Content uploaded by Andrea Matheson
Author content
All content in this area was uploaded by Andrea Matheson on Feb 26, 2016
Content may be subject to copyright.
School Psychology Review,
2005, Volume 34, No. 2, pp. 202-2 19
Training Teachers to Give Effective Commands: Effects
on Student Compliance and Academic Behaviors
Andrea Starkweather Matheson
Iowa Headand Area Education
Agency 11
Mark D+ Shriver
University of Nebraska
Medical Center
Abstract. This study examined the effects of effective command training with teach-
ers on students’ compliance rates and academic engagement. Three target students
were selected who were exhibiting compliance rates substantially below peers.
The students’ teachers were taught how to provide effective commands. Results
indicated that students’ rates of compliance increased with increased use of effec-
tive commands. When verbal praise was added contingent on compliance, stu-
dents’ rates of compliance increased even more. In addition, academic engage-
ment was shown to increase as student compliance increased and disruptive com-
peting behaviors decreased. Implications for consultation and intervention in the
classroom to increase student compliance and academic behaviors are discussed.
Teachers are challenged to set the tone
for instruction in their classrooms and to en-
gage their students in academic lessons. It is
important that students are compliant to teach-
ers’ instructions so that available learning time
is not wasted. Managing students’ inappropri-
ate behaviors is a time-consuming task that re-
duces the amount of time teachers spend on
teaching and the amount of time students spend
on academic tasks. In urban school districts,
student behavior that is incompatible with aca-
demic responses and on-task behaviors occu-
pies 15% to 25% of class time in first through
fourth grades (Greenwood, 199 I). Rhode,
Jenson, and Reavis (1993) suggest that com-
pliance rates below 40% may prevent a child
from benefiting from instructional opportuni-
ties. Effective classroom management can help
ensure student compliance and establish and
maintain an orderly learning environment
(Doyle, 1985). Previous research has demon-
strated positive correlations between well-man-
aged classrooms and student engagement in
academic tasks, more rapid pace of progres-
sion in academic material, and higher levels
of academic achievement (Brophy, 1983;
Gettinger, 1986; Good, 1979). Strategies to
help teachers gain student compliance and in-
crease the amount of time students are engaged
in academic activities are needed. Consulting
with teachers to improve the effectiveness of
teacher commands may be one such strategy.
There is a long history of research on
the use of effective commands by parents that
demonstrates increases in child compliance
(Forehand, 1977; Forehand & McMahon,
Please address correspondence regarding this article to Mark D. Shriver, PhD, Psychology Department,
Munroe-Meyer Institute, 985450 Nebraska Medical Center, Omaha, NE 68 1985450; E-mail:
Mshriver@unmc.edu
Copyright 2005 by the National Association of School Psychologists, ISSN 02796015
202
Effective Commands
198 1; Forehand & Scarboro, 1975; Peed, Rob-
erts, & Forehand, 1977; Roberts, McMahon,
Forehand, & Humphreys, 1978; Roberts &
Powers, 1988; Schoen, 1986; Williams & Fore-
hand, 1984). Effective commands are defined
based on their effect on child behavior. In es-
sence, commands that increase the probability
of child compliance are effective whereas com-
mands that do not positively affect child com-
pliance are not effective. Previous research has
delineated some of the qualities of commands
that are important to establishing child com-
pliance (Forehand, 1977; Forehand &
McMahon, 198 1; McMahon & Forehand,
2003). In particular, Forehand and McMahon
(198 1) identified effective commands as those
that are directly stated, are specific and con-
sist of one step, are developmentally appro-
priate, are phrased positively, and are given one
at a time (e.g., there is at least a 5second wait
time between commands). Training parents to
issue effective commands to improve child
compliance is important, as parents often do
not naturally provide high rates of effective
commands relative to typically high rates of
ineffective commands (Shriver &Allen, 1997).
Unlike the research conducted in clini-
cal settings, there is much less research on the
use of effective commands in classrooms. A
few studies have examined the inclusion of
command training with teachers as part of a
multicomponent intervention package to im-
prove student behavior in the classroom
(DeMartini-Scully, Bray, & Kehle, 2000;
Musser, Bray, Kehle, & Jenson, 2001). The
command format used in these studies was de-
rived from the work of Forehand and
McMahon (1981). Other components of the
intervention included the establishment of posi-
tive classroom rules, increasing frequency of
teacher movement, a token economy or mys-
tery motivator reinforcement program, and re-
sponse cost program. The multicomponent
packages used in these studies were success-
ful in improving student behavior, but the dif-
ferential effects of command training were not
examined.
Several studies have been conducted in
classroom settings that have explored variables
that set the occasion for giving a command or
provide consequences for compliance with a
command. Most notably, eye contact and use
of reinforcement for compliance have been
manipulated. For example, Hamlet, Axelrod,
and Kuerscher (1984) demonstrated that com-
pliance rates increased in mean levels by as
much as 40% over baseline measures when eye
contact was established during instructions and
then compliance was followed by the teacher
saying, “Thank you.” Neef, Sharfer, Egel,
Cataldo, and Parrish (1983) demonstrated that
children with significant developmental dis-
abilities and behavior disorders in a self-con-
tained special education classroom did not gen-
eralize compliant responses when differentially
reinforced for compliance with “do” com-
mands relative to “don’t” commands, but that
reinforcement of both types of commands in-
creased compliance across the command types.
Ford ( 1998) investigated the effective-
ness of training teachers to issue effective com-
mands in the context of an intervention pack-
age that also included verbal praise and posi-
tive physical touch contingent on compliance.
This intervention incorporated other anteced-
ent components in the package to include
teacher proximity and establishment of eye
contact prior to issuing the command. Partici-
pants in this study included three clinic-referred
children in kindergarten and one student in first
grade. All four students’ compliance to teach-
ers’ first time instructions increased with imple-
mentation of the intervention. The extent of
the increases in mean levels of compliance
ranged from 21% to 44% over baseline mea-
sures. Verbal praise and positive physical touch
appeared to function effectively as reinforce-
ment for compliance and appropriate student
behavior.
In summary, the characteristics of effec-
tive commands have been examined in the
context of parent-child interactions, typically
in clinical research settings. Variables that af-
fect compliance when issuing commands, such
as use of teacher proximity, eye contact, and
reinforcement have been examined in class-
room settings. Training teachers to use effec-
tive commands has also been included as part
of multicomponent interventions to improve
student behavior in the classroom. However,
203
the effects of using effective commands
alone (i.e., without positive reinforcement
or establishment of eye contact or as part of
an intervention package) on student com-
pliance in a classroom have not yet been
examined. Likewise, to our knowledge the
subsequent effects on academic behaviors have
not been explored.
Two types of academic behaviors that
have received much attention in educational
research are academic engagement and aca-
demic responding. Academic engagement re-
fers to on-task and attending behaviors in the
context of classroom instructional activities
and teacher expectations for performance. Aca-
demic engagement is a moderate predictor of
academic achievement (Borg, 1980; Karweit,
1983). High rates of academic engaged time
have been identified consistently as a compo-
nent of effective instruction (Anderson, 1984;
Fisher & Berliner, 1985; Good, 1983; Karweit,
1983, 1985). Academic responding includes
the student behaviors of writing on academic
materials, answering teachers’ questions, sup-
plying a motor response to a requested task
(e.g., pointing to the picture), reading aloud or
silently, and asking questions related to aca-
demic material. Academic responding and aca-
demic engagement are behaviors predicated on
student compliance with teacher instructions.
Poor compliance to teacher directions will
likely lead to low rates of academic engage-
ment and responding. Higher rates of student
compliance may be a necessary (if not suffi-
cient) condition of academic engagement and
responding.
As an intervention, training teachers to
provide more effective commands has several
potential benefits. Teacher training in the con-
text of consultation regarding student noncom-
pliance is monetarily cost-free. The interven-
tion takes little physical effort to implement,
is not time-consuming, can be used with the
entire class of students, and is not intrusive to
the classroom environment. Due to these fea-
tures, training teachers to provide effective
commands holds a greater likelihood of accept-
ability and treatment integrity over interven-
tions that require more physical effort to imple-
ment, are more time-consuming, are specific
204
to an individual student, or are intrusive in the
classroom (Gresham, 1989).
Training teachers to provide effective
commands can take place during the consulta-
tion process when a child is referred due to
concerns regarding high rates of noncompli-
ance. Direct behavioral consultation (Watson
& Robinson, 1996) in particular is well suited
as a service delivery model, as it puts empha-
sis on changing teacher behaviors. Specific
skills training including didactic instruction,
modeling, practice, and feedback are central
to carrying out this particular consultation
model (Watson & Robinson, 1996). The cur-
rent study examined the effects of training
teachers to provide effective commands and
praise on students’ compliance rates and aca-
demic behaviors. Training incorporated ele-
ments of a direct behavioral consultation model
including didactic instruction, modeling, prac-
tice, and feedback with teachers.
Method
Participants and Setting
Three students in general education
classrooms and their teachers participated in
this study. Two children were in second grade
(Cory and Nate), and 1 child was in fourth
grade (Andy). The three regular education
teachers were selected to participate based on
the school principal’s recommendations and
teacher report of having a student with non-
compliance concerns that had not been referred
for special education. All teachers were assured
in writing that their willingness to be involved
in the study would not influence their evalua-
tions and no individually identifiable data
would be released. The teachers agreed to par-
ticipate in the two intervention training ses-
sions and permit videotaping of the classroom.
Each of the teachers had 5 or fewer years of
teaching experience. All of the teachers were
Caucasian. Nate and Andy’s teachers were fe-
male and Cory’s teacher was male.
When the principal and teacher referred
a student for participation in the study, the tar-
get students’ parents/legal guardians were
asked to give written informed consent for their
children’s participation. In addition, a letter was
Effective Commands
sent to the parents of the other children in the
target students’ classrooms alerting them that
videotaping would be performed in their child’s
classroom. A release obtaining permission to
videotape in the classroom for research pur-
poses had previously been obtained by the
school; this notice was done as a courtesy and
allowed the parents to request that their stu-
dent not be videotaped if they had objections
to the taping. Only one parent voiced concerns
regarding videotaping, and it was agreed that
the student in question would not be filmed.
Following referral and parent permis-
sion, videotaped observations were conducted
to determine if the students exhibited compli-
ance with initial teacher commands (instruc-
tions presented the first time, not the repeti-
tion of the instruction) at least 20% less than
two randomly chosen same-sex classroom
peers. Random selection of the peers occurred
when the observer (first author) chose two
same-sex peers upon entering the classroom
for a preliminary observation, but prior to any
observation or coding being conducted. No
contact with the students had occurred prior to
this observation. These preliminary observa-
tions were conducted during the same aca-
demic subject period that was subsequently
used throughout the study with the one excep-
tion of the second observation for Nate, which
occurred during spelling.
Cory was observed to exhibit compliance
with 38% of commands provided to him rela-
tive to peers’ mean compliance of 75.5%.
Nate was observed to comply with 52% of
teacher commands relative to peers’ mean
compliance of 91%. Andy was observed to
comply with 33% of teacher commands rela-
tive to peers’ mean compliance of 86%. The
referred students did appear to evidence less
compliance with teacher directions than
peers. In addition, their compliance was less
than the 60% compliance considered clini-
cally significant (McMahon & Forehand,
2003). Previous research has identified a range
of 60% to 90% compliance with parent or
teacher commands as typical (McMahon &
Forehand, 2003). Goals for child compliance
in McMahon and Forehand’s current parent
training program are 75%.
Observation times in the classroom were
chosen based on teacher report of academic
areas in which the student was not exhibiting
academic skill deficits. For Cory and Nate
(both second grade), the academic subject ob-
served was math and for Andy (fourth grade)
it was reading. Observation durations were
based upon the length of time the subject mat-
ter was presented. Observation durations for
Cory ranged from 20 to 38 min with a mean of
29.42 min. Observation durations for Nate
ranged from 17 to 42 min with a mean of 3 1.46
min. Observation durations for Andy ranged
from 28 to 66 min with a mean of 42.58 min.
Cory’s class included 23 students. Desks
were arranged in rows. There were no changes
in class size throughout the study. Nate’s class
included 24 students with desks arranged in
double rows. There were no changes in class
size throughout the study. Andy’s class had
26 students at the beginning of the study and
added 1 more before the end of the study.
The classroom desks were arranged in rows.
All of the classrooms had one teacher. The
instructional activities that took place in-
cluded lecture, independent seat work, and
small group work. These activities did not
appear to change systematically over the course
of the study, although data on this variable were
not collected.
Screenings for Instructional Placement
Curriculum-based measurement (CBM)
probes were used to examine the students’ read-
ing fluency or math fact fluency. This was to
determine if the target students were working
on materials at a level commensurate with
classroom peer performance during the time
periods in which they were observed.
Andy was observed during reading in-
struction. Reading probes from his fourth grade
curriculum were administered by the first au-
thor to determine reading fluency (scored as
total words read correctly in 1 min) and accu-
racy. Andy read 98.5% of the words accurately
at a median rate of 121 words correct per min.
Classroom peer comparisons were not obtained
for reading. Published criteria for determining
appropriate instructional levels were consulted
(Howell & Nolet, 2000; Shapiro, 1996).
205
School Psychology Review, 2005, Volume 34, No. 2
Howell and Nolett’s (2000) criteria for accept-
able performance at the fourth grade level are
95100% accuracy and 140 words correct per
min for independent reading. For fourth grade,
Shapiro ( 1996) presents guidelines indicating
an instructional level for reading to be between
70 to 100 words correct per min with four or
fewer errors, and mastery level to be above 100
words per min correct. Andy’s reading skills
were determined to be at a level appropriate
for Andy to benefit from instruction.
Cory and Nate were observed during
math instruction; all children in both of these
children’s classrooms were given math
probes consisting of math facts in the areas
of addition, subtraction, and multiplication.
The number of digits correct in 2 min was
calculated by the first author and compared
to their respective classroom norms. Cory
and Nate’s performance in addition, subtrac-
tion, and multiplication was comparable to
peer performance in each of these areas.
Cory’s digits correct in 2 min in addition
(19), subtraction (14), and multiplication
(10) was comparable to classroom peers’
median performance in each of the areas (for
addition, median = 25, range = 4 to 94; for sub-
traction, mean = 18.5, range = 7 to 49; and for
multiplication, mean = 3, range = 1 to 19).
Nate’s performance in addition (29), subtrac-
tion (7), and multiplication (0) was also com-
parable with peers’ performance in each of
these math areas (for addition, median digits
correct in 2 min = 18, range = 3 to 47; for sub-
traction, median = 12, range of 0 to 25; for
multiplication, median = 3, range = 0 to 15).
Cory and Nate’s math performance was deter-
mined to be consistent with peers’ math per-
formance in the classroom and consistent with
teacher reports.
Measures and Dependent Variables
Compliance.
Percentage of student
compliance to all teacher commands was the
primary dependent variable. This was calcu-
lated as the number of commands with which
the student complied divided by the total num-
ber of commands given by the teacher to the
student. If the student initiated compliance to
the teacher’s command within 5 s of the com-
mand being given, it was coded as compliance
to the request. If the student did not initiate
compliance to the command within 5 s, the
behavior was coded as noncompliance. Percent
compliance was chosen rather than rate of
compliance, as this is most consistent with
previous literature (McMahon & Forehand,
2003). In addition, the goal of teacher train-
ing was to increase the ratio of effective
commands relative to noneffective com-
mands, not to decrease or increase the over-
all rate of commands. Increased ratio of effec-
tive commands should lead to higher percent-
age of student compliance. However, because
the rate of commands given within a time pe-
riod may also affect student compliance, the
rate of all commands given during each obser-
vation was also monitored.
Academic behaviors.
The percentage
of time during an observation period that the
student was engaged in academic behaviors in-
cluding academic task management activities
(i.e., a measure of academic engagement) and
academic response to materials served as a
dependent variable. Specific academic behav-
iors included: raising hand; play or interacting
appropriately (approved of by the teacher);
manipulating materials; transitioning, walking/
running to a new area or station when comply-
ing (moves); nonacademic task management
(e.g., borrow pencil); ask for help (talk man-
agement); looking directly at the teacher or a
peer (pay attention); marking academic mate-
rials (writing); manipulating elements/materi-
als of an academic task (task participation);
reading aloud; reading silently; and verbaliz-
ing about academic subject/materials, instruc-
tions, or appropriate subject matter topics (talk
academic).
Competing behaviors.
Competing be-
haviors included: aggressive behaviors; disrup-
tive behaviors such as yelling, crying, loud talk,
and destruction of materials; inappropriate
verbal behavior including laughter and talk
about recess or lunch (talk inappropriate); look-
ing away from the academic task (look around);
noncompliance to the teacher’s directive or a
standing class rule; self-stimulating behaviors;
and self-abusive behaviors.
206
Teacher behaviors.
Teacher behaviors
that served as dependent variables included the
ratio of effective commands (termed alpha
commands) relative to ineffective commands
(termed beta commands) provided as well as a
rate of commands and rate of effective com-
mands provided. Teacher’s verbal praise for
student compliance and appropriate behavior
were also coded as dependent variables. Defi-
nitions for each of these behaviors are given
in the section describing Independent Vari-
ables. They are included here to acknowledge
that they should have also been an effect of
the training that was provided to teachers.
Data collection.
Participants were vid-
eotaped during reading or math instructional
periods. The target students were videotaped
for the entire period in which reading or math
instruction occurred. Math and reading instruc-
tion occurred during the same time each day
and videotaped observations occurred approxi-
mately three times per week based on the avail-
ability of the student and excluding days for
which there were schedule conflicts (e.g., as-
semblies, student pictures).
The videotapes were coded using two
systems: (a) a compliance code and (b) an aca-
demic behaviors/competing behaviors code.
Each videotaped observation was coded twice,
once with each system. The compliance
code allowed for data collection on type of
teacher command given (alpha vs. beta),
number of repetitions of command, student
compliance or noncompliance and negative
behavior, and teacher verbal praise (the
compliance coding system and definitions
are available from the second author). The
Mainstream Version of the Code for Instruc-
tional Structure and Student Academic Re-
sponse (MS-CISSAR) taxonomy of the
EcoBehavioral Assessment Systems Soft-
ware (EBASS; Greenwood et al., 1995) was
used to collect data on academic behaviors
and competing disruptive behaviors. Aca-
demic behaviors consisted of the codes for the
task management code of the MS-CISSAR tax-
onomy as well as the academic responding
codes. The specific behaviors are listed above
and coding definitions may be found in the
manual. A 15 s momentary time-sampling pro-
cedure was used to compute the overall per-
centage of time the child was engaged academi-
cally, academically responding, or exhibiting
competing behaviors.
Independent Variables and Treatment
Integrity
Participating teachers were provided two
training sessions individually by the first au-
thor. The first session was to train the teachers
in the delivery of effective commands. The
second session was to train the teachers in the
use of effective commands combined with ver-
bal praise. In the training session on effective
commands, teachers were trained to give ef-
fective (alpha) commands to students. Effec-
tive commands were defined as concise in-
structions that: (a) elicit a distinct outcome, (b)
are precise and temporally isolated, (c) are spe-
cific and direct, and (d) are given one at a time
followed by a 5 s wait period. Effective com-
mands also included the characteristics of quiet
voice tone, directive, stated positively, and
descriptive. Ineffective (or beta) commands are
commands that do not include the qualities of
alpha commands (Forehand, 1977; Forehand
& McMahon, 198 1; Peed, Roberts, & Fore-
hand, 1977).
In addition, alpha and beta commands
were differentiated by the number of verbs in
the command. If the command had only one
verb, it was defined as an alpha command (e.g.,
“Underline ‘that’ and ‘then”’ or “Put the name,
date, and heading on your paper”). If the
command had more than one verb, it was
defined as a beta command (e.g., “When
done, put your pencil at the top of your desk
and stop rolling your pencils” or “Get pa-
per, put on the heading, and write three sen-
tences”). This definition allowed for coding
of chain commands or commands with mul-
tiple steps (McMahon & Forehand, 2003). The
teachers were provided a handout on qualities
of effective commands as part of the review
during didactic instruction.
After didactic instruction on the quali-
ties of effective and ineffective commands, the
teachers were given a written list of beta com-
mands that had been compiled from the
baseline observations in their respective class-
2
School Psychology Review, 2005, Volume 34, No. 2
rooms. They were asked to verbally change the
ineffective commands to effective commands
during this training and were provided imme-
diate verbal feedback from the consultant on
their success or verbal correction and oppor-
tunity to try again. The teachers were also
shown videotapes of their own classrooms
(taped from the baseline phase of the study)
and asked to identify effective or ineffective
commands. For each effective command iden-
tified, the teacher was praised by the instruc-
tor. For each ineffective command identified,
the teacher was asked to describe what he or
she could have said in order to make the com-
mand effective. Materials utilized during the
teacher training are available from the second
author. The teacher training sessions ranged
from 36 min to 48 min in length.
In the second training segment, effec-
tive commands with verbal praise, teachers
were trained to provide verbal praise for ap-
propriate classroom behaviors that occurred at
any time and verbal praise contingent on com-
pliance with teacher commands, while continu-
ing to give commands in the format taught
during the first training session. Verbal praise
was defined as the teacher giving a positive
verbal statement (e.g., “You are doing a nice
job”) or a statement explicitly related to stu-
dent behavior (e.g., “I like the way you are sit-
ting quietly”) to the target student or a group
that included the target student within 5 s of
child compliance or when the student was ob-
served to be displaying appropriate classroom
behavior. A specific schedule of praise was not
required or asked of the teachers. They were
given a general goal of providing at least 10
praise statements during each academic activ-
ity. Training on the use of praise included a
brief explanation that many student behaviors
are motivated by social attention. The teacher
was then asked to identify his or her typical
response to various classroom behaviors and
self-evaluate whether appropriate or inappro-
priate behaviors typically resulted in giving
attention to students.
This self-evaluation was followed by an
explanation that if only great behavior (“great
behavior” being infrequent, atypical positive
behavior exceeding the norm, such as volun-
teering to skip recess in order to clean the ger-
bil cage, compared with more frequent typical
behavior that is normative or expected of stu-
dents such as sitting in one’s seat) is praised,
the number of times a student receives atten-
tion for inappropriate behavior will outnum-
ber the times attention is received for appro-
priate behavior. Reasons to praise, types of
praise, and benefits of labeled and unlabeled
praise were also discussed. Videotaped ex-
amples of praising/not praising students were
shown from videotapes collected during pre-
vious study phases, and the teachers were asked
to practice identifying the students’ acceptable
behaviors that warrant praise. They were pro-
vided with immediate verbal praise for suc-
cess as well as corrective feedback and mod-
eling from the consultant during training. Ma-
terials utilized for this teacher training are
available from the second author. The teacher
training sessions for this segment ranged from
33 to 53 min in length.
To ensure that teachers were implement-
ing the interventions with integrity, teachers’
percentage of commands given in alpha for-
mat, as well as the number of verbal praises,
were monitored during each observation. Ef-
fective commands were monitored on an on-
going basis throughout the study to ensure that
teachers were delivering a level of 80% alpha
commands, as measured by alpha commands/
total commands and that the level of alpha
commands did not spontaneously increase
without training. Once the teachers began the
effective commands phase, the teacher was
coached and given feedback regarding his or
her performance on the previous observation
prior to the next observation period. This
was accomplished by providing the teacher
with a written report on the percentage of
effective commands provided during the
previous observation and specific ineffec-
tive commands that needed to be changed.
During the praise condition, this feedback
also included the number of praises provided.
This written feedback was provided during the
transition between activities prior to the start
of the observation period. This step was
thought to be important because structured fol-
low-up meetings have been found to improve
208
Effective Commands
treatment fidelity and effectiveness (Noell,
Duhon, Gatti, & Connell, 2002).
Written feedback and verbal coaching
were given until a level of 80% alpha com-
mands on 2 consecutive observation days was
reached. The teachers received written feed-
back on each observation day throughout the
rest of the intervention phases. If the teacher’s
level of alpha commands fell below 80% on 2
consecutive observation days at any time fol-
lowing the onset of the effective commands
phase, the teacher was again provided written
and verbal feedback regarding his or her per-
formance until a level of 80% alpha commands
on 2 consecutive observation days was again
reached. Because teachers typically do not
exhibit a high rate of verbal praise (Atwater &
Morris, 19SS), it was not predicted that teach-
ers would begin adding verbal praise before
that phase was taught. However, teacher pro-
vision of verbal praise was monitored through-
out the study to assure that this assumption
remained accurate.
Design
A multiple baseline-across participants
design was utilized. Baseline information on
the compliance behavior of all 3 students was
obtained. Once the baseline data were stable,
intervention was implemented with the first
student while the baseline phase continued for
the remaining students. Stability was defined
as a minimum of 3 consecutive data points that
fell within 15% of the mean and did not show
a systematic trend (Kazdin, 1982). Experimen-
tal control is demonstrated when student com-
pliance shows a positive increase following
introduction of the intervention for that par-
ticular student while the other participants’ data
remain steady until the intervention is intro-
duced respectively for each of the students.
Phase 1 of the study was Baseline. Phase
2 of the study was Effective Command train-
ing alone. A level of 80% alpha commands, as
measured by the percentage of alpha com-
mands out of total commands during the in-
structional period observed, was needed for this
phase to be considered implemented with in-
tegrity. Phase 3 of the study was the Effective
Commands with verbal praise component.
Phase 3 was to be implemented only if 90%
total compliance was not achieved in Phase 2
(Effective Command training alone). Because
this level of total compliance was not achieved
for any of the 3 students, Phase 3 (Effective
Commands with verbal praise) was imple-
mented for all students. Phase 3 was contin-
ued until the compliance data stabilized or until
the end of the school year, as the study was
conducted in the late spring.
Interobserver Agreement
Interobserver agreement was coded for
20% of the classroom observations, which re-
sulted in 12 sessions being randomly selected
for each participant. Coding was initially com-
pleted by the first author. An independent coder
assisted with interobserver agreement. For the
compliance code, a Kappa statistic was calcu-
lated to provide an estimate of agreement be-
tween observers corrected for chance (Cohen,
1960). These data yielded an average Kappa
of .87 (range = .82 to .94). Agreement was rated
on the identification of alpha and beta com-
mands and compliance to commands. A Kappa
statistic was also calculated for the agreement
on the academic behaviors of task management
and academic responding as well as compet-
ing behaviors. Interrater agreement for the aca-
demic behaviors and competing behaviors
yielded an average Kappa of .90 (range = .87
to .95).
Results
Teacher Behavior
The percentage of effective teacher com-
mands and rate of teacher praise for each ob-
servation period are presented in Figures 1 and
2, respectively. Across all 3 teachers, the per-
cent of effective commands provided (relative
to total number of commands provided) prior
to training was low (see also Table 1). All 3
teachers increased use of effective commands
when the Effective Commands phase was in-
troduced.
Cory’s teacher was showing an increas-
ing trend in effective commands towards the
end of baseline. She demonstrated an immedi-
ate increase in level of effective commands
209
School Psychology Review, 2005, Volume 34, No. 2
Table 1
Means and Ranges of Percentage of Effective Commands for Each Phase
Baseline Effective Commands Effective Commands
Plus Praise
Cory’s
Teacher
Nate’s
Teacher
Andy’s
Teacher
Total
Total
Total
Mean Range
46.25 22-63
64.00 47-84
51.80 35-7 1
Mean Range Mean Range
83 67-93 78.78 70-83
81.38 67-9 1 84.75 66-93
86.5 83-93 92.5 77-100
once this phase was initiated and increased sta- strated less than 80% effective commands trig-
bility in her use of effective commands across gering a coaching session.
observations. This was largely maintained into Nate and Andy’s teachers had fairly
the effective commands plus praise condition stable data with no specific trends in baseline.
although a small decrease in level of effective They both evidenced an immediate increase
commands was observed and there were two in level when the effective commands condi-
consecutive occasions on which she demon- tion was introduced and data continued to re-
100 -
90 -
80 -
70 -
60 -
50 -
40 -
30 -
20 -
10 r
Baseline
OL
Effective Commands Effective Commands with Praise
ICI0
90
:;: ;
40 -
30 -
20
10 1
OJ
Cory’s Teacher
100
90
80
o:,,,., ,,,,,,,,,
v-
Nate’s Teacher
Andy’s Teacher
1 2 3 4 5 6 7 8 9 10 II I2 13 I4 I5 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
School Days
Figure 1, Percentage of Effective Commands Provided by Teacher.
210
Effective Commands
Table 2
Mean Frequency, Rate Per Minute, and Range of Total
Commands Given for Each Phase
Baseline Effective Effective Commands
Commands Plus Praise
Freq Rate (Range) Freq Rate (Range) Freq Rate (Range)
-_-~
Cory ‘s
Teacher 24 .84 (.64-1.04) 32 1.12 (.82-1.50) 30 .95 (.42-1.26)
Nate’s
Teacher 22 .70 (.44-1.03) 21 .71 (.55-1.00) 26 .76 (.52-.97)
Andy’s
Teacher 15 .29 (.17-0.44) 13 .33 (.18-0.43) 14 .38 (.29-.50)
main stable and above 80%. Some initial de-
crease in effective commands was noted when
the effective commands and praise condition
was initiated, but this did not persist and coach-
ing was not triggered by more than 1 day of
less than 80% provision of effective com-
mands.
The number of commands given per ob-
servation was also collected to monitor any
systematic changes in the rate of commands
given that may have affected participants’ com-
pliance and other behavior. The frequency and
rate of commands given per observation are
reported in Table 2 across each student/teacher
and phase. Cory’s teacher did evidence an in-
creased rate of commands during the effective
commands phase. Her rate of commands de-
creased some during the effective commands
plus praise condition, but was still higher than
baseline.
The frequency and rate of verbal praise
teachers provided prior to training were low
(see Figure 2 and Table 3). However, Cory’s
teacher did demonstrate an increased rate of
praise during the effective commands condi-
tion. For all three teachers, there was an im-
mediate increase in the rate of praise follow-
ing verbal praise training (Figure 2). Increases
in praise were maintained throughout this
phase for Cory and Nate’s teachers. Andy’s
teacher showed a downward trend in praise
after an initial increase.
Student Compliance
Percent of compliance for each obser-
vation across conditions for each of the par-
ticipants is presented in Figure 3. Each of the
students demonstrated a stable or downward
trend in compliance during baseline condi-
tions. Cory and Nate appeared to evidence
an immediate increase in level when the ef-
fective commands condition was introduced.
The data remain stable throughout the phase.
Andy did not evidence an immediate increase
in level, although the overall level of data in the
effective commands condition was higher. He
evidenced more variability in his compliance
relative to baseline. This variability in data con-
tinued into the praise condition with a slight in-
crease in level of data, but no discernable
trend. Cory and Nate also demonstrated
slight increases in compliance during the
effective commands plus praise conditions
with data stability comparable to previous
conditions and no discernable trend in the data.
Mean compliance percentages and stan-
dard deviations for each condition and effect
sizes for the baseline to effective commands
conditions, effective commands to praise con-
ditions, and baseline to praise conditions are
presented in Table 4. Effect sizes were com-
puted based on a method described by Busk
and Serlin (1992; Approach One). Effect sizes
indicate the degree to which a treatment effect
approaches or exceeds one standard deviation
211
Baseline
0.8 -
0.7 -
0.6 -
0.5 -
0.4 -
0.3 -
0.2 -
0.1 -
O- q
PI
rcc 0.8 r
0
0.7 -
d)
$ij 0.6 -
p! 0.5 -
0.4 -
0.3 -
0.2 -
Effective Commands Effective Commands plus Praise
Cory’s Teacher
0.8 -
0.7 -
0.6 -
0.5 -
0.4 -
0.3 -
0.2 -
0.1 -
Nate’s Teacher
Andy’s Teacher
0
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 31
School Days
Figure 2. Rate of Praise Provided by Teacher.
Effective Commands
Table 3
Mean Frequency, Rate Per Minute, and Range of Verbal Praise
Given for Each Phase
Baseline Effective Effective Commands
Commands Plus Praise
-
Freq Rate (Range) Freq Rate (Range) Freq Rate (Range)
Gory’s
Teacher 1 .03 (.OO-.09) 4.38 .15 (.05-.33) 13.56 .39 (.24-.61)
Nate’s
Teacher 1 .32 (.OO-JO) SO .17 (.OO-.03) 6.50 .19 (.ll-.31)
Andy’s
Teacher .40 .Ol (.OO-.02) .25 .o 1 (.OO-.03) 3.75 .lO (.04-.13)
of the mean of the baseline data. An effect size
of 1 .OO or greater indicates that treatment data
are one or more standard deviations above the
baseline mean. The effect sizes presented in
Table 4 would be considered to indicate mod-
erate to good treatment effects.
Academic and Competing Behaviors
Academic-related behaviors and com-
peting behaviors are presented in Figure 4. The
academic behaviors and competing behaviors
for all 3 participants were relatively stable dur-
ing baseline observations. An increase in level
was observed for all 3 participants with intro-
duction of the effective commands condition.
Neither downward nor upward trends were
evidenced for any of the participants in the ef-
fective commands condition. Small increases
in level of academic behaviors and decreases
in level of competing behaviors were observed
with the introduction of the effective com-
mands plus praise condition.
Means, standard deviations, and effect
sizes are presented in Table 4. The effect
sizes computed are indicative of good treat-
ment effects for academic behaviors and
moderate to good treatment effects for the
competing behaviors.
Discussion
Results of this study indicate that im-
provement in rates of compliance and academic
behaviors were observed when teachers in-
creased their use of effective commands. Fur-
ther improvements in compliance and aca-
demic behaviors were noted when teachers
increased their use of praise in conjunction with
using more effective commands. These im-
provements in compliance and academic be-
haviors are supportive of the hypothesis that
increasing effective teacher commands would
result in increased rates of student compliance
and academic behaviors. The study demon-
strated that teacher training in effective com-
mands did improve teachers’ use of effective
commands. Likewise, training in use of con-
tingent verbal praise was effective in improv-
ing teacher use of verbal praise.
Positive changes in students’ compliance
occurred with the introduction of teacher use
of effective commands. Consistent with the lit-
erature in parent training demonstrating that
parents’ use of effective commands improves
children’s compliance, teachers’ use of effec-
tive commands may also improve children’s
compliance in the classroom. In addition, stu-
dents were more likely to engage in academic-
related behaviors and less likely to engage in
competing disruptive behaviors. Students who
are more likely to do what the teacher tells them
to do and what the teacher expects them to do
(i.e., classroom rules, routines) may be more
likely to engage in academic-related behaviors
(e.g., read a book, get out needed materials,
213
School Psychology Review, 2005, Volume 34, No. 2
00000000000
OOlCO~\OIA~M~”
4 00000000000
OOl00bD~TtMC\1”
FH
-
214
Effective Commands
M
&“I”“@‘1
00000000000
OC3OO~USv)TJ-~N-
F-4
00000000000
OO\OOtbV>v,T3-Cr,N-
F-4
215
School Psychology Review, 2005, Volume 34, No. 2
Table 4
Means, Standard Deviations, and Effect Sizes Across Phases for Compliance,
Academic Behaviors, and Competing Behaviors
I3 aseline Effective Effective Commands
Commands Plus Praise
Cory
Compliance
Academic
Competing
Nate
Compliance
Academic
Competing
Andy
Compliance
Academic
Competing
Mean StDev Mean StDev ES Mean StDev ES2 ES”
49.50 10.38 56.5 6.72 .67 66.44 8.53 1.63 1.48
59.25 3.77 68.5 7.69 2.45 74.89 7.27 4.15 .83
40.50 3.87 31.28 7.93 2.36 25.22 7.28 3.95 .76
56.63 7.71 71.00 7.86 1.86 84.00 3.16 3.55 1.65
71.25 5.31 76.44 6.21 .98 84.75 4.50 2.54 1.34
28.5 5.32 24.69 7.29 .72 15.25 4.50 2.49 1.29
44.50 5.07 61.25 16.23 3.30 67.50 16.68 4.54 .39
59.2 6.98 74.88 14.12 2.25 83.50 11.12 3.48 .61
40.80 6.30 25.13 14.27 2.49 16.50 11.12 3.86 .60
ES1 = Effect Size between Baseline and Effective Command phase; ES2 = Effect Size between Baseline and Effective
Command Plus Praise phase; ES3 = Effect Size between Effective Command and Effective Command Plus Praise phase.
complete a worksheet). There is a degree of
overlap in the response classes of compliance
and academic behaviors such that complying
with teacher instructions regarding classroom
activities is also an academic behavior and
actively engaging in academic behaviors is a
form of compliance.
All 3 teachers indicated that they were
comfortable with the intervention of giving
effective commands and 2 of the 3 teachers
indicated that they felt comfortable with in-
creasing their use of verbal praise in conjunc-
tion with effective commands. One teacher
who did not feel comfortable with increasing
praise indicated that students at the fourth grade
level should not need frequent praise and only
implemented the intervention as a requirement
of the study. All 3 teachers agreed that they
perceived that the interventions had benefited
the target students as well as the other students
in the classroom. One parent asked if the in-
tervention could be continued the following
year due to the positive changes she had ob-
served in her son in the home setting since the
interventions were implemented. The
videographer who assisted in videotaping only
and was not trained in the area of education
consistently reported that things appeared to
be improving in Cory’s classroom and the
teacher appeared more in control of the class-
room. Although these reports are anecdotal, all
indicate observable improvements for students
and are consistent with observed changes in
behavior that occurred as a function of inter-
vention.
For teachers who do not naturally use
effective commands in their classroom, this
intervention represents a change in behavior.
216
The teacher will need to change the form of
the commands s/he provides and/or change the
ratio of alpha commands relative to beta com-
mands s/he provides. Although seemingly a
fairly simple strategy, direct training, monitor-
ing, and frequent feedback during the consul-
tation process may be necessary to improve
treatment integrity. Frequent feedback was pro-
vided during this study to assist with maintain-
ing higher ratios of alpha commands.
For school psychologists consulting with
teachers regarding students who are
noncompliant, it is important to consider how
the teacher provides commands. Consulting
with teachers about how to provide effective
commands may improve students’ rate of com-
pliance. Consultation may require direct in-
struction to include didactic teaching, model-
ing, practice, and feedback with the teacher.
To be most effective, however, it is important
to remember that increasing rates of reinforce-
ment for compliance has a long history of dem-
onstrated effectiveness in the classroom and
verbal praise has been found to be effective as
a reinforcer for many students (Madsen,
Becker, & Thomas,l968). Follow-up by the
consultant through frequent classroom obser-
vation and teacher feedback may be necessary
to initially establish and maintain treatment
integrity.
Limitations of the current study are the
small sample size, which may limit generaliz-
ing results to other children in terms of devel-
opmental functioning and to other teachers and
classrooms. In addition, although peer com-
parisons were utilized early in the participant
selection process, peer comparison data were
not collected during or after the study. Peer
comparison data may assist in providing in-
formation regarding whether changes in
teacher behavior affected other students as well
as the target students. These data may also as-
sist in assessing the degree to which the target
student behaves like a more typical peer.
Data on instructional activities were not
collected. Instructional activities may have
changed across sessions and phases and con-
tributed to some of the variability seen in the
data. It may be that a target student or students
are more likely to comply during some instruc-
tional activities (e.g., group activity vs. inde-
pendent seatwork) or comply more frequently
during preferred activities (reading vs.
worksheet). Given that changes in student com-
pliance and academic behaviors occurred only
when changes in teacher behavior were intro-
duced, it is assumed that any variability in in-
structional activities does not affect the over-
all interpretation of results of this study. How-
ever, future research will need to control or
more systematically monitor instructional ac-
tivities over time as a possible influence on
student compliance and academic behavior.
Future research may also evaluate the
effects on student behavior of various compo-
nents of effective commands alone and/or in
combination. Additional studies might evalu-
ate the components of effective commands
with close proximity without contingent rein-
forcement or a combination of effective com-
mands with close proximity and eye contact
without contingent reinforcement. There are
many possible combinations and discovering
which components or combinations of com-
ponents result in the greatest improvements in
compliance and academic behaviors will pro-
vide information on how to best train teachers
to manage their classrooms.
In summary, teachers can be taught to
provide more effective commands in their
classrooms. The use of effective commands
alone or in combination with verbal praise
for appropriate classroom behavior could be
recommended to teachers as an intervention
to improve student compliance or as part of
a multicomponent package to improve stu-
dent compliance. In addition, positive ef-
fects may be observed on other academic
related behaviors.
References
Anderson, L. W. (1984). Instruction and time-on-task: A
review. In L. W. Anderson (Ed.), fime andschool learn-
ing (pp. 143-163). New York: St. Martin’s Press.
Atwater, J. B., & Morris, E. K. (1988). Teachers’ instruc-
tions and children’s compliance in preschool class-
rooms: A descriptive analysis. Journal of Applied Be-
haviorAnalysis, 21, 157-167.
Borg, W. R. (1980). Time and school learning. In C.
Derham & A. Lieberman (Eds.), Time to learn (pp.27-
72). Washington, DC: National Institute of Education.
217
School Psychology Review, 2005, Volume 34, No. 2
Brophy, J. E. (1983). Classroom organization and man-
agement. The Elementary School Journal, 83, 265
285.
Busk, P. L., & Serlin, R. C. (1992). Meta-analysis for
single-case research. In T. R. Kratochwill & J. R. Levin
(Eds.), Single-case research design and analysis: New
directions for psychology and education (pp. 187-2 12).
Hillsdale, NJ: Lawrence Erlbaum Associates.
Cohen, J. (1960). A coefficient of agreement for nominal
scales. Educational and Psychological Measurement,
20, 37-46.
DeMartini-Scully, D., Bray, M. A., & Kehle, T. J. (2000).
A packaged intervention to reduce disruptive behav-
ior in general education students. Psychology in the
Schools, 37, 149-156.
Doyle, W. (1985). Recent research on classroom manage-
ment: Implications for teacher preparation. Journal of
Teacher Education, 36, 3 l-35.
Fisher, C. W., & Berliner, D. C. (Eds.). (1985). Perspec-
tives on instructional time. New York: Longman.
Ford A. D. (1998). An evaluation of the contributions of
effective instruction delivery to a compliance training
package for children. Unpublished doctoral disserta-
tion, University of Southern Mississippi, Hattiesburg.
Forehand, R. L. ( 1977). Child noncompliance to parent
commands: Behavior analysis and treatment. In M.
Hersen, R. M. Eisler, & P. M. Miller (Eds.), Progress
in behavior modification (vol. 5, pp. 11 I- 147). New
York: Academic Press.
Forehand, R. L., & McMahon, R. J. (198 1). Helping the
noncompliant child. New York: Guilford Press.
Forehand, R., & Scarboro, M. E. (1975). An analysis of
children’s oppositional behavior. Journal ofAbnormal
Child Psychology, 3, 27-3 1.
Gettinger, M. (1986). Issues and trends in academic en-
gaged time of students. Special Services in the Schools,
2, 1-17.
Good, T. L. (1979). Teacher effectiveness in the elemen-
tary school. Journal of Teacher Education, 30, 52-64.
Good, T. L. (1983). Classroom research: A decade of
progress. Educational Psychologist, l&3), 127- 144.
Greenwood, C. R. (1991). Longitudinal analysis of time,
engagement, and achievement in at-risk versus non-
risk students. Exceptional Children, 57, 521-535.
Greenwood, C. R., Carta, J. J., Kamps, D., Delquadri, J.,
Terry, B., Finney, R., Hou, S., Lowe, A., Montagna,
D., Peterson, P., Dawson, H., & Schick, C. (1995).
Ecobehavioral assessment systems software: (EBASS).
Lawrence, KS: Juniper Gardens Children’s Project,
University of Kansas.
Gresham, F. M. (1989). Assessment of treatment integrity
in school consultation and prereferral intervention.
School Psychology Review, 18, 37-50.
Hamlet, C. C., Axelrod, S., & Kuerscher, S. (1984). Eye
contact as an antecedent to compliant behavior. Jour-
nal of Applied Behavior Analysis, 17, 553-557.
Howell, K. W., & Nolet, V. (2000). Curriculum-based
evaluation: Teaching and decision making (3”’ ed.).
Belmont, CA: Wadsworth/Thomson Learning.
Karweit, N. L. (1983). Time on task: A research review
(Report No. 332). Baltimore, MD: Center for Social
Organization of Schools, Johns Hopkins University.
Karweit, N. L. (1985). Should we lengthen the school
term? Educational Researcher; 14(6), 9-15.
Kazdin, A. E. ( 1982). Single-case research designs: Meth-
ods for clinical and applied settings, New York: Ox-
ford.
Madsen, C. H., Becker, W. C., & Thomas, D. R. (1968).
Rules, praise, and ignoring: Elements of elementary
classroom control. Journal ofApplied BehaviorAnaly-
sis, I, 139-150.
McMahon, R. J., & Forehand, R. L. (2003). Helping the
noncompliant child: Family-based treatment for op-
positional behavior (2nd ed.). New York: Guilford Press.
Musser, E. H., Bray, M, A., Kehle, T. J., & Jenson, W.
(2001). Reducing disruptive behaviors in students with
serious emotional disturbance. School Psychology
Review, 30, 294-304.
Neef, N. A., Sharfer, M. S, Egel, A. L., Cataldo, M. F., &
Parrish, J. M. (1983). The class specific effects of com-
pliance training with “do” and “don’t” requests. Ana-
logue analysis and classroom application. Journal of
Applied Behavior Analysis, 16, 8 l-89.
Noell, H. G., Duhon, G. J., Gatti, S. L., & Connell, J. E.
(2002). Consultation, follow-up, and implementation
of behavior management interventions in general edu-
cation. School Psychology Review, 31,2 17-234.
Peed, S., Roberts, M., & Forehand, R. (1977). Evaluation
of the effectiveness of a standardized parent training
program in altering the interaction of mothers and
noncompliant children. Behavior Modification, 1, 323-
350.
Rhode, G., Jenson, W. R., & Reavis, H. K. (1993). The
tough kid book. Longmont, CO: Sopris West, Inc.
Roberts, M. W., McMahon, R. J., Forehand, R., &
Humphreys, L. (1978). The effect of parental instruc-
tion-giving on child compliance. Behavior Therapy,
21, 257-27 1.
Roberts, M. W., & Powers, S. W. (1988). The compliance
test. Behavioral Assessment, IO, 375-398.
Schoen, S. F. (1986). Decreasing noncompliance in a
severly multihandicapped child. Psychology in the
Schools, 23, 88-94.
Shapiro, E. S. (1996). Academic skills problems: Direct
assessment and intervention (2nd ed.). New York:
Guilford Press.
Shriver, M. D., &Allen, K. D. (1997). Defining child non-
compliance: An examination of temporal parameters.
Journal of Applied Behavior Analysis, 30, 173- 176.
Watson, T. S., & Robinson, S. L. (1996). Direct behav-
ioral consultation: An alternative to traditional behav-
ioral consultation. School Psychology Quarterly, II,
267-278.
Williams, C. A., & Forehand, R. L. (1984). An examina-
tion of predictor variables for child compliance and
noncompliance. Journal of Abnormal Child Psychol-
ogy, 12,491-504.
218
Effective Commands
Andrea Starkweather Matheson received her PhD from Iowa State University in School
Psychology in 2001. She presently serves as a school psychologist and assessment con-
sultant for Heartland Area Education Agency 11 in Johnston, Iowa. Her professional in-
terests focus on educational reform and intervention studies.
Mark D. Shriver received his PhD in School Psychology from the University of Nebraska-
Lincoln in 1994. Currently he is an Associate Professor in Pediatrics at the University of
Nebraska Medical Center and provides clinical pediatric psychology services at the Munroe-
Meyer Institute for Genetics and Rehabilitation. His clinical and research interests are in
the areas of parent training, early childhood education, assessment, and intervention to
improve academic and behavioral functioning of children with disabilities in the home,
school, and community. He is co-editor of the Journal of Evidence Based Practices for
Sch002s.
