+1443 776-2705 panelessays@gmail.com
  

1 pg summry on article attached below

Evaluation of a telehealth parent training program in teaching
self-care skills to children with autism

Ariana R. Boutain, Jan B. Sheldon and James A. Sherman
Department of Applied Behavioral Science, University of Kansas

The present study used synchronous video conferencing to remotely deliver a behavioral skills
training-based (BST) parent training program to 3 parents of children with autism in the family
home. Parents were taught to implement graduated guidance to teach their children several
important self-care skills. Parents did not correctly implement graduated guidance after receiving
detailed written instructions only. After parents received the BST parent training package, how-
ever, all parents implemented graduated guidance with near-perfect levels of fidelity, and all chil-
dren completed the targeted self-care skills with substantially higher levels of accuracy and
independence. Furthermore, parents reported high levels of satisfaction with graduated guidance,
the telehealth BST training package, and their children’s ability to complete self-care skills.
Key words: telehealth, parent training, behavioral skills training, self-care skills, autism

Research has demonstrated that early and
intensive interventions that use the principles of
applied behavior analysis (ABA) can be effective
in teaching skills to children with autism spec-
trum disorder (ASD) (Matson & Smith, 2008).
Unfortunately, parents of children with ASD
often experience difficulty obtaining ABA services
for their children (Kogan et al., 2008) due to
barriers such as cost, lengthy waitlists, or geo-
graphic isolation (Koegel et al., 2002; Stahmer &
Gist, 2001). One way to address these issues is
to train parents of children with ASD to be effec-
tive behavior-analytic teachers for their children.
A number of authors have discussed the benefits

of training parents to implement ABA programs
with their children. For example, several authors
have indicated that parent involvement in ASD
interventions may help promote generalization and

maintenance of child skills and is often more cost-
effective than a purely therapist-based treatment
intervention (e.g., Ingersoll & Gergans, 2007;
Lerman et al., 2020). Parent training and involve-
ment has also been shown to reduce parental stress
related to caring for a child with ASD (Koegel
et al., 1996), increase parent optimism and reported
leisure time (Koegel et al., 1982), and increase par-
ents’ reported sense of competence related to their
parental abilities (Connell et al., 1997). Further-
more, the National Research Council (NRC)
suggested that a fundamental component of effec-
tive ASD intervention programs is parent involve-
ment (NRC, 2001).
One method that has been shown to be effective

in teaching new skills to parents is behavioral skills
training (BST). BST is a teaching procedure that
involves the use of instructions, modeling, rehearsal
(or role-play), and feedback (Miltenberger, 2004).
BST has been used to teach parents of children with
ASD to implement a variety of interventions
including shaping and prompting procedures
(e.g., Koegel et al., 1978), functional analyses
(e.g., Stokes & Luiselli, 2008), function-based treat-
ments to address problem behavior (e.g., Robertson
et al., 2013; Tarbox et al., 2002; Vollmer
et al., 1994), the Early Start Denver Model
(ESDM) (e.g., Vismara et al., 2009), toilet training

This study is based on a dissertation submitted by the
first author, under the supervision of the second and third
author, in the Department of Applied Behavioral Science
at the University of Kansas. We would like to thank
Jessica Winne, Britney Latham, and Florence DiGennaro
Reed for their assistance with various aspects of this
project.
Address correspondence to: Ariana Boutain, who is

now with KGH Autism Services, 1161 Lake Cook
Road, Deerfield, IL 60015 (email: [email protected]
services.com)
doi: 10.1002/jaba.743

Journal of Applied Behavior Analysis 2020, 53, 1259–1275 NUMBER 3 (SUMMER)

© 2020 Society for the Experimental Analysis of Behavior

1259

protocols (e.g., Kroeger & Sorensen, 2010), joint
attention programs (e.g., Rocha et al., 2007),
discrete trial instruction (e.g., Lafasakis &
Sturmey, 2007), multiple phases of the picture
exchange communication system (PECS) (e.g.,
Ben-Chaabane et al., 2009), and a variety of natu-
ralistic behavioral interventions such as natural
language paradigm, milieu teaching, pivotal
response teaching, reciprocal imitation training,
embedded teaching, and incidental teaching
(e.g., Charlop-Christy & Carpenter, 2000; Inger-
soll & Gergans, 2007; Koegel et al., 2002). It is
often the case, however, that parents are unable to
receive adequate training due to obstacles such as
geographic location, cost, and a shortage of quali-
fied professionals (e.g., Board Certified Behavior
Analysts®, BCBAs). One way to address these
issues is to use telehealth to remotely train parents.
Telehealth (also called ‘telemedicine’ or

‘telepractice’) is the use of communication
technology to provide critical health services
remotely to people who may not have direct
access to the professionals who can provide
these services. Telehealth can involve the use of
telephones, email, asynchronous videos or pic-
tures, or synchronous video conferencing
(e.g., Doxy®, Zoom®, FaceTime®). A common
telehealth model consists of a practitioner using
synchronous video conferencing to deliver live,
real-time services to a client in their home.
Research has shown that therapists can effec-
tively teach parents via telehealth to implement
a variety of behavior-analytic programs with
their children (e.g., for reviews, see Ferguson
et al., 2019; Tomlinson et al., 2018). For
example, clinicians in numerous studies have
used BST via synchronous video conferencing
to teach parents to implement functional assess-
ment and treatment procedures to address
problem behavior (e.g., Lindgren et al., 2016;
Machalicek et al., 2016; Monlux et al., 2019;
Suess et al., 2014; Suess et al., 2016; Tsami
et al., 2019; Wacker et al., 2013a; 2013b) and
programs aimed at increasing social and
communication behaviors (e.g., Baharav &

Reiser, 2010; McDuffie et al., 2013; Vismara
et al., 2012; 2013; Wainer & Ingersoll, 2015).
Although research regarding the necessary and

sufficient components of BST is mixed, a number
of studies have found that written instructions alone
are not effective in teaching individuals to correctly
and consistently implement new skills (e.g.,
Ducharme & Feldman, 1992; Gardner, 1972;
Ward-Horner & Sturmey, 2012). Interestingly,
however, recent studies examining the effectiveness
of self-instruction packages in isolation have rev-
ealed some promising results. For example, a study
by Graff and Karsten (2012) found that “enhanced”
written instructions (EWI) effectively trained
teachers to implement stimulus preference assess-
ments with children with ASD. The EWI consisted
of jargon-free, step-by-step instructions, a detailed
data sheet, and diagrams that detailed how to con-
duct the preference assessments. A follow-up study
by Shapiro et al. (2016) similarly found that EWI
were sufficient to teach five of seven undergraduates
with no prior coursework in ABA and four of five
in-home behavior technicians to conduct stimulus
preference assessments. Nonetheless, more research
is needed to further investigate the range of skills
that can be taught using a similar type of detailed
written instructions. If effective, detailed written
instructions could be used within telehealth parent
training programs alone or in concert with other
BST components (e.g., model, role-play, feedback)
to teach parents more effectively and efficiently.
Currently, no studies have examined the effective-
ness of detailed written instructions in teaching
parents.
Furthermore, despite the growing number of

studies demonstrating the effectiveness of
telehealth parent training models, no studies
have targeted the skills needed to teach children
multistep, self-care skills. This is an important
area because many children with ASD exhibit
pronounced deficits in self-care skills (Flynn &
Healy, 2012). Deficits in these skills can hinder
a child’s integration into daycare or school
settings and result in a greater reliance on
parents to complete these necessary skills

Ariana R. Boutain et al.1260

(Jasmin et al. 2009). Additionally, no telehealth
parent training studies have targeted graduated
guidance, which is commonly used to teach
children with ASD to complete self-care skills
(Demchak, 1989; 1990).
The purpose of the current study was to

evaluate a telehealth program to teach parents
of children with ASD to implement graduated
guidance to teach their children to indepen-
dently complete three important self-care skills
(i.e., washing hands, washing face, and apply-
ing lotion). One goal was to determine if
detailed written instructions would be sufficient
to teach parents to implement the graduated
guidance procedure. If detailed written instruc-
tions were not sufficient, the second goal was
to determine if a parent training package deliv-
ered via telehealth would be effective for teach-
ing parents to implement graduated guidance
with acceptable levels of fidelity and concur-
rently produce positive changes in child
behavior.

Method

Participants, Setting, and Materials
Three children with ASD, two boys and one

girl, and their parents participated. Children
ranged from 4 to 5 years of age and were par-
ticipating in a center-based ABA program that
serves children with ASD located at a Midwest-
ern university. To participate in this study,
families had to meet the following criteria:
(1) children had an ASD diagnosis from an
independent agency; (2) children were between
the ages of 3 and 5 years; (3) at least one parent
was available to participate in two to three
20-30 min sessions per week for the duration
of the study; (4) another adult (e.g., other par-
ent, older sibling, neighbor, babysitter) was
available to participate in several role-play ses-
sions throughout the study; and (5) the family
home was equipped with a wireless router and
a high-speed Internet connection (i.e., at least
1 gigabyte/s upload and download) for the

duration of the study. One parent from each
parent–child dyad (referred to as the teaching
parent) was selected to teach self-care skills; the
other parent is referred to as the nonteaching
parent.
Jesse was a 5-year-old boy diagnosed with

ASD who had received an average of 35 hr a
week of ABA therapy for 16 months prior to
participating. The Assessment of Basic Learning
and Language Skills (ABLLS; Partington, 2006)
indicated that Jesse was demonstrating 58% of
all assessed skills. More specifically, Jesse was
demonstrating 71% of grooming skills, 90% of
gross motor skills, and 89% of fine motor skills.
Jesse’s mother was the teaching parent in this
study. At the beginning of the study, Jesse’s
mother was 36 years old, had completed some
college coursework, and did not have any prior
experience implementing discrete-trial instruction
or graduated guidance procedures. Jesse’s mother
was married and had a full-time job outside of
the family home.
Bobby was a 4-year-old boy diagnosed with

ASD who had received an average of 35 hr a
week of ABA therapy for 13 months prior to
participating in the current study. An ABLLS
assessment indicated that Bobby was demon-
strating 35% of all assessed skills. More specifi-
cally Bobby was demonstrating 86% of
grooming skills, 83% of gross motor skills, and
75% of fine motor skills. Bobby’s mother was
the teaching parent. At the beginning of the
study, Bobby’s mother was 34 years old, had
completed her bachelor’s degree, and did not
have any prior experience implementing dis-
crete trial instruction or graduated guidance
procedures. Bobby’s mother was married and
had a full-time job outside of the family home.
Laura was a 5-year-old girl diagnosed with

ASD who had received an average of 35 hours
a week of ABA therapy for the past 11 months.
An ABLLS assessment indicated that Laura was
demonstrating 4% of all assessed skills. More
specifically, Laura was demonstrating 0% of
grooming skills, 13% of gross motor skills, and

1261Telehealth Parent Training Program

21% of fine motor skills. Laura was the only
participant who engaged in problem behaviors.
Laura’s problem behaviors included aggression
and property destruction. Functional analyses
of Laura’s aggression and property destruction
were completed prior to the current study and
indicated that both were maintained by escape
from demands. Laura’s mother was the teach-
ing parent in this study. Laura’s mother was
30 years old, had completed her master’s
degree, and did not have any prior experience
implementing discrete trial instruction or grad-
uated guidance procedures. Laura’s mother was
also married and had a full-time job outside of
the family home.
The experimenter served as the parent

trainer for the duration of the study. She was a
master’s-level BCBA who had 8 years of experi-
ence working with children with autism and
their families and 4 years of experience con-
ducting parent training. She had not received
any formal training in telehealth or using
telehealth to deliver behavior analytic services.
All teaching sessions and observations took

place in the participants’ homes. The experi-
menter conducted all sessions using FaceTime®

video conferencing technology on password-
protected iPad® minis that allowed the
experimenter and parents to see, hear, and
communicate with each other in real time. Fac-
eTime® was selected as it was familiar, easily
accessible by the parent participants, and
authorized by the University of Kansas Institu-
tional Review Board. Because FaceTime® is
not HIPAA compliant, the risks associated with
using FaceTime® were discussed with the par-
ents who signed a consent form agreeing to its
use before starting in the study.
Each parent also received an iPad® mini

and Otterbox® case that functioned as a stand
for the iPad®. To ensure that only the teach-
ing parent (rather than both the teaching par-
ent and the child) heard the experimenter’s
instructions and feedback during sessions,
teaching parents were provided with Jabra®

wireless ear buds to wear during all telehealth
sessions.
To determine which self-care skills would be

taught to each parent–child dyad, an assess-
ment questionnaire was distributed to all parent
participants that asked them to individually rate
their satisfaction with their child’s ability to
demonstrate a number of skills and how impor-
tant they believed it was for their child to learn
each skill. All parents reported low satisfaction
with their children’s ability to independently
wash their hands, wash their face, and apply
lotion. Additionally, all parents reported high
importance with respect to their children inde-
pendently completing these self-care skills.
None of these self-care skills were, or had been,
directly taught as part of the children’s ABA
program.

Response Measurement
Parent Behavior
The primary dependent variable for all par-

ent participants was correct implementation of
the graduated guidance procedures. This vari-
able was scored using a parent behavior check-
list (see Table 1). For each graduated guidance
teaching session, the percentage of parent
behavior steps correctly completed by the par-
ent was calculated by dividing the total number
of parent behavior steps correctly completed by
the total number of applicable steps and multi-
plying that number by 100.
Laura’s mother was also taught to implement

constant prompt delay probe trials within the
graduated guidance teaching sessions to better
assess Laura’s progress in learning each self-care
skill. Thus, the correct implementation of the
constant prompt delay probe trial steps was
scored for Laura’s mother using a parent behav-
ior checklist that is available as supplementary
material on the publisher’s website. For each
constant prompt delay prompt probe trial, the
percentage of parent behavior steps correctly
completed by Laura’s mother was calculated by

Ariana R. Boutain et al.1262

dividing the total number of parent behavior
steps correctly completed by the total number
of applicable steps and multiplying that num-
ber by 100.
After each parent–child dyad had completed

training for the three self-care skills, all teaching
parents conducted posttraining probe trials to
assess their child’s performance of the self-care
skills when parents simply told their children
to complete each skill. Parent implementation
of postraining probe trials also was scored
using a parent behavior checklist (available as

supplementary material on the publisher’s
website). The percentage of parent behavior
steps correctly completed was calculated by
dividing the total number of parent behavior
steps correctly completed by the total number
of applicable steps and multiplying that num-
ber by 100. All teaching parents correctly com-
pleted 100% of posttraining probe steps after
receiving the oral instructions from the primary
experimenter.

Child Behavior
The dependent variables for all child partici-

pants were the independent completion of self-
care skill steps and the occurrence of problem
behavior. Independent skill completion was
defined as the child (a) correctly performing all
steps of the skill in the absence of prompts,
and (b) not engaging in any problem behaviors
during the skill. Of the three child participants,
only Laura displayed problem behavior during
sessions. The behaviors included aggression,
defined as anytime Laura’s open or closed
hands or head made contact with any part of
her parent’s body with enough force to produce
a sound or a mark, or anytime Laura’s open
mouth or teeth made contact with any part of
her parent’s body and left a visible mark; and
property destruction, defined as grabbing and
releasing an object so that it traveled more than
1 foot or made an audible sound when it made
contact with another object. For each graduated
guidance teaching session, constant prompt
delay probe trial (for Laura), and posttraining
probes, the percentage of self-care skill steps
independently completed by the child was cal-
culated by dividing the number of self-care skill
steps independently completed by the total
number of self-care skill steps and multiplying
by 100. Additionally, for each graduated guid-
ance teaching session (and constant prompt
delay trial for Laura), the percentage of self-care
skill steps that contained problem behavior was
calculated by dividing the number of self-care
skill steps during which the child displayed

Table 1

Parent Behavior Checklist for Graduated Guidance Teaching
Procedure

1. Did the teaching parent present at least two different pieces of
the child’s preferred edibles that the child chose during prior a
preference assessment?

2. Did the teaching parent ask the child to pick what edible he or
she wanted to work for?

3. Did the teaching parent put five pieces of the edible the child
chose into a small container?

4. Did the teaching parent bring the child into the bathroom?
5. Did the teaching parent close the bathroom door?
6. Did the teaching parent give the child the correct instruction
(e.g., “Please wash your hands”)?

7. Did the teaching parent start teaching by implementing the
appropriate level of prompt for all steps of the skill (e.g., last
level of prompt used in the previous teaching session or trial)?

8. Did the teaching parent provide vocal praise each time the
child correctly completed a skill step (i.e., step completed
without problem behavior, with or without help from parent)?

9. Did the teaching parent decrease the level of prompt one level
when the child correctly completed each step of the skill (i.e.,
each skill step completed without problem behavior, with or
without help from parent) three consecutive times?

10. Did the teaching parent provide vocal praise, physical
touches, and one serving of the child-selected edible each
time the child correctly completed (i.e., each step completed
without problem behavior, with or without help from parent)
all steps of a skill?

11. Did the teaching parent return to the previous prompt level
for the remaining skill steps if the child made an error (e.g.,
omitted a step, did not fully complete a step) or displayed
problem behavior?

12. Did the teaching parent remain calm and continue to prompt
the child through the skill if the child attempted to pull
away, protest, or resist the physical prompts?

13. Did the teaching parent ask the child to complete the skill
five times?

Note: Each item was scored as either “Y”-Yes, “N”-No, or
“NA”-Not Applicable

1263Telehealth Parent Training Program

problem behavior by the total number of self-
care skill steps and multiplying by 100.

Interobserver Agreement and Treatment
Integrity
A second observer independently scored data

on parent and child behavior for 30% of ses-
sions for each experimental condition for each
parent–child dyad. To assess interobserver
agreement (IOA) on the implementation of
teaching procedures for all three parents, a
point-by-point comparison was conducted for
each behavioral step on the graduated guidance
parent behavior checklist and the constant
prompt delay parent behavior checklist for
Laura’s mother. To evaluate IOA for child
independent completion of self-care skill steps
and the occurrence of problem behavior during
self-care skill steps (for Laura), a point-by-point
comparison was conducted for each self-care
skill step. IOA was calculated by dividing the
number of agreements by the number of agree-
ments plus disagreements and converting this
number into a percentage. Mean IOA for grad-
uated guidance and constant prompt delay par-
ent behavior across all parents was 98% (range,
86%-100%). Mean IOA for child independent
completion of skill steps across all children was
99% (range, 86%-100%). Mean IOA for child
problem behavior was 99% (range, 95%-
100%) for Laura.
To assess treatment integrity, a second

observer scored the experimenter’s implementa-
tion of the BST graduated guidance parent
training package and constant prompt delay
parent training package for 30% of all sessions
using experimenter checklists (available as sup-
plementary material on the publisher’s website).
For both procedures, treatment integrity was
calculated by dividing the number of steps
implemented correctly by the number of cor-
rect plus incorrect steps and multiplying by
100. Procedural integrity for both the gradu-
ated guidance parent training package and

constant prompt delay parent training package
was 100%.

Experimental Design and Procedures
A nonconcurrent multiple baseline design

across parent–child dyads and across self-care
skills within parent–child dyads was used to
evaluate the effectiveness of the telehealth par-
ent training program.

Graduated Guidance Teaching Procedure
The experimenter taught the teaching parent

to implement the 13 steps reflected in the
checklist in Table 1. The teaching parent tau-
ght her child one self-care skill at a time and
each graduated guidance teaching session con-
sisted of the teaching parent conducting five
graduated guidance teaching trials. Following
each teaching trial, the parent permitted the
child to consume the edible (if earned) and to
take a 2-min to 3-min break to allow the teach-
ing parent to reset the self-care trial and the
experimenter to provide feedback to the teach-
ing parent. The parent conducted teaching ses-
sions once per day, and each session lasted
20 min to 30 min.
The levels for gradually removing the physi-

cal prompts from the most controlling to the
least controlling were as follows: (1) initially,
the teaching parent used hand-over-hand, full
physical prompts to gently guide the child
through each step of the skill; (2) the teaching
parent used partial physical prompts by using
only her thumb and index finger to gently
guide the child through each step of the skill;
(3) the teaching parent used shadow prompts
by “shadowing” the child’s hands within
approximately one inch for each step of the
skill; finally, (4) the teaching parent presented
only the initial instruction to compete the skill.
Training for the skill ended when the teaching
parent completely removed her physical pro-
mpts and the child independently completed at
least 90% of the self-care skill steps for three

Ariana R. Boutain et al.1264

consecutive sessions. The graduated guidance
procedure was based on that described by
Demchak (1989).

Constant Prompt Delay Probe Trials
To better assess Laura’s progress in learning

self-care skills, Laura’s mother was taught to
conduct constant prompt delay probe trials.
During delayed feedback graduated guidance
teaching sessions for the first self-care skill,
washing face, Laura’s mother was implementing
graduated guidance with near 100% fidelity,
yet she was not able to successfully fade out her
physical prompts due to either the occurrence
of problem behavior or because Laura consis-
tently needed at least partial physical prompts
to complete certain skill steps (e.g., pumping
soap onto wash cloth, rinsing her face, turning
off the water). As a result, Laura did not have
the opportunity to independently complete the
skill. Thus, Laura’s mother was instructed to
conduct a constant prompt delay probe trial
prior to every third graduated guidance teach-
ing session to assess skill mastery. Laura’s
mother was instructed to use this additional
assessment procedure for the remaining two
self-care skills (i.e., applying lotion, wash
hands) during all teaching phases (e.g., detailed
written instructions, immediate feedback with
child, delayed feedback with child).
Laura’s mother was taught to implement

the following steps each time she conducted a
constant prompt delay probe trial: 1) give
Laura the correct instruction (e.g., “Please
wash your face”); 2) deliver a full physical
prompt to help Laura complete each skill step
that Laura did not complete within 5 s of the
initial instruction or within 5 s of completing
of the previous skill step; 3) deliver a full
physical prompt to help Laura complete each
skill step during which Laura displayed prob-
lem behavior (e.g., throwing wash cloth, hit-
ting parent, biting parent); 4) provide vocal
praise each time Laura correctly completed a
skill step without problem behavior, with or

without help from parent; 5) provide vocal
praise, physical touches, and one serving of
the child-selected edible if Laura correctly
completed all steps without problem behavior,
with or without prompts from the parent).

Parent Training Procedures
The parent training program consisted of a

baseline phase, a detailed written instructions
phase, and, if necessary, a BST parent training
package phase.
Baseline. The experimenter told the teach-

ing parent to use graduated guidance to teach
her child to complete a self-care skill and did
not answer any questions or provide feedback.
Detailed Written Instructions. The experi-

menter gave the teaching parent detailed writ-
ten instructions that explained how to
implement the graduated guidance teaching
procedure to teach a skill. The detailed written
instructions also included a task analysis detail-
ing the specific skill steps for the self-care skill
(e.g., get wash cloth, turn on water, get wash
cloth wet). The detailed written instructions
were the same for each self-care skill except for
the specific task analysis detailing the steps for
the self-care skill. The experimenter then told
the teaching parent to use graduated guidance
to teach her child to complete the self-care skill
and did not provide any feedback on her per-
formance. If the teaching parent attempted to
ask a question regarding the graduated guid-
ance teaching procedures before or during grad-
uated guidance teaching trials, the experimenter
stated to “Just do your best.” For each self-care
skill, if the teaching parent implemented gradu-
ated guidance with her child across five consec-
utive sessions with at least 90% fidelity after
receiving the written instructions and the child
independently demonstrated at least 90% of
the skill steps for three consecutive sessions,
training on the skill was considered complete
and the parent conducted no further training
on the skill. If not, the experimenter delivered
the BST parent training package.

1265Telehealth Parent Training Program

BST Parent Training Package. Training
consisted of a graduated guidance teaching
overview, modeling, role play, and immediate
and delayed feedback. First, the experimenter
orally described the graduated guidance teach-
ing procedure and the skill that the parent
would teach. The experimenter listed the skill
steps from the task analysis for the self-care skill
and provided rationales for why it is important
to teach the skill (e.g., “If we teach Laura to
wash her face by herself, she will be able to do
so when her face gets dirty and she will also be
able to maintain good hygiene”). Then, the
experimenter read aloud the steps of the gradu-
ated guidance procedure. Finally, the experi-
menter assessed the teaching parent’s
knowledge and understanding of the procedure
by giving the teaching parent an oral quiz that
involved answering aloud a series of questions.
The teaching parent had to answer all questions
correctly to move on to the modeling phase.
Next, the experimenter correctly modeled,

with a research assistant playing the child, the
parent behavior steps in using the graduated
guidance prompting procedure to teach the skill.
Before each teaching trial, the experimenter
instructed the research assistant on how to per-
form during the trial (e.g., complete all self-care
steps correctly, complete a step incorrectly, omit
a step, engage in problem behavior during skill).
First, the experimenter correctly modeled all
13 parent behavior steps. Next, the experimenter
correctly modeled nine parent behavior steps and
incorrectly modeled four parent behavior steps
with the research assistant still playing the role of
the child. A random number generator (http://
www.random.org) was used to determine which
four graduated guidance steps were incorrectly
modeled. Following each model, the experi-
menter asked the teaching parent to state the
steps that were correctly and incorrectly
implemented by the experimenter. The teaching
parent had to correctly identify all steps that were
correctly and incorrectly implemented by the
experimenter to move to role play.

During role play, the teaching parent’s
spouse (i.e., nonteaching parent) played the
role of the child, and the experimenter
“coached” the teaching parent through her
implementation of the entire graduated guid-
ance teaching procedure by providing immedi-
ate and on-going positive and corrective
feedback after each parent behavior step. The
feedback consisted of the experimenter praising
the parent for steps that were completed cor-
rectly (e.g., “Perfect hand-over-hand prompt-
ing!”) and giving corrective feedback for errors
(e.g., “Remember to deliver praise, physical
touches, and an edible if Jesse correctly com-
pletes all the steps”). Prior to each trial, the
experimenter instructed the nonteaching parent
on how to perform the self-care skill during the
trial and ensured that the teaching parent prac-
ticed given a variety of child responses. After
the teaching parent role-played the entire grad-
uated guidance teaching procedure and per-
formed five consecutive graduated guidance
teaching trials with no corrective feedback from
the experimenter, the teaching parent began to
implement the graduated guidance procedure
with her child.
During the immediate feedback phase, the

teaching parent conducted gra