Journal of Teaching in Physical Education, 2012, 31, 261-278

© 2012 Human Kinetics, Inc.

The Effects of Peer-Administered Token

Reinforcement on Jump Rope Behaviors
of Elementary Physical Education Students1
Andrew E. Alstot
Pacific University

Token economies have a long research and applied history within clinical settings
and classroom education (Kazdin, 1982). However, despite reported successes in
improving physical activity behaviors (Alstot, 2012), research examining token
reinforcement implemented specifically in physical education is virtually nonex-
istent. Therefore, the purpose of the current study was to examine the effects of
a peer-administered token economy on the jump rope behaviors of elementary
physical education students. An alternating treatments design was used to assess
the effects of the intervention. Participants were alternated between five baseline
and five token economy sessions while response differentiation between the two
phases was assessed. Results indicated that nine out of ten participants showed
an increase in the number of successful jump rope practice trials during token
reinforcement sessions as compared with baseline sessions. Based on the results
of the study, it was concluded that peer-administered token economies can be
useful tools for physical educators.

Keywords: token economy, behavior analysis, practice

Physical education research repeatedly supports the relationship between

practice and student achievement (Ashy, Lee, & Landin, 1988; Buck, Harrison, &
Bryce, 1990; Silverman, 1985). Because the link between practice and achieve-
ment is so strongly established, it is important for teachers to incorporate optimal
practice time into their classes. Therefore, the development of teaching techniques
and strategies that can be used to increase student practice within a physical edu-
cation setting should be of great value to physical educators. One tool, the token
economy, which has its foundations in behavior analysis, may be beneficial in
improving student practice.

Behavior Analysis in Physical Education

Behavior analysis, which is founded in Skinner’s (1953, 1974) Behaviorism, has a
long tradition in physical education literature dating back to Siedentop and Rushall’s

Alstot is with the Department of Exercise Science, Pacific University, Forest Grove, OR.

  261
262  Alstot

(1972) introduction of a behavioral model for motor skill acquisition. From this
point, Behaviorism became a frequently used theoretical framework within physi-
cal education and sport research (see review articles by Lee, 1993, and Ward &
Barrett, 2002). This theory views behavior as an individual phenomenon that is
controlled by stimuli extrinsic to the individual. Therefore, if one were to systemati-
cally manipulate these extrinsic stimuli, corresponding behavior changes would be
observed. By reinforcing desired behaviors and punishing undesired behaviors, a
teacher, for example, could systematically increase and decrease desirable and unde-
sirable behaviors, respectively. The development of teaching techniques grounded
in behavioral theory can provide practitioners with an applicable science to imple-
ment within physical education to improve skills and increase physical activity.
Interventions based in applied behavior analysis have been used in physical
education and sport settings to improve practice behaviors and improve a variety
of physical activity and skill-related behaviors (Donahue, Gillis, & King, 1980;
Lee, 1993; Ward & Barrett, 2002). Reinforcement, a fundamental component of
applied behavior analysis, occurs when a stimulus is presented to or removed from
an individual upon his/her engagement in a desired behavior. This consequence
increases the likelihood of the desired behavior occurring again in a similar situ-
ation (Cooper, Heron, & Heward, 2007). Several studies in physical education
and sport settings have used reinforcement principles, either as an independent
intervention or a component of a package intervention, to improve tennis skills
(Buzas & Ayllon, 1981; Allison & Ayllon, 1980), volleyball skills (Crouch, Ward,
& Patrick, 1997; Ward, Crouch, & Patrick, 1998; Ward, Smith, Makasci, & Crouch,
1998), locomotor and manipulative skills (Houston-Wilson, Dunn, van der Mars,
& McCubbin, 1997), basketball skills (Kladopoulos & McComas, 2001), football
skills (Komaki & Barnett, 1977), hockey skills (Anderson, Crowell, Doman, &
Howard, 1988), and baseball skills (Heward, 1978). Both social reinforcers, such
as-praise (Buzas & Ayllon, 1981; Allison & Ayllon, 1980), and tangible reinforcers,
such as money (Heward, 1978), have been effective to increase achievement of
motor skill-related behaviors. Several behavior analysis-based studies also found
peer-mediated accountability, which included peer-administered assessments and
public postings as well as additional physical activities as reinforcers, increased
the number of students’ practice trials in physical education (Crouch et al., 1997;
Ward, Crouch, & Patrick, 1998; Ward, Smith, Makasci, & Crouch, 1998).

Token Economies
The token economy, which may be useful to physical education practitioners, is
a tool derived from the field of behavior analysis. First introduced by Ayllon and
Azrin (1968), the token economy has been shown to be useful across a variety of
settings to improve numerous behaviors, including academic behaviors, such as class
participation (Boniecki & Moore, 2003), math homework completion (McGinnis,
Friman, & Carlyon, 1999), and reading behaviors (Ayllon & Roberts, 1974), as
well as physical activity behaviors (e.g., Alstot, 2012; Bennett, Eisenman, French,
Henderson, & Schultz, 1989; DeLuca & Holborn, 1985, 1990, 1992; Mangus,
Henderson, & French, 1986), among others. Having been described as “one of the
most widely used and effective behavior management strategies to evolve from
behavior analysis” (McKenzie, 1979, p. 102), token economies are most often
 Token Economy in Elementary Physical Education   263

reinforcement-based systems, consisting of three major segments: (a) specifically

defined target behavior(s); (b) tokens or points to be awarded to participants upon
their engagement in the target behavior(s); and (c) back-up reinforcers for which
participants can exchange their tokens (Cooper et al., 2007). More specifically, when
a student engages in the desired behavior, he/she is rewarded with a token. After a
specified amount of time or amount of tokens earned, each student can exchange
his/her tokens for a variety of reinforcers (e.g., tangible prizes, edible rewards,
privileges, etc.); these are referred to as “back-up reinforcers.” When implemented
properly, a token economy can be “…a system that will allow teachers to be effec-
tive” (Stainback, Payne, Stainback, & Payne, 1973, p. 4).
Specifically within physical education and physical activity settings, the use of
a token economy can be an effective tool with several advantages. First, the educa-
tional process does not have to be stopped to administer reinforcement; tokens or
points can be administered quickly with little interruption to the practice or game
activities of the students (Rushall & Siedentop, 1972). Second, the students’ earned
tokens can be used to purchase from a selection of back-up reinforcers, minimizing
the chances of satiation on an individual reinforcer (Rushall & Siedentop, 1972).
Finally, tokens can be administered immediately following engagement in the
target behavior. It may not always be convenient to administer tangible reinforc-
ers during physical activity. Therefore, the use of token reinforcement delays the
presentation of tangible reinforcers until a more convenient time (Lavay, French,
& Henderson, 2006).
Token economies have proven useful across several physical activity settings to
improve a variety of exercise and activity behaviors, such as pedaling a stationary
bike (Bennett et al., 1989; DeLuca & Holborn, 1985, 1990, 1992), walking distance
(Wiggam, French, & Henderson, 1986), one-mile walk/jog times (Trocki-Ables,
French, & O’Connor, 2001), time spent exercising (Bernard, Cohen, & Moffett,
2009), pole vaulting (Brock, Brock, & Willis, 1972), and attentive behavior during
physical activity (Reitman, Hupp, O’Callaghan, Gulley, & Northup, 2001). The
range of populations addressed in these studies varied greatly as well. Token
economies have been shown effective in improving the physical activity behaviors
of children with ADHD (Reitman, et al., 2001; Trocki-Ables et al., 2001), adult
women with Down Syndrome (Bennett et al., 1989), children with Cystic Fibro-
sis (Bernard et al., 2009), senior citizens (Wiggam et al., 1986), and obese and
nonobese boys (DeLuca & Holborn, 1985, 1990, 1992). Unfortunately, few of the
aforementioned token economy studies that address physical activity behaviors have
been conducted in school settings, while even fewer have been done specifically
within physical education classes.
Although token economy systems have been shown to be effective in improving
physical activity behaviors of individuals across a variety of settings and behaviors
and have been recommended for use in physical education (Lavay et al., 2006;
Rushall & Siedentop, 1972), there is little empirical support for the beneficial
effects of token economies on typically developing students administered by teach-
ers in physical education settings. McKenzie (1979) stated, “despite the increased
number of reported successes and the wide acceptance of token economy systems
in special and regular classrooms…token economy applications remain relatively
unnoticed by practitioners and researchers in sport and physical education” (p.
110). Despite that statement being presented more than 30 years ago, paired with
264  Alstot

the recommendations for using token systems (Lavay et al., 2006; Rushall & Sie-
dentop, 1972), as well as the reported empirical successes in improving physical
activity behaviors outside of physical education (Bernard et al., 2009; DeLuca &
Holborn, 1985, 1990, 1992; Reitman et al., 2001), little token economy research
has been conducted in school settings, specifically within physical education.
Therefore, the purpose of the current study was to examine the effect of a token
economy administered by an elementary physical education teacher on the number
of successful jump rope practice trials performed by typically developing students
within physical education.

Method
Participants
Participants were selected from one intact third grade physical education class.
Third grade was chosen as the target population due to the students’ developmental
level as related to the nature of the jump rope task being examined; the task was
one that necessitated persistence and practice in order for children as young as
third grade to show improvement. Upon approval from the university institutional
review board, informed consent was requested from the teacher and from each
student’s parent or legal guardian; informed assent also was obtained from each
participating student. The physical education teacher was then given a document
describing the Generic Levels of Skill Proficiency (Graham, Holt/Hale, & Parker,
2007) and asked to characterize each student’s skill level as one of the following
levels: precontrol, control, or utilization (Graham et al., 2007) based on the criteria
found in the document. Ten students (i.e., 5 girls and 5 boys), from a class of 20,
identified by their physical education teacher to be at the precontrol, control, or
utilization levels were selected from one class to serve as participants. The study was
originally intended to target only low skilled students (i.e., precontrol or control), the
population that is in greatest need of skill improvement; however, too few potential
participants were identified as low skilled. Therefore, three students characterized
as at the utilization level were added as participants. Table 1 describes participants’
characteristics, including demographic information (i.e., gender and age) and skill
level as determined by the teacher. Pseudonyms were used for all participants.

Setting and Personnel

The study was conducted in an elementary school located in a suburban city in
southeastern United States. All sessions took place in the school’s gymnasium
during third grade physical education classes with the exception of session two,
which was conducted outside on the school’s outdoor tennis courts. The physical
education classes usually met once a week for approximately 45 min (occasion-
ally, depending on the school’s rotating schedule, the class met twice in a single
week). Each of the 10 sessions included in the study occurred over a two-month
period. Personnel included the physical education teacher (female, with eight years
of physical education teaching experience) who taught the class and implemented
the intervention and the researcher who was present during all sessions to col-
lect pertinent data. The presence of the researcher did not appear to influence the
 Token Economy in Elementary Physical Education   265

Table 1  Participant information and Generic Level of Skill

Proficiency as Rated by the Teacher
Participant Gender Age Skill Level
Allison Female 8 Control
Kendra Female 8 Control
Carrie Female 8 Control
Wendy Female 8 Utilization
Carla Female 8 Precontrol
Eddie Male 9 Utilization
Daniel Male 8 Utilization
Levi Male 8 Precontrol
Isaiah Male 9 Precontrol
Doug Male 8 Precontrol
Note. Generic Level of Skill Proficiency was rated for each participant by the teacher before the onset
of the study.

participants’ behavior. The school is located in close proximity to a university; the

students are, therefore, accustomed to frequent observers in classes.

Data Collection and Equipment

All ten sessions were recorded using two video cameras; a Sony Handycam DCR-
SR47 digital video camera served as the primary recording device while a Kodak
Zi6 Pocket video camera served as a backup. Additional equipment included jump
ropes for each student in the class, tokens (i.e., 3/4 inch round plastic “Bingo” chips)
to be distributed during the intervention, personalized token containers (i.e., 4 inch
square plastic boxes which included removable lids with a one inch round holes
for easy token entry) to store each student’s earned tokens, and a menu of back-up
reinforcers (e.g., stickers, yo-yos, balls, erasers, glow sticks, etc.) for which tokens
were exchanged.

Experimental Design and Procedures

The dependent variable of interest was successful jump rope practice trials. The
jump rope skill was selected due to the nature of performing a jump rope task;
persistence is needed to improve the jump rope skill. The study was designed to
examine the reinforcing effects of a token economy on students’ successful jump
rope practice trials. Jumping rope is a skill that may be difficult for some children
without adequate practice.
Behavioral theory views behavior as an individual phenomenon; the use of
this perspective necessitates the use of a single-subject research design (Kennedy,
2005). Therefore, an alternating treatments design (Barlow & Hayes, 1979) was
used to assess the effects of the intervention. Alternating treatments designs switch
between treatment conditions to assess whether levels of the target behavior are
266  Alstot

different under the varying treatments. The two conditions implemented in this
study were baseline and token economy conditions. A functional relation can be
determined if response differentiation occurs between the two conditions (Kennedy,
2005). Barlow and Hayes provided a rationale for utilizing alternating treatment
designs by citing two advantages: it does not require the researchers to completely
withdraw the treatment (as one would do with a reversal design) and a comparison
between the treatments can be made very quickly.
Before the beginning of each day’s class, the researcher randomly selected one
of the conditions (i.e., baseline or token economy) to be implemented. In total, five
baseline and five token sessions were completed. The entire class participated in the
condition, although videos were only recorded of the participants. The students did
not have prior knowledge of the experimental condition that was to be presented
each day. Upon entering the physical education class, the teacher informed the
students which condition was to be performed during that class period. On occa-
sion, to increase the number of sessions included in the study, two sessions were
conducted during the same day, one at the beginning of the class period and one
at the end. This occurred two times throughout the duration of the study: sessions
three and four and sessions seven and eight. On these occurrences, the researcher
randomly selected which condition was to be implemented at the beginning of
the class. The opposing condition was then implemented at the end of the class
session. For example, the researcher flipped a coin to determine which condition
was implemented at the beginning of the class. If a baseline session was selected,
it was implemented following the procedures described below. Upon its comple-
tion, the teacher conducted a condensed version of her usual class (i.e., from the
instructional unit the teacher was implementing, such as throwing or gymnastics;
not necessarily lessons on jumping rope). Then, during the final 10–15 min of class
period, the teacher conducted another session, this time a token economy session.
Baseline.  The teacher had previously taught several lessons on performing the
basic forward jump rope skill before the onset of the study; all participants received
instruction on the proper way to perform the skill. Upon the commencement of the
study, all students in the class were divided into pairs by the researcher (although
data were collected on the selected participants only). While one member of each
pair performed a jump rope skill, the partner counted the total number of jumps
(before the first baseline session, the teacher gave thorough instructions on how
to properly count jumps). The teacher prompted the students to begin jumping
and cued them to stop after 30 s; a 30 s rest period then was given (i.e., one cycle,
consisting of 30 s of jump rope and 30 s of rest, for a total of one minute). Each
participant continued for five cycles (i.e., 5 min total) before switching roles with
their partners. Once the partners switched roles (i.e., one jumping while the other
counts jumps) the process was repeated, giving each partner a chance to jump for
a complete 5 min session. Each baseline session took approximately 10–15 min.
Token Economy.  During the intervention phase, the participants followed a
similar procedure as was followed during the baseline sessions, with the addition
of the administration of token reinforcement. During the 30-s rest period described
above, the partner rewarded the participant with tokens based on the number of
practice trials he/she performed. Partners were instructed to give one token for
every ten times the participant swung the rope from behind his/her body over-
 Token Economy in Elementary Physical Education   267

head to the front of the body and attempted to jump over the rope with both feet
(e.g., 20 jump attempts yielded 2 tokens, 35 jumps yielded 3, 48 jumps yielded
4, etc.). During the 30 s activity period, the partner counted total practice trials.
When the 30 s rest period began, the partner retrieved the appropriate amount of
tokens from a central location (i.e., a large container, located on the gym floor,
which held an ample amount of tokens) and deposited them into the participant’s
token container which was located on the ground near where the participant was
jumping rope. Dropping plastic tokens into a plastic container provided each par-
ticipant with multiple stimuli (i.e., visual and auditory) to be associated with token
reinforcement.
Throughout the duration of the study, the participants did not retain the same
partner. Pairs were changed four times by the researcher, resulting in each partici-
pant having five different partners throughout the study.
Students had an opportunity approximately once a week to exchange their
tokens for a variety of back-up reinforcers at the “store.” The teacher requested
not to conduct a reinforcer preference assessment before stocking the token store.
In lieu of the preference assessment, the teacher suggested items that she thought
would be of interest to the participating students. The store consisted of four contain-
ers that each held items of different value: 10, 15, 20, or 30 tokens. Smaller items
(e.g., small stickers and erasers) were the least expensive, costing only 10 tokens,
while larger items (e.g., yo-yos and glow sticks) were the most expensive, costing
30. Students had the option to spend their tokens at the store or to save until a later
date, accumulating tokens to exchange them for more “expensive” reinforcers.
Teacher Training.  Before implementation of the intervention, the researcher
trained the physical education teacher on all procedures of the study. The teacher’s
competency of the procedures of the study was assumed when the teacher was able
to fully describe the steps in the implementation of the components of the study
to the researcher without error. Teacher training lasted approximately 30 min.
During the study, the researcher was present for all sessions, available to answer
the teacher’s questions as well as correct any errors in the implementation of the
study’s procedures.
Token Training.  Before the intervention was introduced, the researcher and
teacher conducted a short token training session with the students, lasting
approximately 20 min. According to Cooper et al. (2007), initial token training
with high-functioning children can primarily consist of verbal instructions and
modeling. Therefore, the researcher explained to the class how they were able to
earn tokens and showed them what could be purchased with their earned tokens;
the teacher modeled to them how tokens were to be earned and distributed.
Social Validity.  Upon the conclusion of the intervention, the baseline and inter-
vention data were shown to the physical education teacher. Then, a questionnaire
designed by the researcher was administered to inquire of her perception of the
effectiveness of the intervention as well as her opinion regarding the feasibility
of using token economies in physical education. The questionnaire also addressed
the cost of implementing a token economy, the teacher’s future intentions of using
a token economy, and an open-ended response area for the teacher to share any
additional thoughts she had on the study. See the Appendix for the social validity
questionnaire that was completed by the teacher.
268  Alstot

Data Analysis
Data from all sessions were analyzed using the recordings from the primary camera
with the exception of the third session, which necessitated the use of the backup
camera due to a user error in the recording process. For each session of the study,
videos were viewed in slow motion and the number of successful and unsuccessful
jump rope practice trials were recorded for each participant. A jump rope practice
trial was coded as “successful” if each of the following elements of the skill were
performed: (a) hands holding the handles of the rope on each side of the body, (b)
jump rope starts behind the body, (c) rope will swing in a circular motion above
head with rope ending in front of the body, and (d) rope passes under both feet. A
trial was recorded as “unsuccessful” if one or more of the above criteria were not
properly performed. After each trial was coded as successful or unsuccessful; the
total numbers of successful and unsuccessful trials were then recorded for each
session for each individual participant.
Data analysis was ongoing throughout the duration of the study. The study
was ceased when the two conditions showed a consistent and stable sequence as
evident in a visual analysis of the graphs.
Interobserver Agreement.  Interobserver agreement (IOA) was assessed for
approximately 30% the sessions (Cooper et al., 2007). Using video tape data, a
trained independent observer coded each practice trial of a session as successful or
unsuccessful. Percentage agreement was calculated by dividing the total number
of agreements by the total number of agreements plus disagreements and multi-
plying by 100%. Overall agreement was 95.9%, ranging from 87.8% to 99.0%.
Treatment Integrity.  A treatment integrity measure of the independent variable
was conducted. For each token session, the researcher calculated the absolute
percent error (APE) of token distribution to determine the accuracy of treatment
administration. APE was calculated by subtracting the criterion amount (i.e., how
many tokens the participant should have received for the session) from the actual
amount (i.e., how many tokens the participant actually received), dividing by the
criterion amount and multiplying by 100. The resulting APE represents the percent
error with which tokens were administered to each participant for the given session.
Across all sessions, only one participant, Levi, was given tokens with less
than 80% accuracy (i.e., APE higher than 20%), while most participants received
tokens with more than 90% accuracy (i.e., APE less than 10%). Figures 1, 2, and
3 show the APE (i.e., the amounts inside the parentheses) for each participant for
each token session. Table 2 shows the mean absolute percent error (MAPE), that
is, the mean of all sessions’ APE amounts per participant, and the APE across all
sessions (i.e., the total APE if all token sessions were combined).

Reults
Successful Jump Rope Practice Trials
The influence of the token economy intervention on the number of successful jump
rope practice trials is indicated in the line graphs in Figures 1, 2, and 3. Response
differentiation between baseline and token sessions is evident in nine out of ten
 Token Economy in Elementary Physical Education   269

Figure 1 — Number of successful jump attempts across all sessions and success rate (%) of
jump attempts by session type for participants rated at the Pre-Control level. Note. Numbers
in parentheses () represent the absolute percent error (APE) of token distribution per session.

participants. Carrie, who was classified as control level, was the only student
whose data did not indicate differences between the two conditions. With this
lone exception, differences were present in the mean number of successful jumps
during baseline sessions and token sessions for all participants. Kendra, Allison,
and Wendy each increased their mean number of jumps per session by more than
50 (M = 63.40, 56.92, and 53.25, respectively) during token sessions as compared
with the baseline condition. Isaiah, Daniel, and Eddie improved their mean by over
30 jumps (M = 37.20, 34.60, and 33.05, respectively) while Carla improved by
21.80 jumps per session, Doug added 17.00 jumps during token sessions, and Levi
increased by 10.85 jumps. Carrie showed only a minimal change by increasing by
1.60 jumps per session.
270  Alstot

Figure 2 — Number of successful jump attempts across all sessions and success rate (%)
of jump attempts by session type for participants rated at the Control level. Note. Numbers
in parentheses () represent the absolute percent error (APE) of token distribution per session.

Graphical Trends
Trendlines for each condition (i.e., baseline and token economy) were calculated
using the split middle technique (Gast, 2010) and added to the line graphs in
Figures 1, 2, and 3. Four distinct patterns were present in the graphical analysis.
First, Levi, Wendy, and Carla’s graphs each showed both baseline and token data
to be trending upward, while still maintaining level differentiation between the
two conditions. Second, Eddie, Daniel, and Doug’s token condition data trended
upward, while their baseline levels were falling. Again, response differentiation
between conditions was present. Third, although notable differences between con-
ditions existed, Isaiah and Allison showed a decreasing trend in both baseline and
token conditions. Finally, Kendra’s baseline was falling while the token economy
trendline remained relatively unchanging.

Success Rate
There were noteworthy differences in the success rate between conditions as well.
Success rate was calculated for each condition by dividing the total number of
successful jumps by the total number of jump attempts and multiplying by 100.
 Token Economy in Elementary Physical Education   271

Figure 3 — Number of successful jump attempts across all sessions and success rate (%) of
jump attempts by session type for participants rated at the Utilization level. Note. Numbers
in parentheses () represent the absolute percent error (APE) of token distribution per session.

The bar graphs presented in Figures 1, 2, and 3 indicate an improvement in the

success rate of nine out of ten participants; Carrie was the sole exception. Of the
participants who showed an increased success rate, the improvement ranged from
a 2.00% increase (i.e., Wendy) to a 9.00% improvement (i.e., Kendra). In addition,
participants showed an average increase of 33 successful jumps per token economy
session when compared with baseline sessions.

Social Validity
The physical education teacher’s responses to the social validity questionnaire
were consistently positive. Her perception was that the token economy was “very
effective” in helping her students learn jump rope skills while stating the imple-
mentation of the token economy was “very easy.” The researcher calculated the
total monetary cost of operating the token store (i.e., less than approximately 75
cents per student across a two month period); the teacher suggested on the ques-
tionnaire that she thought the positive effects of the intervention were worth the
cost of its implementation. However, she was only “somewhat likely” to operate a
token economy in her future classes due to the “budget.”
272  Alstot

Table 2  Token Administration Accuracy

Actual # # Tokens
Total # Tokens Should Have APE Across
Tokens Received Received MAPE all Sessions
Participant Received per Session per Session (%) (%)
Allison 108 27.00 22.50 20.84 20.00
Kendra 93 18.60 18.40 5.16 1.09
Carrie 104 20.80 19.00 11.29 9.47
Wendy 125 25.00 23.40 10.06 6.84
Carla 48 9.60 9.20 25.33 4.35
Eddie 85 17.00 16.40 6.26 3.66
Daniel 83 20.75 20.50 10.80 1.22
Levi 22 4.40 6.20 30.71 29.03
Isaiah 108 21.60 19.80 12.83 9.09
Doug 32 6.40 7.20 28.83 11.11
Note. MAPE = mean absolute percent error; APE = absolute percent error

Discussion
The primary objective of this study was to examine the effectiveness of a token economy
on successful jump rope practice trials performed by typically developing third grade
students in a physical education class. The results indicated the token economy had
a positive effect on the number of successful jumps as compared with baseline levels
in nine out of ten participants. The only exception, Carrie, did not show response
differentiation between baseline and token economy sessions. By definition, positive
reinforcement occurs when a stimulus is presented after engagement in a behavior
and the presentation of the stimulus increases the frequency with which the behavior
occurs again in similar circumstances (Cooper et al., 2007). Because Carrie’s baseline
and token economy session responses did not differ, it can be assumed that the tokens,
and subsequently the back-up reinforcers available at the token store, did not have
reinforcing properties for Carrie. This reflects one of the major theoretical founda-
tions of behaviorism—that behavior is an individual phenomenon (Skinner, 1953).
What is reinforcing to one student may not be reinforcing to another. In this case, the
teacher requested to forego reinforcer sampling, which would have specified items
of interest for each student to be included in the token store. Further inquiry into
Carrie’s specific reinforcer preferences may have yielded different results; if the
store was stocked with items of particular interest to Carrie, she may have also
increased the number of responses during token sessions. Teachers, when instructing
a class of numerous individuals, should understand that what controls one student’s
behavior may not impact another student in the same way, due to the individualistic
nature of reinforcement. However, the results of the study indicated that nine of the ten
participants showed an increase in successful jumps. For these remaining participants,
the back-up reinforcers available in the token store provided some reinforcement for
the engagement in jump rope practice trials, thus, an increase in the number of jumps.
 Token Economy in Elementary Physical Education   273

Because the literature consistently supports the relationship between practice

and achievement in physical education (Ashy et al., 1988; Buck et al., 1990; Sil-
verman, 1985), it is important to examine tools that improve and increase practice
time in the gym; the current study reveals token economies to be one of these tools.
Buck et al. (1990) demonstrated a relationship between achievement and correctly
performed practice trials where the outcome of the trial was successful (i.e., the
process and outcome were both successful). In the current study, the operational
definition for a “successful” jump attempt included both a correctly performed
process as well as a successful outcome. The token economy sessions produced
a higher number of successful jump attempts as compared with baseline sessions
(i.e., an average increase of approximately 33 successful jumps per session). Across
the five token sessions in the current study, the average student increased their
amount of successful trials by nearly 165 when compared with sessions without
token reinforcement. Based on this finding, one could assume that if a teacher
implemented a token economy for 10 days during jump rope sessions, the average
student may produce 330 more successful trials than if the unit was taught without
the token economy.
Results also indicated that participants from all initial skill levels (i.e., Pre-
Control, Control, and Utilization) increased their number of successful jumps when
compared with baseline. There appears to be, however, a difference in the amount of
increase during token sessions based on skill level. Higher skilled students tended to
have a greater increase during token sessions, which may be attributed to their abil-
ity level; Silverman (1993) found that higher skilled students performed more total
trials during physical education. Although higher skilled participants increased by a
greater amount, Figure 1 shows that lower skilled students improved performance
as well. In addition, sessions that included token reinforcement also produced more
efficient practice sessions (i.e., success rate tended to be higher when receiving
token reinforcement). If success rate is used as the achievement measure, it can
be inferred that a relationship between the increase in the number of successful
jump attempts and student achievement (i.e., success rate) exists in the current data
as well. This supports the literature showing that successful performance during
practice produces greater achievement (Silverman, 1985, 1990). The results also
indicate that token economies have the potential to positively influence students’
practice within physical education settings, regardless of initial skill level, thereby
increasing student achievement.
This study supports Rushall and Siedentop’s (1972) arguments for the use
of token systems in physical education settings. First, tokens were administered
during built in rest periods. The administration of the token reinforcement did not
interrupt any educational process or skill practice; students were reinforced without
any interruption to the planned class activities. Second, because the tokens could
be exchanged for a wide variety of items, a general decrease in jump responses
during token sessions was not observed. Results demonstrated that the choices
available in the token store provided enough variety in reinforcement to continue
to result in an increase in jump rope trials throughout the study. Two participants,
however, showed a downward trend during token sessions, both of which may be
explained by influences outside of the control of the study. During Allison’s last
token session (i.e., session 10), she wore large boots to class; she jumped for the first
part of the session while wearing these boots, but removed them and finished the
274  Alstot

session in socks. If this data point is removed, Allison’s overall trend is increasing.
Isaiah’s last two token sessions (i.e., sessions 8 and 10) were notably lower than
his previous token sessions. During these sessions, the researcher observed Isaiah
struggle with the length of the jump rope he chose to use. If a jump rope of proper
length was used, Isaiah’s results may have been different. These two participants’
downward trends may be anecdotally explained away. The results, however, still
provide enough evidence to support Rushall and Siedentop’s claims.
A third argument for the use of token economies in physical education settings,
presented by Lavay et al. (2006), states that tangible reinforcers are not always con-
venient to administer during physical activity sessions; tokens provide a way to delay
reinforcement until a more convenient time. Again, the current study supports
this rationale. Tokens were administered without any interruption of learning
activities. However, if tangible reinforcers were given during activity time, it
may have inhibited the learning environment. The token store was opened after
the learning activities were completed, thus delaying tangible reinforcement
until a more convenient time that did not disrupt the academic environment. In
addition to these three reasons for using token systems in physical education,
the results of the current study provide evidence for one additional rationale.
A token economy that includes an assortment of back-up reinforcers serves a
wide variety of individuals. Not only will a variety of reinforcers reduce satiation
within a singular student (Rushall & Siedentop, 1972), but it will also provide a
greater chance that more students will find an item with individually reinforcing
properties, therefore servicing the wide range of individuals found within a single
physical education class.
Another facet of the current study involves the accuracy with which tokens
were administered by peers. Ward and colleagues (Crouch et al., 1997; Ward,
Crouch, & Patrick, 1998; Ward, Smith, Makasci, & Crouch, 1998) conducted a
series of studies examining peer-mediated accountability, which used peers to
assess performance. Students’ assessments were found to be reasonably accurate
(i.e., above 80% accuracy across all three studies). Mangus et al. (1986) also used
peers to administer reinforcement and found they provided token reinforcement
with greater than 90% accuracy. The current study supports these findings in that
student peers were able to administer token reinforcement with a relatively high
degree of accuracy. The mean absolute percent error (MAPE) across all participants
was 16.21%, indicating the participants in the study administered reinforcement
with nearly 84% accuracy. As with the aforementioned research, before the onset
of the current study, peers were trained to properly administer reinforcement.
However, training sessions took very little time and were completed in less than
one class period. Therefore, when analyzing the time it took to train the students,
the time saved by having students reinforce each other, and the accuracy with
which tokens were administered, peer-administered reinforcement appears to be a
feasible and reasonable option for physical education teachers to implement into
their instructional activities.
One of the potential barriers to the incorporation of a token economy is related
to administrative issues in its implementation (Kazdin, 1982). This study, however,
showed that the token economy in a physical education setting was implemented
with relative ease. The teacher indicated on the social validity questionnaire that
the token system was “very easy” to execute. Initial token and peer training took
 Token Economy in Elementary Physical Education   275

some effort on the part of the teacher and researcher, but by the second session the
token system basically managed itself, with students taking responsibility for
administering the intervention. Although operation of the token system took
approximately 15 min or less per session, it can be assumed that implementation
in an actual physical education class would require less time if the nuances of
data collection and research-related procedures were removed. Nevertheless,
when asked if she would use a token system in her class in the future, the teacher
indicated she was only “somewhat likely,” noting the cost of the token store
may have been too high, despite her stated belief that the benefits of the token
system were worth its cost. She was, however, willing to attempt another token
economy using free back-up reinforcers, such as line leader privileges and free
choice time. Overall, the teacher perceived the token economy as positive, stating,
“the token economy study seemed not only to improve student participation, but
also helped to motivate the students. They were eager to ask me if the following
PE day would be a ‘token’ day.”
The main limitation of the current study is related to the issue of external
validity. From a behavior analysis perspective, individuals behave, whereas groups
of people do not. Due to this theoretical perspective, behavior analysis research is
typically conducted using single subject design (Kennedy, 2005). Because of this,
the external validity of the study is restricted. However, proponents of single subject
design advocate for a strengthening of the external validity of findings through
direct and systematic replication of the study. Therefore, unless the current study
is replicated, generalization should be considered with caution.
Further research should be conducted in the area of token economies in physical
education in a variety of grades and with more teachers. In addition, more areas of
potential token economy in physical education research became evident through
the conducting of the current study. First, the physical education teacher agreed
that the use of inexpensive reinforcers (i.e., privileges, choices, etc.) may be more
feasible than continually purchasing items to keep the token store stocked. How-
ever, this needs to be examined more closely before its widespread use. Second,
the use of tangible tokens and token containers may be too cumbersome for some
teachers. Less intrusive methods of implementing a token economy, such as using
points or check marks on a poster board, instead of tangible tokens dropped into
personalized containers should be examined. It may be an even easier method to
incorporate token reinforcement in physical education, if it is deemed effective.
Finally, it is unknown if the positive effects of the token economy lasted beyond
the duration of the study or if the benefits associated with token reinforcement were
due to a novelty effect associated with the relatively short data collection period;
further studies should examine the long-term effects of token reinforcement on
jump rope behaviors.
This study’s results reveal the implementation of a token economy system in
a physical education class was effective in increasing the number of jump rope
practice trials in nine out of ten third grade participants. These results indicate
that token reinforcement can positively influence motor behaviors within a
physical education setting as well as distinctly improve the total number of
successful trials across the entire unit of instruction. The current study provides
evidence of the potential of token systems to be an effective and feasible tool for
physical educators.
276  Alstot

Note
1. This study was completed as part of the author’s doctoral work at Middle Tennessee State
University.

References
Allison, M. G. & Ayllon, T. (1980). Behavioral coaching in the development of skills in
football, gymnastics, and tennis. Journal of Applied Behavior Analysis, 13, 297, 314.
Alstot, A.E. (2012). Implications for the use of token economies in physical education: A
literature review. Physical and Health Education Academic Journal, 4(1). Retrieved
May 7, 2012, from http://ojs.acadiau.ca/index.php/phenex/article/view/1449.
Anderson, D.C., Crowell, C.R., Doman, M., & Howard, G.S. (1988). Performance posting,
goal setting, and activity-contingent praise as applied to a university hockey team. The
Journal of Applied Psychology, 73, 87–95. doi:10.1037/0021-9010.73.1.87
Ashy, M.H., Lee, A.M., & Landin, D.K. (1988). Relationship of practice using correct
technique to achievement in a motor skill. Journal of Teaching in Physical Education,
7, 115–120.
Ayllon, T., & Azrin, N. (1968). The Token Economy: A Motivational System for Therapy
and Rehabilitation. New York: Appleton.
Ayllon, T., & Roberts, M.D. (1974). Eliminating discipline problems by strengthening
academic performance. Journal of Applied Behavior Analysis, 7, 71–76. doi:10.1901/
jaba.1974.7-71
Barlow, D.H., & Hayes, S.C. (1979). Alternating treatments design: One strategy for com-
paring the effects of two treatments in a single subject. Journal of Applied Behavior
Analysis, 12, 199–210. doi:10.1901/jaba.1979.12-199
Bennett, F., Eisenman, P., French, R., Henderson, H., & Shultz, B. (1989). The effect of a
token economy on the exercise behavior of individuals with Down Syndrome. Adapted
Physical Activity Quarterly, 6, 230–246.
Bernard, R.S., Cohen, L.L., & Moffett, K. (2009). A token economy for exercise adherence
in pediatric Cystic Fibrosis: A single-subject analysis. Journal of Pediatric Psychology,
34, 354–365. doi:10.1093/jpepsy/jsn101
Boniecki, K.A., & Moore, S. (2003). Breaking the silence: Using a token economy to
reinforce classroom participation. Teaching of Psychology, 30, 224–227. doi:10.1207/
S15328023TOP3003_05
Brock, S., Brock, D., & Willis, J. (1972). The effect of tangible and token rewards on the pole
vaulting behavior of high school students. School Applications of Learning Theory, 4, 32–37.
Buck, M., Harrison, J.M., & Bryce, G.R. (1990). An analysis of learning trials and their
relationship to achievement in volleyball. Journal of Teaching in Physical Education,
10, 134–152.
Buzas, H.P., & Ayllon, T. (1981). Differential reinforcement in coaching tennis skills.
Behavior Modification, 5, 372–385. doi:10.1177/014544558153006
Cooper, J.O., Heron, T.E., & Heward, W.L. (2007). Applied Behavior Analysis (2nd ed.).
Upper Saddle River, NJ: Pearson.
Crouch, D.W., Ward, P., & Patrick, C.A. (1997). The effects of peer-mediated accountability
on task accomplishment during volleyball drills in elementary physical education.
Journal of Teaching in Physical Education, 17, 26–39.
DeLuca, R.V., & Holborn, S.W. (1985). Effects of a fixed-interval schedule of token reinforce-
ment on exercise with obese and non-obese boys. The Psychological Record, 35, 525–533.
DeLuca, R.V., & Holborn, S.W. (1990). Effects of fixed-interval and fixed-ratio schedules
of token reinforcement on exercise with obese and nonobese boys. The Psychological
Record, 40, 67–82.
 Token Economy in Elementary Physical Education   277

DeLuca, R.V., & Holborn, S.W. (1992). Effects of a variable-ratio reinforcement schedule
with changing criteria on exercise in obese and nonobese boys. Journal of Applied
Behavior Analysis, 25, 671–679. doi:10.1901/jaba.1992.25-671
Donahue, J.A., Gillis, J.H., & King, K. (1980). Behavior modification in sport and physical
education: A review. Journal of Sport Psychology, 2, 311–328.
Gast, D.L. (2010). Single Subject Research in Behavioral Sciences. New York: Taylor and Francis.
Graham, G. Holt/Hale, S. A., & Parker, M. (2007). Children Moving: A Reflective Approach
to Teaching Physical Education (7th ed.). New York: McGraw Hill.
Heward, W.L. (1978). Operant conditioning of a. 300 hitter? The effects of reinforcement
on the offensive efficiency of a barnstorming baseball team. Behavior Modification,
2, 25–40. doi:10.1177/014544557821002
Houston-Wilson, C., Dunn, J.M., van der Mars, H., & McCubbin, J. (1997). The effect of
peer tutors on motor performance in integrated physical education classes. Adapted
Physical Activity Quarterly, 14, 298–313.
Kazdin, A.E. (1982). The token economy: A decade later. Journal of Applied Behavior
Analysis, 15, 431–445. doi:10.1901/jaba.1982.15-431
Kennedy, C.H. (2005). Single-Case Designs for Educational Research. Boston: Allyn and Bacon.
Kladopoulos, C.N., & McComas, J.J. (2001). The effects of form training on foul-shooting
performance in members of a womens college basketball team. Journal of Applied
Behavior Analysis, 34, 329–332. doi:10.1901/jaba.2001.34-329
Komaki, J., & Barnett, F.T. (1977). A behavioral approach to coaching football: Improv-
ing the play execution of the offensive backfield on a youth football team. Journal of
Applied Behavior Analysis, 10, 657–664. doi:10.1901/jaba.1977.10-657
Lavay, B.W., French, R.W., & Henderson, H. (2006). Positive Behavior Management in
Physical Activity Settings (2nd ed.). Champaign, IL: Human Kinetics.
Lee, C. (1993). Operant strategies in sport and exercise: Possibilities for theoretical develop-
ment. International Journal of Sport Psychology, 24, 306–325.
Mangus, B., Henderson, H., & French, R. (1986). Implementation of a token economy by
peer tutors to increase on-task physical activity time of autistic children. Perceptual
and Motor Skills, 63, 97–98. doi:10.2466/pms.1986.63.1.97
McGinnis, J.C., Friman, P.C., & Carlyon, W.D. (1999). The effect of token rewards on
intrinsic motivation for doing math. Journal of Applied Behavior Analysis, 32, 375–379.
doi:10.1901/jaba.1999.32-375
McKenzie, T.L. (1979). Token economy research: A review for the physical educator. Motor
Skills: Theory into Practice, 2, 102–114.
Reitman, D., Hupp, S.D., O’Callaghan, P.M., Gulley, V., & Northup, J. (2001). The influ-
ence of a token economy and methylphenidate on attentive and disruptive behavior
during sports with ADHD-diagnosed children. Behavior Modification, 25, 305–323.
doi:10.1177/0145445501252007
Rushall, B.S., & Siedentop, D. (1972). The Development and Control of Behavior in Sport
and Physical Education. Philadelphia: Lea and Febiger.
Siedentop, D., & Rushall, B. (1972). An operant model for skill acquisition. Quest, 17, 82–90.
Silverman, S. (1985). Relationship of engagement and practice trials to student achievement.
Journal of Teaching in Physical Education, 5, 13–21.
Silverman, S. (1990). Linear and curvilinear relationships between student practice and
achievement in physical education. Teaching and Teacher Education, 6, 305–314.
doi:10.1016/0742-051X(90)90023-X
Silverman, S. (1993). Student characteristics, practice, and achievement in physical education.
The Journal of Educational Research, 87, 54–61. doi:10.1080/00220671.1993.9941166
Skinner, B.F. (1953). Science and human behavior. New York: MacMillan Company.
Skinner, B.F. (1974). About behaviorism. New York: Vintage Books.
Stainback, W.S., Payne, J.S., Stainback, S.B., & Payne, R.A. (1973). Establishing a Token
Economy in the Classroom. Columbus, OH: Merrill.
278  Alstot

Trocki-Ables, P., French, R., & O’Connor, J. (2001). Use of primary and secondary reinforc-
ers after performance of a 1-mile walk/run by boys with Attention Deficit Hyperactivity
Disorder. Perceptual and Motor Skills, 93, 461–464.
Ward, P., & Barrett, T. (2002). A review of behavior analysis research in physical education.
Journal of Teaching in Physical Education, 21, 242–266.
Ward, P., Crouch, D.W., & Patrick, C.A. (1998). Effects of peer mediated accountability
on opportunities to respond and correct skill performance by elementary school
children in physical education. Journal of Behavioral Education, 8, 103–114.
doi:10.1023/A:1022868808642
Ward, P., Smith, S.L., Makasci, K., & Crouch, D.W. (1998). Differential effects of peer-
mediated accountability on task accomplishment in elementary physical education.
Journal of Teaching in Physical Education, 17, 442–452.
Wiggam, J., French, R., & Henderson, H. (1986). The effects of a token economy on dis-
tance walked by senior citizens in a retirement center. American Corrective Therapy
Journal, 40, 6–12.

Appendix
Social Validity—Teacher Questionnaire
Please circle the most accurate response regarding your perception(s) of the inter-
vention.
1. How effective do you think the token economy was in helping students improve
jump rope skills?
Not at all Somewhat Somewhat Very
effective ineffective Neutral effective effective
2. How difficult/easy was it to implement the token economy in your class?
Very Somewhat Somewhat Very
Difficult difficult Neutral easy easy
3. Taking into consideration the total cost of implementing the token economy
and the benefits of its implementation, were the effects worth the cost?
Yes No
4. After this participating in this study, how likely are you to implement a token
economy in your class in the future?
Not at all Somewhat Somewhat Very
likely unlikely Neutral likely likely
5. Please provide any additional comments you would like to share.
Copyright of Journal of Teaching in Physical Education is the property of Human Kinetics Publishers, Inc. and
its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.