.
We intend to complete this study this winter and anticipate working with your school. We
appreciate your time in considering working with us on this endeavor. Again, please feel
free to contact Dr. Houlihan at (507)389-6308 if you have any questions. If you have
questions regarding the rights and treatment of human subjects participating in research
studies, you can contact the Dean of Graduate Studies, Dr. Barry Ries at (507) 389-2321.
We would greatly appreciate you working with us and request that you complete the
below form giving permission for schools in your district to participate in our research.
Sincerely,
Dan Houlihan, Ph.D. Britta Fiksdal, M.A.
Professor and Director of School Psychology Doctoral
Graduate Student
69
------------------------------------------------------------------------------------------------------------
---------
I have read the above description of the research study on Classroom Management to be
conducted by Dan Houlihan, Ph.D. in conjunction with students from the School
Psychology Doctoral Program at Minnesota State University, Mankato. In addition, I
understand that participation in this study is voluntary and I can discontinue at any time
with no penalty or negative consequences.
In signing this form, I give permission for researcher to contact school principals in the
_______________________district to participate in this research study.
___ I do wish to receive a copy of this consent form
___ I do not wish to receive a copy of this consent form
Agency
Signature______________________________________________ Date
___________________
70
Agency Letter of Intent
71
APPENDIX B:
OBSERVATION FORMS AND TEACHER MATERIALS
1. Teacher Manual
2. Academic Engagement and Disruptive/Problem Behavior Observation Form
3. Academic Performance Observation Form
4. Treatment Integrity Observation Form
72
Teacher Manual
Implementation, Recording, and Observation
Purpose
The purpose of the study is to compare the effectiveness of a response cost condition, a
token economy condition, and a combination condition for both reducing problem
behavior and increasing academic behavior among fourth grade students. The procedure
will involve training teachers in on three different classroom management strategies;
response cost, token economy, and a combination of response cost and token economy.
Data on the instances of problem behaviors and academic performance will be recorded
by the teacher and analyzed by the researchers.
Implementation and Recording of Data
Problem Behavior
The observation form utilized in this study will be used to collect data regarding the
instances of problem behaviors among students. The teacher will indicate when a student
exhibits a problem behavior by recording which student and which problem behavior was
exhibited. The different problem behaviors to be recorded will be discussed with each
teacher to determine what behavior each specific teacher sees as being problematic.
Additionally, the data collection sheets will be developed with the teacher to increase
understanding of data collection and ensure feasibility of the data collection procedure.
Academic Performance and Understanding
Teachers will also record the percentage of time students were academically engaged
using a simple likert scale that will be developed with the teacher’s input prior to the
study beginning. Each classroom is different and so the data collection will be tailored to
each class. Currently, the teachers are doing a Paws up For Understanding at the end of
each lesson. This data is already being collected by the teacher but will be handed to the
researcher for academic understanding analysis. Overall, all data will be collected by the
teacher and analyzed by the researchers. The Paws up for Understanding can be recorded
at a later time and is already collected. The data for academic engagement will consist of
the teacher simply circling a number that best corresponds with their class’ engagement
during the condition. The data for problem behaviors will be collected using a simple
clicker counter and the number on the counter at the end of the condition will simply be
written down on the top of the academic engagement data sheet. Data collection will be
very short and easy for the teachers.
73
Conditions
During each condition the teacher will observe and record the instances of problem
behavior and deliver/remove tokens depending on the condition. Each day, the teacher
will implement one of the classroom management strategies and collect data for either the
entire day or half of the day depending on the teacher preference and daily schedule (see
below for teacher instructions).
Informed Consent and Debriefing
Teachers that will be participating have been provided informed consents. We will plan
to hold update meetings as we complete each phase of the study.
Design
For the first phase, the teachers will alternate between the four different conditions
throughout four weeks. The order of conditions for each week will be determined through
a Latin squares design and provided to the teacher at the beginning of the study. For
example, the first week the schedule could look like this: Monday implement Baseline,
Tuesday implement Token Economy, Wednesday implement Combination, Thursday
implement Response Cost, and Friday implement Baseline. Teachers will have the option
of either implementing the condition for an entire day or half day depending on their
schedule. Phase 2, weeks five through six, will consist of rotating between the two
conditions that data indicates is the most effective at reducing problem behaviors and
increasing academic engagement and understanding. Phase 3, the seventh week, will
consist of the teacher implementing the final condition that data indicates is superior at
reducing problem behaviors and increasing academic engagement and understanding. As
with the first four weeks, the teacher will have the option of implementing the condition
for an entire day or half day. It is important to note that during the Baseline condition,
teachers will go about their teaching day as normal and only have to collect the data on
problem behaviors and academic engagement and understanding. They will not have to
deliver or take away tokens nor will there be tokens to exchange at the end of the session
Observer Training (see Appendix D)
Researchers involved in the study will be trained to accurately observe and code the
problem behaviors identified by each teacher. First, observers will learn the operational
definitions of the different problem behaviors (will be determined with the teacher based
on the problem behaviors they see in their classroom). In addition to recording the
occurrences of each problem behavior, academic engagement and academic
understanding will also be recorded. To keep the student information confidential, the
teacher will enter the names of each student with a teacher identified ID number and once
the data sheet is filled out, will remove the student names from the sheet keeping only the
ID numbers before sending the information to us. The teacher will be responsible for
collecting data, however, data collection will be made as easily as possible by developing
74
the data collection sheet with the teacher to identify preferences and feasibility. The
prizes that tokens can be exchanged for will be identified by the teachers.
Phase 1: Alternating Treatment between 4 Conditions Daily
For the first part of the study, the teacher will alternate between the four conditions on a
daily basis. The order of conditions will be randomly assigned at the beginning of the
study. See Design section above for example of weekly schedule. The four conditions
will be Baseline, Token Economy, Response Cost, and Combination. See below for
details on each condition.
Phase 2: Alternating Treatment between 2 Conditions Daily
Based on the data from Phase 1 of the study, the two most effective/efficient conditions
will be continued and the teacher will alternate between the two conditions on a daily
basis. The order of the conditions will be randomly assigned immediately following
Phase 1 and before phase 2 starts. The most effective/efficient conditions will be
determined by comparing the rate of problem behavior during each condition along with
academic engagement and academic understanding.
Phase 3: Final Condition
Based on the data from Phase 2 of the study, the most effective/efficient condition will be
continued and the teacher will implement that one condition each day for the remaining
week. The most effective/efficient condition will be determined by comparing the rate of
problem behavior during each condition along with academic engagement and academic
understanding.
Token Economy Condition
During this condition, the teacher will start the lesson by giving the students the
following instructions, “During this lesson, you will have the opportunity to earn a token
for “good” behavior. When you earn a token, I will place it in your envelope (or cup)
located at your desk. Your tokens cannot be taken away right now, you can only earn
them for good behavior. At the end of the lesson you can exchange your tokens for items
in the reward box.” The teacher will then start the lesson as normal and deliver tokens to
students contingent upon desirable behavior. When delivering a token, the teacher will
briefly state what behavior the student is earning the token for (e.g. “I like the way you
are reading quietly in your seat.”). The teacher will continue to deliver tokens throughout
the chosen lesson for the day. At the end of the lesson, the teacher will allow students to
exchange their tokens for a reward. The magnitude and size of the reinforcer will be
determined by the number of tokens the student exchanges.
Response Cost Condition
During this condition, the teacher will start the lesson by giving the students the
following instructions, “During this lesson, each of you will be given ten tokens in your
envelope (or cup) located at your desk. Each time you misbehave, I will come and take a
75
token away. You cannot earn tokens back right now; you can only keep them if you do
not engage in any problem behaviors. At the end of the lesson you can exchange
whatever tokens you have left for items in the reward box.” The teacher will then gave
each student three tokens in their envelope (or cup) located at each student’s desk and
start the lesson as normal. Throughout the lesson, anytime a student engages in problem
behavior (as identified in the problem behavior definition list) the teacher will go over to
the student and quietly take away a token from their envelope (or cup) and tell the student
what they did wrong (e.g. “I do not like the way you are twirling your book, instead you
should be reading chapter 4.”). The teacher will continue to take away tokens throughout
the condition contingent on problem behavior. At the end of the condition, the teacher
will allow students to exchange their tokens for a reward. The magnitude and size of the
reinforcer will be determined by the number of tokens the student has left and is
exchanging.
Combination Condition
During this condition, the teacher will start the lesson by giving the students the
following instructions, “During this lesson, you will have the opportunity to earn a token
for “good” behavior. When you earn a token, I will place it in your envelope (or cup)
located at your desk. Your tokens can be taken away if you engage in any problem
behaviors. So throughout this lesson, you can earn tokens for good behavior AND get
your tokens taken away for bad behavior. At the end of the lesson you can exchange
however many tokens you have for items in the reward box.” The teacher will then start
the lesson as normal and put a token in student’s cups contingent on good behavior with a
brief, quiet description of what behavior the student is earning the token for (e.g. “I like
the way you are reading quietly in your seat.”) and take away a token contingent on
inappropriate behavior with a brief, quiet description of what behavior the student is
getting a token taken away for (e.g. “I do not like the way you are singing and looking
around instead of reading your book.”). The teacher will continue to deliver and take
away tokens throughout the lesson. At the end of the lesson, the teacher will allow
students to exchange their tokens for a reward. The magnitude and size of the reinforcer
will be determined by the number of tokens the student has left and is exchanging.
Baseline Condition
During baseline, the teacher will start the lesson by giving the students the following
instructions, “During this lesson, you will not be given any tokens nor will you be able to
lose any tokens. I still want you all to be on your best behavior.” The teacher will then
teach the lesson as normal, without delivering any type of tangible reinforcement
contingent on behavior. At the end of the lesson, the teacher will transition the kids to the
next activity since there will be no tokens for students to exchange.
76
Academic Engagement and Disruptive/Problem Behavior Observation Form
Day: Subject: Problem Behaviors:
Condition: Total PB:
Min :00 :15 :30 :45 Min :00 :15 :30 :45 Min :00 :15 :30 :45
0
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
15
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
30
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
1
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
16
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
31
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
2
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
17
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
32
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
3
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
18
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
33
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
4
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
19
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
34
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
5
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
20
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
35
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
6
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
21
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
36
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
7
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
22
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
37
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
8
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
23
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
38
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
9
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
24
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
39
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
10
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
25
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
40
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
11
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
26
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
41
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
12
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
27
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
42
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
13
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
28
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
43
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
14
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
29
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
44
AE DS
DB
AE DS
DB
AE DS
DB
AE DS
DB
77
Academic Performance Observation Form
Date:___________ Condition:______________ Teacher:________________
Student ID
Points Earned
Class Average
1
Day Class Average =
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Additional Notes:
78
Treatment Integrity Observation Form
Date: _______________ Condition:____________ Class:_____________
Treatment
Component
Correct
Incorrect
Not Observed
Additional Notes/Comments
Read Instructions for
Condition – at
beginning of math
class
Handed out/took
away tokens
contingent on
behavior – according
to that day’s
condition
Gave students 3 point
quiz at end of math
class
Allowed students to
exchange tokens for
backup reinforcers
immediately after
math class
79
APPENDIX C:
DATA COLLECTION FORMS:
1. Student Preference Interview
2. Social Validity Teacher Interview
80
Student Preference Interview
1. Of the four conditions, which one did you like the most?
2. Why did you like that one the most?
3. Of the four conditions, which one did you like the least?
4. Why did you like that one the least?
5. Which one would you like your teacher to continue next week?
81
Social Validity Teacher Interview
1. Which procedure did you like administering the best? Why?
2. Which procedure did you like administering the least? Why?
3. What were some advantages of the different strategies to your
students/classroom/teaching/etc.?
4. What were some disadvantages of the different strategies to your
students/classroom/teaching/etc.?
5. Did you notice your students behaving better or more academically engaged
during any of the conditions? If so which one(s)?
6. Did you notice your students misbehaving more or academically engaged less
during any of the conditions? If so, which one(s)?
7. What did you like about employing the different strategies/participating in the
study?
82
8. What didn't you like about employing the different strategies/participating in the
study?
9. If you could make any changes to the strategies what would they be?
10. Which one would you be most likely to do in the future? Why? Are you currently
implementing any of them in your classroom?
11.
Which one would you be least likely to do in the future? Why?
83
References
Ayllon, T., & Azrin, N. H. (1968). The Token Economy. New York: Appleton-Century-
Crofts.
Birnbrauer, J. S., Wolf, M. M., Kidder, J. D., & Tague, C. E. (1965). Classroom behavior
of retarded pupils with token reinforcement. Journal of Experimental Child
Psychology, 2, 219-235.
Broden, M., Hall, R. V., Dunlap, A., & Clark, R. (1970). Effects of teacher attention and
a token reinforcement system in a junior high school special education class.
Exceptional Children, 36, 341-349.
Bushell, D., Wrobel, P. A., & Michaelis, M. L. (1968). Applying “group” contingencies
to the classroom study behavior of preschool children. Journal of Applied
Behavior Analysis, 1, 55-63.
Conyers, C., Miltenberger, R., Maki, A., Barenz, R., Jurgens, M., Sailer, A., Haugen, M.
et al. (2004). A comparison of response cost and differential reinforcement of
other behavior to reduce disruptive behavior in a preschool classroom. Journal of
Applied Behavior Analysis, 37, 411-415.
Iwata, B. A., & Bailey, J. S. (1974). Reward versus cost token systems: An analysis of
the effects on students and teacher. Journal of Applied Behavior Analysis, 7, 567-
576.
84
Jones, M. L., Eyberg, S. M., Adams, C. D., & Boggs, S. R. (1998). Treatment
acceptability of behavioral interventions for children: an assessment by mothers
of children with disruptive behavior disorders. Child and Family Behavior
Therapy, 20, 15-26.
Kahng, S. W., Boscoe, J. H., & Byrne, S. (2003). The use of an escape contingency and a
token economy to increase food acceptance. Journal of Applied Behavior
Analysis, 36, 349-353.
Kazdin, A. E. (1972). Response cost: The removal of conditioned reinforcers for
therapeutic change. Behavior Therapy, 3, 533-546.
Kazdin, A. E. (1977). The token economy: A review and evaluation. New York: Plenum
Press.
Kazdin, A. E. (1982). The token economy: A decade later. Journal of Applied Behavior
Analysis, 15, 431 – 445.
Kazdin, A. E., & Bootzin, R. R. (1972). The token economy: An evaluative review.
Journal of Applied Behavior Analysis, 5, 343-372.
Kaufman, K. F., & O’Leary, K. D. (1972). Reward, cost, and self-evaluation procedures
for disruptive adolescents in a psychiatric hospital school. Journal of Applied
Behavior Analysis, 5, 293-309.
Kratochwill, T. R., Hitchcock, J., Horner, R. H., Levin, J.R., Odom, S. L., Rindskopf, D.
M., et al. (2010). Single-case designs technical documentation. Retrieved from
What Works Clearinghouse website: http://ies.ed.gov/ncee/wwc/pdf/wwc_scd.pdf
85
Kratochwill, T. R., Stoiber, K. C., Christenson, S., Durlak, J., Levin, J. R., Talley, R., et
al. (2002). Procedural and coding manual for review of evidence-based
interventions. Task force on the Evidence Based Interventions in School
Psychology, Sponsored by Division 16 of the American Psychological Assocation
and Society for the Study of Psychology.
Leitenberg, H. (1965). Is time-out from positive reinforcement an aversive event? A
review of the experimental evidence. Psychological Bulletin, 64, 428-441.
Little, L. K., & Kelley, M. L. (1989). The efficacy of response cost procedures for
reducing children’s noncompliance to parental instructions. Behavior Therapy,
20, 525-534.
Maggin, D. M., Chafouleas, S. M., Goddard, K. M., & Johnson, A. H. (2011). A
systematic evaluation of token economies as a classroom management tool for
students with challenging behavior. Journal of School Psychology, 49, 529 – 554.
Matson, J. L. & Boisjoli, J. A. (2009). The token economy for children with intellectual
disability and/or autism: A review. Research in Developmental Disabilities, 30,
240-248.
McGoey, K. E., & DuPaul, G. J. (2000). Token reinforcement and response cost
procedures: Reducing the disruptive behavior of preschool children with
attention-deficit/hyperactivity disorder. School Psychology Quarterly, 15, 330-
343.
86
McLaughlin, T., & Malaby, J. (1972). Reducing and measuring inappropriate
verbalizations in a token economy. Journal of Applied Behavior Analysis, 5, 329-
333.
O’Leary, K. D., & Becker, W. C. (1967). Behavior modification of an adjustment class:
A token reinforcement program. Exceptional Children, 33, 637-642.
O’Leary, K. D., Becker, W. C., Evans, M. B., & Saudargas, R. A. (1969). A token
reinforcement program in a public school: A replication and systematic analysis.
Journal of Applied Behavior Analysis, 2, 3-13.
O’Leary, K. D., & Drabman, R. (1971). Token reinforcement programs in the classroom:
A review. Psychological Bulletin, 75, 379-398.
Pace, D. M., & Forman, S. G. (1982). Variables related to the effectiveness of response
cost. Psychology in the Schools, 19, 365-370.
Panek, D. M. (1970). Word association learning by chronic schizophrenics on a token
economy ward under conditions of reward and punishment. Journal of Clinical
Psychology, 26, 163-167.
Pemberton, J. R., & Borrego, J. (2007). Increasing acceptance of behavioral child
management techniques: What do parents say? Child and Family Behavior
Therapy, 29, 27-45.
Phillips, E. L. (1968). Achievement place: Token reinforcement procedures in a home-
style rehabilitation setting for “pre-delinquent” boys. Journal of Applied Behavior
Analysis, 1, 213-223.
87
Phillips, E. L., Phillips, E. A., Fixsen, D. L., & Wolf, M. M. (1971). Achievement place:
Modification of the behaviors of pre-delinquent boys within a token economy.
Journal of Applied Behavior analysis, 4, 45-59.
Reynolds, L. K., & Kelley, M. L. (1997). The efficacy of a response cost-based treatment
package for managing aggressive behavior in preschoolers. Behavior
Modification, 21, 216-230.
Rodriguez, J. O., Montesinos, L., & Preciado, J. (2005). A 19
th
century predecessor of the
token economy. Journal of Applied Behavior Analysis, 38, 427.
Schmauk, F. J. (1970). Punishment, arousal, and avoidance learning in sociopaths.
Journal of Abnormal psychology, 76, 325-335.
Siegel, G. M., Lenske, J., & Broen, P. (1969). Suppression of normal speech disfluencies
through response cost. Journal of Applied Behavior Analysis, 2, 265-276.
Skinner, B. F. (1966). Contingencies of reinforcement in the design of a culture.
Behavioral Science, 11, 159-166.
Walker, H. M., Mattson, R. H., & Buckley, N. K. (1969). Special class placement as a
treatment alternative for deviant behavior in children. University of Oregon,
Monograph No. 1
Weiner, H. (1962). Some effects of response cost upon human operant behavior. Journal
of Experimental analysis of behavior, 5, 201-208.
Winkler, R. C. (1970). Management of chronic psychiatric patients by a token
reinforcement system. Journal of Applied Behavior Analysis, 3, 47-55.
88
Wolery, M., Bailey, D. B., & Sugai, G. (1988). Effective teaching: Principles and
procedures of applied behavior analysis with exceptional students. Boston: Allyn
Bacon.
Wolf, M. M., Giles, D. K., & Hall, R. V. (1968). Experiments with token reinforcement
in a remedial classroom. Behavior research and Therapy, 6, 51-64.