An Investigation of General and Descriptive Praise in Teaching  
Language to Children with Autism 
 
by 
 
Amy Susanne Polick 
 
 
 
 
A dissertation submitted to the Graduate Faculty of  
Auburn University 
in partial fulfillment of the  
requirements for the Degree of 
Doctor of Philosophy 
 
Auburn, Alabama 
May 9, 2011 
 
 
 
 
Keywords: Autism, Early Intervention, Praise, Applied Behavior Analysis 
 
Copyright 2011 by Amy Susanne Polick 
 
Approved by 
 
James E. Carr, Chair, Associate Professor of Psychology 
M. Christopher Newland, Alumni Professor of Psychology 
Linda A. LeBlanc, Associate Professor of Psychology 
Jennifer Gillis, Assistant Professor of Psychology 
 
 
 
 
 
ii 
 
 
 
 
 
 
Abstract 
 
 
 Research has shown that traditional operant behavioral procedures have been effective in 
enhancing language performance (Lovaas, 1987). In addition, the use of descriptive praise has 
been widely recommended in the area of language intervention for individuals with autism 
spectrum disorders (ASDs) (e.g., Leaf & McEachin, 1999; Maurice, Luce, & Green, 1996).  
However, few published studies have directly evaluated the use of descriptive praise in teaching 
language to children with ASDs (i.e., Sellers & Higbee, 2009; Stevens, Sidener, Reeve, & 
Sidener, 2011). The purpose of the study was to expand upon the Sellers and Higbee and Stevens 
et al. evaluations to further investigate the effects of descriptive versus general praise on the 
acquisition of language skills in children diagnosed with ASDs.  
 Results of the study showed that there were minimal, if any, advantages of descriptive 
praise in acquisition efficiency. In 5 out of 9 comparisons (55%), participants reached mastery 
faster with descriptive praise; however, the differences were often minor. In 3 out of 9 
comparisons (33%), participants met mastery (or failure) in the exact same number of sessions. 
Only 1 of the 9 comparisons (11%) resulted in faster skill acquisition with general praise, and 
again, the difference was small. From this investigation it appears that recommendations for 
using descriptive praise in teaching language to children with ASDs exceed the evidence to 
support them. Thus, it cannot be confidently concluded that descriptive praise provides a greater 
benefit than general praise for skill acquisition in children with ASDs, but rather the 
iii 
 
effectiveness of descriptive praise (and praise, in general) as a teaching strategy seems to be 
idiosyncratic. 
   
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
iv 
 
 
 
 
 
 
Acknowledgements 
 
  
 The author would like to thank her parents Jane and Ray for their unconditional love and 
support, her brothers Matt and Tim for their encouragement over the years, her best friend Liz 
Micheli for her continued friendship, and Dr. James Carr for his never-ending expertise and 
guidance. She would also like to thank Nicole Hanney for her assistance in conducting the 
experiment, and her committee members Drs. M. Christopher Newland, Jennifer Gillis-Mattson, 
and Linda LeBlanc for their constructive comments and advice. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
v 
 
 
 
 
 
 
Table of Contents 
 
 
Abstract ......................................................................................................................................... ii 
Acknowledgments........................................................................................................................ iv  
List of Tables ............................................................................................................................... vi  
List of Figures ............................................................................................................................. vii  
List of Abbreviations ................................................................................................................. viii  
Chapter I An Investigation of General and Descriptive Praise in Teaching Language to  
 Children with Autism   ...................................................................................................... 1 
Chapter II Method ....................................................................................................................... 16 
Chapter III Experimental Results................................................................................................ 30 
Chapter IV Discussion ................................................................................................................ 37 
References   ................................................................................................................................. 46 
Appendix A Experimental Visit Data Sheet ............................................................................... 64 
Appendix B IOA Data Sheet ...................................................................................................... 65 
Appendix C MSWO Data Sheet  ................................................................................................ 66 
Appendix D Praise Assessment Data Sheet  ............................................................................... 67 
Appendix E Skill Probe Data Sheet ............................................................................................ 68 
Appendix F 1-array MSWO Data Sheet ..................................................................................... 69 
Appendix G Procedural Integrity Data Sheet  ............................................................................ 70
vi 
 
List of Tables 
 
 
Table 1 Intraverbal Skills Targets for Instruction ....................................................................... 57
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
vii 
 
 
 
 
 
 
List of Figures 
 
 
Figure 1 Reinforcer (praise) assessment results for Brad, Shaun, Tina, and Brett  .................... 58 
Figure 2 Percentage correct during baseline, treatment, and follow-up conditions for Brad ..... 59 
Figure 3 Percentage correct during baseline, treatment, and follow-up conditions for Shaun ... 60 
Figure 4 Percentage correct during baseline, treatment, and follow-up conditions for Tina ...... 61 
Figure 5 Percentage correct during baseline and treatment conditions for Brett ........................ 62 
Figure 6 Number of sessions to meet mastery criteria and follow up ........................................ 63 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
viii 
 
 
 
 
 
 
List of Abbreviations 
 
 
ABA Applied Behavior Analysis    
ASD Autism Spectrum Disorder  
EIBI Early and Intensive Behavioral Intervention
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2 
 
 
1 
 
 
 
 
 
 
 
CHAPTER I 
An Investigation of General and Descriptive Praise in Teaching Language to Children with 
Autism 
 
Developmental Disabilities 
 Developmental disabilities are a diverse group of severe chronic conditions accompanied 
by social, communicative, neurological, behavioral, or physical impairments (Centers for 
Disease Control and Prevention [CDC], 2004). A developmental disability is considered 
developmental because delays, disorders, or impairments appear during the human 
developmental period that usually occurs between birth and 22 years of age (Odom, Horner, 
Snell, & Blacher, 2007). Individuals with developmental disabilities (e.g., cerebral palsy, 
intellectual disability, epilepsy, Down syndrome) may experience problems with major life 
activities such as language, mobility, learning, self-help, and independent living (CDC). While 
impairments in general intellectual functioning may be associated with and can be markers for 
developmental disabilities, individuals with developmental disabilities can have average or 
above average intelligence depending upon the severity and type of disability experienced 
(Odom et al.). One area of developmental disability that is not always accompanied by 
deficiencies in intellectual impairment are the autism spectrum disorders.  
 Autism is a life-long developmental disorder that appears by the age of 3 and is 
characterized by impairments in social and communicative behavior (e.g., gaze aversion, 
2 
 
mutism, language delays, absence of communicative gestures), as well as the presence of a 
restricted and repetitive range of behaviors, such as an insistence on sameness in routine and 
stereotypy (American Psychiatric Association, 2000). Autism (also referred to as autistic 
disorder), Asperger?s syndrome, and pervasive developmental disorder-not otherwise specified 
constitute what is commonly referred to as the autism spectrum disorders (ASDs) (CDC, 2006). 
Once thought to be rare, ASDs have recently been estimated to occur in approximately 66 per 
10,000 or 1 in 150 people in the United States (CDC, 2007).  
 Specific biological or biomedical etiologies of ASD have not yet been identified; 
however, research has suggested that genetic and/or environmental factors may be involved in 
their development (Nebel-Schwalm & Matson, 2008). To date, a number of therapeutic 
modalities have been developed for treating symptoms associated with the disorder. These 
include biological (e.g., use of medications such as risperidone), biomedical (e.g., use of vitamin 
supplements such as B6), and behavioral (e.g., use of operant teaching techniques) approaches to 
treatment (Hanson et al., 2007). While some practices have shown moderate efficacy and safety 
with some individuals with ASD (e.g., medical use of risperidone), other strategies (e.g., 
megavitamin therapy) lack the same scientific support (Romanczyk, Gillis, White, & Digennaro, 
2008). The approach to ASD treatment that has the strongest research support is early and 
intensive behavioral intervention (Eikeseth, 2009). 
Behavioral Intervention for Autism Spectrum Disorders 
 Researchers in the 1960s were among the first to demonstrate the potential of operant 
learning techniques in treating children with autism (e.g., Ferster & DeMyer, 1961; Lovaas, 
Berberich, Perloff, & Schaeffer, 1966; Wolf, Risley, & Mees, 1964). This research gave rise to 
the development of comprehensive applied behavior-analytic (ABA) interventions which focus 
3 
 
on targeting specific individualized behavioral objectives, utilizing ongoing data collection to 
measure and analyze treatment, examining environmental factors that influence behavior, 
developing function-based procedures for reducing problem behavior, and supervising and 
evaluating staff to ensure treatment integrity (Faja & Dawson, 2006; Steege, Mace, Perry, & 
Longenecker, 2007).  
 Current research on behavioral treatment for ASD suggests that early intensive 
intervention, begun when a child is between 2 and 5 years old, can have a significant and lasting, 
positive impact in the majority of children with ASD (Maurice, Green, & Luce, 1996). In 
addition, researchers have found that beginning behavioral intervention earlier in life (e.g., age 2) 
may produce greater positive outcomes than in later years (e.g., age 5) (Fenske, Zalenski, Krantz, 
& McClannahan, 1985; Harris & Handleman, 2000). Sometimes referred to as early and 
intensive behavior intervention (EIBI), this treatment approach is based on the principles of 
learning (e.g., reinforcement, discrimination, generalization) and it includes skill-based curricula 
that are hierarchically organized and developmentally sequenced with an intense delivery 
schedule of one-to-one instruction for up to 40 hrs per week over several years (Love, Carr, 
Almason, & Petursdottir, 2009).  
 Although applications of ABA methods for children with autism began in the 1960s, 
EIBI for autism gained notoriety when Lovaas (1987) treated children with autism intensively 
with ABA for several years and found significant improvements in 47% of the children receiving 
treatment. A number of large-n EIBI outcome studies have since been published.  These studies 
have documented improvement in ABA-treated groups compared to control groups not receiving 
the high-intensity intervention (Sherer & Schreibman, 2005). Other studies have also reported 
significant gains when children with ASD were treated intensively with ABA at an early age 
4 
 
(e.g., Eikeseth, Smith, Jahr, & Eldevik, 2007; Harris, Handleman, Gordon, Kristoff, & Fuentes, 
1991; Remington et al., 2007), when ABA was provided at low-intensity levels (12 hrs per week) 
(Eldevik, Eikeseth, Jahr, & Smith, 2006), and when intensive intervention was provided with 
older children aged 4 to 7 years (Eikeseth, Smith, Jahr, & Eldevik, 2002). In addition to these 
empirical efforts, multiple treatment and assessment manuals have been developed that outline 
EIBI strategies for young children with autism for disseminating the technology to parents and 
professionals (e.g., Leaf & McEachin, 1999; Lovaas, 2003; Maurice et al., 1996; Sundberg, 
2008; Sundberg & Partington, 1998). 
 Overall, the application of ABA-based treatment has been shown to increase 
communicative, social, play, cognitive, and other skills in children with ASD as well as reduce 
serious behavioral excesses, such as self-injurious behavior and aggression (Schreibman & 
Anderson, 2001). It is now widely acknowledged that, to date, the forms of treatment for ASD 
with the broadest empirical validation for effectiveness are based upon ABA (Schreibman, 
2000). Behavior-analytic intervention programs have been so successful in ASD treatment that in 
1999 the Surgeon General of the United States concluded that ABA was the treatment of choice 
for children with ASD (Rosenwasser & Axelrod, 2001). Furthermore, the results of a recent 
meta-analysis of EIBI programs found significant improvements in intelligence and adaptive-
behavior scores after receiving intensive ABA treatment across a number of well-designed 
studies, providing further evidence that EIBI should be the intervention of choice for children 
with ASD (Eldevik et al., 2009).  
Essential Components of Early and Intensive Behavior Intervention   
 While hundreds of studies have demonstrated the effectiveness of behavioral procedures 
in building important skills and reducing problem behaviors with individuals with autism, it is 
5 
 
possible that the implementation of EIBI by therapists in different settings (e.g., schools, clinics, 
special education classrooms) may vary (McClannahan & Krantz, 2004). Thus, researchers have 
begun investigating common elements found in EIBI. Early and intensive behavior intervention 
programs generally include the following important components: (a) intervention at an early age 
(before 3.5 years); (b) individualized and comprehensive intervention that addresses all skill 
domains (based on assessments of current skill levels); (c) use of behavioral procedures to build 
new skill repertoires and reduce interfering behavior (e.g., differential reinforcement, prompting, 
discrete-trial instruction, incidental teaching, activity-embedded trials); (d) one or more therapy 
agents with advanced training in ABA and experience with autism; (e) selection of intervention 
goals and objectives guided by developmental sequences; (f) parents serving as co-therapists 
with their children; (g) 1:1 intervention delivered with transitions to group instruction; (h) 
intervention beginning in the home with transition to community and school settings; (i) 
intensive, year-round programming with 20 to 30 hrs per week of structured sessions; and (j) 
duration of treatment of about 2 or more years (Green, Brennan, & Fein, 2002; Eikeseth, 2009; 
Eldevik et al., 2009).  
 McClannahan and Krantz (2004) also provided a view into the essential components of 
EIBI. According to McClannahan and Krantz, practices that contribute to EIBI success include: 
(a) providing the child the opportunity to engage in activities and with other people; (b) 
providing the child multiple opportunities to respond to instructional stimuli across settings and 
with multiple instructional strategies; (c) using descriptive praise contingent upon correct 
responding; (d) building relationships by pairing therapists with reinforcement; (e) including 
programs focused on hygiene and self-help skills; (f) working on social skills; (g) developing 
6 
 
prevention strategies for problem behavior; (h) involving the child?s family in intervention; and 
(i) intervening at an early age and at a high intensity.  
 Although EIBI is well researched and demonstrated to be effective, it appears that 
researchers differ in determining its essential components. Perhaps these differences are due to a 
number of variations in treatment priorities and procedures that have developed during the 
widespread dissemination of EIBI (Love et al., 2009). Love et al. surveyed 211 supervisors of 
EIBI programs to determine specific clinical practices in use across providers. The authors found 
the majority of program supervisors reported using more than one EIBI treatment manual for 
developing their program curricula, suggesting that no single EIBI manual met all program 
needs. In addition, they found a number of other differences across providers, including the use 
of preference assessments, task interspersal methods, the involvement of parents, and the setting 
in which services were provided (e.g., home, school, clinic). Thus, it is evident that EIBI 
programs employ a number of different techniques, and researchers and clinicians may hold 
different views about which components are most important to the overall enterprise. However, 
one EIBI component that appears to be used consistently across settings and providers is the 
reinforcement of adaptive behavior. One type of reinforcer commonly recommended for use in 
EIBI programs is praise (McKlannahan & Krantz, 2004).  
Praise as an Important Teaching Component   
 From the science of behavior, it is well understood that reinforcement is necessary for 
individuals to acquire new behavior (Heflin & Alberto, 2001). Therefore, it is no surprise that 
praise, a potential form of social reinforcement, is a common element of most teaching strategies 
often employed by teachers, parents, peers, and employers following a desirable behavior 
(Dozier, 2006). According to Brophy (1981), to praise means to ?commend the worth of or to 
7 
 
express approval or admiration? (p.5) while expressing a positive affect (showing delight) and/or 
giving information to a person about the value of his behavior. 
 Early applications of praise showed its utility in increasing appropriate behavior of 
disruptive students (e.g., Hall, Lund, & Jackson, 1968) and increasing task engagement with 
children in general education classrooms (e.g., Broden, Bruce, Mitchell, Carter, & Hall, 1970). 
Later studies evaluated praise with students with emotional and behavioral disorders (e.g., 
Sutherland, Wehby, & Copeland, 2000), learning disabilities (e.g., Alber, Heward, & Hippler, 
1999), and developmental disabilities (e.g., Craft, Alber, & Heward, 1998; McGee, Krantz, & 
McClannahan, 1984). The main focus of recent studies (e.g., Alber et al.; Craft et al.) has been 
on increasing a teacher?s use of praise because earlier studies reported that regardless of support 
for the use of praise in classrooms, teachers were found to provide low rates of praise across 
several studies (range = 0.2 to 4.4 per hr) (Brophy, 1981).    
 Praise has been referred to as a naturalistic, cost effective, and nonintrusive intervention 
for increasing appropriate behavior (Sutherland et al., 2000). In addition to its effectiveness in 
teaching the populations mentioned above, studies have demonstrated that utilizing praise can 
increase a number of both academic and social behaviors (either used in isolation or in 
conjunction with other strategies such as a token economy). For example, praise has been shown 
to contribute to increases in work productivity and accuracy (Craft et al., 1998), verbal behavior 
(Sigafoos, Doss, & Reichle, 1989), academic performance (Good, Eller, Spangler, & Stone, 
1981), on-task behavior (Ferguson & Houghton, 1992), student attention (Broden et al., 1970), 
positive self-statements (Phillips, 1984), social skills (Foxx, McMorrow, & Schloss, 1983), and 
self-initiated behavior (Rigsby-Elderidge & McLaughlin, 1992). From these studies, it appears 
8 
 
that praise can be a critical component of teaching interventions for many students for whom 
praise functions as a reinforcer (Keller, Brady, & Taylor, 2005). 
Types of Praise 
 Researchers have distinguished between two types of praise statements, descriptive and 
general, in order to further evaluate the use of praise in teaching. Descriptive praise refers to a 
praise statement that identifies to the learner the behavior for which he is being praised1 (Reinke, 
Lewis-Palmer, & Martin, 2007). General praise refers to a verbal praise statement that does not 
specify to the learner the desired behavior for which he is being praised (Sutherland et al., 2000). 
According to Wright (2008), the main components of a descriptive-praise statement include (a) 
gaining the student?s attention, (b) recognizing the appropriate behavior, and (c) providing the 
praise statement. These components are necessary for a social interaction to qualify as 
descriptive praise or else the statement becomes a general-praise statement. 
 The difference between the two types of praise is illustrated in the following example. If 
a teacher would like Johnny to raise his hand more often to answer questions, she could praise 
him immediately after he raises his hand to answer a question. To use descriptive praise, the 
teacher could say ?Good job raising your hand to answer the question, Johnny.? To use general 
praise the teacher could say ?Good job, Johnny.? The defining difference between the two 
statements is that the general praise statement is nonspecific compared to the descriptive praise 
statement. 
 The majority of studies have evaluated general, rather than descriptive, praise.  In fact, all 
of the studies previously cited employed general praise.  However, there do exist a few examples 
of research on descriptive praise.  For example, Sutherland et al. (2000) examined the rates of a 
teacher?s use of descriptive praise with a classroom of nine students with emotional and 
9 
 
behavioral disorders. The authors found that students? on-task behavior increased when the 
teacher?s descriptive praise increased and decreased with the removal of descriptive praise, 
suggesting that descriptive praise was an effective component of the teacher?s instructional 
approach. Novak and Hammond (2001) evaluated the use of descriptive praise in conjunction 
with a self-reinforcement token economy on reading performance of children in a fourth-grade 
general-education classroom. Upon completing a reading problem, students were given a 
detailed description of a job well done and of future consequences for their performance, such as 
?You sure got a lot correct. You should get a good grade in reading this term.? Results showed 
that using descriptive praise in conjunction with the token economy resulted in greater academic 
gains compared to using the token economy alone in which tokens were delivered by either the 
teacher or the student (i.e., self-administered).  
 Chalk and Bizo (2004) investigated the effects of descriptive praise and positive praise 
(i.e., any kind of praise descriptive or general) on student on-task behavior, academic self-
concept, and task enjoyment with four classrooms of 109 children in both general and special 
education. The results of the study showed that both student on-task behavior and academic self 
concept (as measured by the ?Myself as Learner? scale) were increased during both the 
descriptive praise and positive praise conditions with a slightly greater increase when teachers 
were instructed to give descriptive praise. However, the authors found no effect of type of praise 
on the enjoyment ratings made by the students. Finally, Wright (2008) assessed the effect of 
adding descriptive praise to the Good Behavior Game and reported that the addition resulted in 
slightly greater on-task behavior of students across five elementary education classrooms.    
 Based on the assumption that descriptive praise does increase desirable behaviors, other 
studies have investigated ways to increase teachers? use of descriptive praise. Research has 
10 
 
shown that teachers? descriptive praise statements can be increased using an audiotape self-
evaluation intervention (Keller et al., 2005), using visual performance feedback (Reinke et al., 
2007), and by teaching the students to recruit descriptive praise from their teacher (Alber et al., 
1998; Craft et al., 1998). 
 Very few studies have evaluated descriptive praise with children with ASD; those that 
have included descriptive praise as a component of a treatment package. For example, McGee et 
al. (1983) included descriptive praise in a modified incidental teaching package and found 
increases in newly acquired language skills. In addition, Brown, Krantz, McClannahan, and 
Poulson (2008) used descriptive praise as part of a script-fading procedure that was shown to 
successfully teach scripts to children with ASD who attended community-shopping trips. 
Although descriptive praise may have been a critical component of these packages, the exact 
utility of descriptive praise with learners with ASD has not yet been clearly evaluated.    
 Overall, studies have showed some utility of using descriptive praise in a number of 
educational contexts, yet research specifically evaluating descriptive praise is particularly 
lacking. In comparison, there are a greater number of studies investigating general praise than 
descriptive praise. Despite the lack of evidence, recommendations for using descriptive praise 
flourish, especially those that suggest it is superior to general praise.  
Recommendations to use Descriptive Praise  
 In comparison to descriptive praise, general praise may be perceived as vague and 
insincere and researchers have reported that identifying the performed behavior is an important 
component of effective praise (Brophy, 1981). Brophy reviewed the use of praise by teachers and 
derived several guidelines for using effective praise. His primary recommendations for 
delivering praise were to ?specify the particulars of the accomplishment? being praised and to 
11 
 
provide ?information to students about their competence or the value of their accomplishment? 
(p. 26). Alternatively, he stated that ineffective praise was characterized by global positive 
reactions, bland uniformity, and no information at all about performance. Thus, Brophy strongly 
recommended using descriptive praise over general praise during instruction and his assertion 
has been widely cited by other researchers (Chalk & Bizo, 2004; Keller et al., 2005; Shute, 2008; 
Sutherland et al., 2000). 
 In a review of evidence-based practices for classroom management, Simonsen et al. 
(2008) recommended that teachers should use a continuum of strategies to acknowledge 
appropriate behavior in the classroom. One of these was the use of specific, contingent praise 
because ?the effects of praise may be bolstered when the praise is specific (i.e., describes the 
desired behavior) and used in conjunction with other strategies? (p. 363). Similarly, Heflin and 
Alberto (2001) recommended that reinforcement be specific to a response when teaching 
individuals with ASD, implying that descriptive praise would be more appropriate than general 
praise for teaching new skills to this population.  
 Recommendations for the use of descriptive praise over general praise are prevalent in 
the ASD literature, in particular among EIBI manuals and web sites related to ASD and ABA. In 
a popular EIBI manual, Leaf and McEachin (1999) recommended the following in regards to 
using descriptive praise (referred to as ?labeling behavior?): 
 In early teaching, label the behavior that is being reinforced. This helps your child 
understand the behavior that is being reinforced and that you would like him to repeat. It 
also strengthens the connection between the reinforcer and the behavior. Labeling the 
behavior serves as a prompt for yourself as the teacher, keeping you focused on the 
12 
 
purpose of trials. Later on it will be less important to specifically label the behavior 
because your child will be able to understand the contingencies. (p. 33) 
Another example can be found in a chapter by Anderson, Taras, and Cannon (1996) that can be 
found in another popular clinical manual (Maurice et al., 1996): 
Praise is the most natural, convenient, and universal reinforcer that you can use. 
Unfortunately praise is not naturally reinforcing to many children with autism. 
Nevertheless by providing it simultaneously with the delivery of small portions of food or 
other reinforcing consequences, it also may gain reinforcement value to the child. Make 
your praise statement very specific and descriptive of the target behavior (e.g., ?Good 
sitting? vs. ?Good boy?). (p. 187) 
 In addition to recommendations found in EIBI manuals, a number of web sites advocate 
for descriptive praise compared to general praise. For example, the Behavior Analysis 
Association of Michigan website (n.d.) provides an example of how descriptive praise may be 
commonly explained to parents, teachers, or professionals: 
Descriptive praise is better than simple praise because it increases the duration and 
quality of the praise, and identifies the behavior being reinforced. By using the person's 
name, you increase the chances that the person will notice that you are delivering a verbal 
reinforcer and directly engage with you. If you are reinforcing the behavior of a person 
who is non-verbal, the descriptive praise will attach verbal labels to the activity, perhaps 
teaching the person more words. Even a completely non-verbal person will benefit. 
Longer social contact is likely to be more reinforcing than a brief contact (even if the 
person doesn't understand). 
13 
 
 It is evident that descriptive praise is widely recommended over general praise and this 
notion has become what appears to be a natural assumption of EIBI practice. However, all 
recommendations for descriptive praise over general praise have been made without sufficient 
evidence to support them; thus, questions remain about the scientific basis of the 
recommendation for descriptive over general praise. 
 The Scientific Basis of Descriptive Praise Recommendations 
 As mentioned previously, it appears that the most salient recommendation for using 
descriptive versus general praise originated with Brophy (1981). However in his research review, 
Brophy cautioned his readers that most of the data he reviewed were correlational and that 
inferences he made regarding the effectiveness of praise were often drawn from teachers? and 
students? thoughts and behavior rather than direct evidence. He referred to his paper as a ?logical 
analysis based on integration of a broad range of indirect data and not an empirical? analysis of 
praise (p. 25).  
 Although Brophy (1981) did not have direct evidence to support his recommendations, 
the notion of descriptive praise being superior to general praise is commonplace, even today 
(e.g., Shute, 2008). However, few studies have directly evaluated the use of descriptive praise 
versus general praise in an educational context (and even moreso with children with 
developmental disabilities). As mentioned previously, Chalk and Bizo (2004) evaluated different 
types of praise, but weaknesses of their study (e.g., unclear differentiation of descriptive versus 
general praise, lack of treatment integrity measures, and the inclusion of only four observations 
per teacher) make it impossible to clearly interpret their findings.  
 Few studies exist (published and unpublished) that have compared the effects of 
descriptive and general praise (Scheer, 1976; Sellers & Higbee, 2009; Stevens et al., 2010; 
14 
 
Zahler, 1975). Both Scheer and Zahler were unpublished dissertations in which both authors 
evaluated descriptive and general praise in simple motor tasks and found minimal if any effects 
of descriptive praise over general praise on student performance (i.e., with typically-developing 
4th graders and college students). Recently, Stevens et al. evaluated the effects of descriptive and 
general praise on the acquisition of tacts with 2 children diagnosed with ASDs (i.e., autism and 
PDD-NOS). In the study they paired descriptive and general with tokens and prompting and 
found no differential effects of either type of praise on the acquisition of skills. From their 
evaluation it is unknown if the type of praise would have had a different effect on learning if 
presented alone. Sellers and Higbee (2009) evaluated descriptive and general praise with 3 young 
children with autism and also found no greater effect of either the use of descriptive or general 
praise on the rate of skill acquisition of receptive actions (i.e., auditory-visual conditional 
discriminations). Specifically, 2 of the 3 children did not acquire many of the targeted skills 
demonstrating the type of praise was equally ineffective (or that praise was not an effective 
reinforcer). The 3rd participant in their study mastered the same number of skills with both types 
of praise, further showing no difference when learning was achieved. 
Purpose of the Current Study  
 There is a clear need for additional research into the use of descriptive praise with 
children in an instructional context, in particular those with ASD. From the present review, it 
appears that recommendations for using descriptive praise with children with ASD as part of an 
EIBI program have exceeded any evidence collected to support them. The purpose of the present 
study, therefore, was to investigate the effects of descriptive praise and general praise on the 
acquisition of language skills of children diagnosed with ASD.  
15 
 
 To further investigate general and descriptive praise, the current study expanded upon 
Stevens et al. (2010) and Sellers and Higbee (2009) in a couple of ways. First, Stevens et al. 
evaluated the effects of descriptive versus general praise in conjunction with prompting and 
additional reinforcement (i.e., tokens). They did not directly evaluate the effects of either type of 
praise in isolation. Thus, the current study included an initial intervention phase of praise alone 
(either descriptive or general) plus prompting, which was followed by the introduction of 
additional reinforcement if learning was not achieved. Second, Sellers and Higbee included 
participants in their study for whom praise did not function as a reinforcer, as evidenced by 2 out 
of 3 participants? failure to reach skill mastery, and also found no effect of either descriptive or 
general praise on the rate of skill acquisition for a participant who mastered the same number of 
targets regardless of praise type. In the current study, an additional experimental condition (i.e., 
praise plus a highly preferred item) was included for children whom praise may not have been 
sufficient for increasing language performance in order to further determine any effects of praise 
type. Third, the participants in Sellers and Higbee were young learners (aged 3 and 4 yrs) with 
autism who may not have had a sufficient listener repertoire. From a conceptual perspective, one 
would not expect praise to be an effective reinforcer with young children who lack a strong 
listener repertoire because they would be unable to benefit from the added language in 
descriptive praise. Therefore, in the present study we also assessed the use of descriptive versus 
general praise in teaching language to learners who were young (between 3 to 4 years of age) but 
had a strong listener repertoire (as determined via direct assessment) in order to determine what 
effects these types of praise, if any, have on the acquisition of language skills.
  
16 
 
 
 
 
 
 
 
CHAPTER II 
Method 
Participants and Setting 
 Participants were selected from a local preschool that delivered EIBI services to students 
with and without developmental delays. Seven preschool students (six boys and one girl) 
diagnosed with an autism spectrum disorder were chosen to participate in the study based on 
parental consent and willingness to participate. Greg, Drew, Jesse, Brad, Shaun, Brett, and Tina 
were 3.7, 4.7, 4.11, 4.8, 3.9, 4.11, and 4.11 years of age respectively at the time of the study. 
Inclusion in the study required that participants (a) were able to follow simple directions and (b) 
did not have any significant impairment (e.g., severe problem behavior or limited verbal ability) 
that could hinder their participation in the study. Of the 7 children who enrolled in the study, 4 
participants (Brad, Shaun, Tina, and Brett) completed the study. The other participants did not 
finish the experiment due to switching schools and/or meeting exclusionary criteria after 
beginning the study (e.g., severe problem behavior).   
 All experimental sessions were conducted in a small therapy room at the pre-school 3 to 
5 days per week with an experimenter, a research assistant, and the participant present. The room 
contained a table, two large cabinets, and three chairs. The materials necessary for conducting a 
session (i.e., a video camera and data sheets) were set up in the room prior to participant entry. 
All instruction given during a session consisted of one-to-one teaching that took place seated at a 
table or sitting on the floor. The 
17 
 
experimenter conducted multiple trials per session and session duration was determined by the 
duration of time to complete nine trials. More than one session was conducted within an 
experimental visit and visits lasted 30 to 60 min. Sessions continued until a) the participant 
mastered the skill sets or b) until the participant reached 3 failures (as described in the 
procedures section below). In total, the study was conducted over a nine month period, with 
participants taking part in sessions for 12 to 20 weeks each.  
Data Collection 
 Throughout the study, each participant was exposed to sessions consisting of language 
trials in which the child was given an instruction (the discriminative stimulus, SD) and data were 
collected on the participants? responses emitted after the SD during each trial. Each response was 
scored as correct, incorrect, or ?no response.? A sample data sheet is depicted in Appendix A. A 
correct response was defined as the participant responding correctly and independently (without 
prompting) to the instruction within 5 s. All incorrect or non-responses within 5 s resulted in the 
presentation of a least-to-most prompting sequence (described later). All responses were 
manually recorded on a data sheet immediately after the participants? responses were emitted or 
up to 5 s after the participant was given the opportunity to emit a response (for prompted or non-
responses only).  
 One skill set was assessed during each session.  A skill set consisted of three specific 
language targets (i.e., intraverbals) presented for 3 trials each. Thus, each session consisted of 9 
trials.  Two dependent measures were derived from the data collected. The first was the 
percentage of correct responses per target skill set per session and the second was the number of 
sessions to reach the mastery criterion per target set (defined as three consecutive sessions at 
89% correct or better across two consecutive visits).  
18 
 
Interobserver Agreement 
 Interobserver agreement (IOA) was assessed for 33%, 37%, 48%, and 33% of the 
treatment-evaluation sessions for Brad, Shaun, Tina, and Brett respectively. To facilitate IOA, 
experimental visits were video recorded and later scored by a second trained observer. These 
data were compared to the in-vivo data collected by the experimenter who conducted the session 
to calculate point-by-point agreement (i.e., video data to live data; see Appendix B). An 
agreement was defined as the two independent observers agreeing on the events that occurred 
during the session (e.g., the participants? responses and the prompts given by the experimenter). 
Interobserver agreement was calculated by dividing the total number of agreements by the total 
number of agreements plus disagreements and converting the ratio to a percentage.  The mean 
and range of IOA scores for each participant were as follows: Brad 97% (range, 85-100%), 
Shaun 100% (range, 100-100%), Tina 98% (range, 89-100%), and Brett 100% (range, 100-
100%). Procedures 
 Initial assessments. The experimenter conducted four initial assessments with each 
participant prior to the experimental evaluation. First, a stimulus preference assessment was 
conducted to identify preferred items and activities that could serve as reinforcers to be used in 
the experiment. Second, a language assessment was used to determine each student?s eligibility 
to participate in the study along with their current language abilities.  Third, a reinforcer 
assessment was conducted to evaluate the effectiveness of praise as a reinforcer. Fourth, skill 
probe assessments were used to determine specific language skill targets to teach during the 
study. All assessments took place in the same location as the experimental sessions and included 
the same personnel. 
19 
 
 Stimulus preference assessments.  Upon receiving parental consent for research 
participation, a stimulus preference assessment was conducted with each participant. Interviews 
were conducted with the participants? parents, therapists, or teacher to identify potentially 
preferred toys and leisure items that could be further assessed during a subsequent direct-
observation preference assessment. After potentially preferred items were identified, a subset of 
them were then included in a brief multiple-stimulus without replacement (MSWO) preference 
assessment (Carr, Nicolson, & Higbee, 2000).  
 During the MSWO assessment, toys from both the interviews and novel classroom toys 
were arranged in a semi-circle on the floor in front of the child. The child was directed through 
the array and all toys were introduced and demonstrated. The child was then instructed to select a 
toy by the experimenter instructing him to ?Pick one,? or asking ?What do you want?? If the 
child failed to respond within 5 s the instruction was repeated and any attempt to pick more than 
one item was blocked. Upon making a choice, the participant received access to the selected item 
for up to 30 s and the experimenter recorded the selection on a data sheet (see Appendix C). 
When an item was chosen it was removed from the array and the remaining items were 
rearranged quasi-randomly. A second trial then took place during which the child was asked to 
select another item. The trial ended with access to the item chosen for up to 30 s and removal of 
that item as described above. Trials were repeated until all items were selected or until the child 
stopped selecting. The entire MSWO procedure was repeated two additional times to determine 
each child?s relative preferences among the included items.  The results of the MSWO 
preference assessment were used to select items to include in the experiment during play breaks 
between sessions and the entire assessment was completed within one experimental visit.  
20 
 
 Language assessment.  The Verbal Behavior Milestones Assessment and Placement 
Program (VB-MAPP) (Sundberg, 2008) was used to assess each child?s language skills. The VB-
MAPP is a comprehensive criterion-based language assessment that allows a tester to identify 
language abilities based on Skinner?s (1957) analysis of verbal behavior.  The VB-MAPP was 
either reviewed (if completed within 1 month of beginning the study) or conducted by the 
experimenter. The VB-MAPP was used to determine if the participant met criteria for continuing 
in the study and it also aided in determining the child?s language skill areas in need of further 
instruction. During visits in which the VB-MAPP was completed the experimenter directly 
assessed areas of the VB-MAPP with the participant during brief 5-min sessions with play breaks 
in between.  
 In each VB-MAPP trial, the child was given an instruction (the SD) and data were 
collected on the participants? responses emitted after the SD.  Each trial was scored as correct, 
incorrect, or ?no response? and the child?s responses were coded according to the point values 
given in the VB-MAPP protocol. The results of the VB-MAPP are scored against three 
developmental levels (Level 1: 0-18 mos., Level 2: 18-30 mos., Level 3: 30-48 mos.).  For the 
present study, the experimenter used the VB-MAPP to evaluate the following repertoires: 
listener responding (following instructions), echoics (vocal imitation), and intraverbals 
(elementary conversational skills). 
 Upon completion of the assessment, the experimenter was able to determine that all 
participants (Brad, Brett, Shaun, and Tina) were eligible to take part in further phases of the 
study. The participants? performance in a remaining area of the VB-MAPP (intraverbals) was 
later used to determine target skill sets to teach each child during the experimental visits.    
21 
 
 Reinforcer (praise) assessment. To assess whether praise functioned as a reinforcer for 
the children?s behavior, each participant was instructed to engage in a simple task unrelated to 
the other tasks in the experiment. The responses targeted for this assessment were a) placing a 
plain white sticker (i.e., an adhesive paper reinforcement label) on a piece of construction paper 
for Brad, Tina, and Brett and b) putting blocks in a box for Shaun. Responses were reinforced on 
a progressive ratio (PR) 1 schedule. See Appendix D for a sample data sheet. In the praise 
assessment three different conditions were assessed using a multielement design. The first 
condition was a general praise condition in which the participant was provided a general praise 
statement (?Good job?) when the schedule requirement was met. The second condition was a 
descriptive praise condition in which the participant was provided a descriptive praise statement 
(?Good job putting the sticker on the paper?) when the schedule requirement was met. The third 
condition was a no-praise condition which served as a control condition because no programmed 
consequences were delivered for responding. Each session began with the experimenter 
modeling the required response (putting a sticker on a piece of paper for Brad, Brett, and Tina 
and putting a block in a box for Shaun) and providing a verbal instruction ?Do this.? 
Programmed responses were then delivered per the experimental condition. Sessions ended (a) 
after 5 min had elapsed), (b) until the participant made four attempts to leave his or her seat, or 
(c) after 2 min of no responding elapsed. 
 Skill probe assessment. Using the VB-MAPP as a basis of skill level, the experimenter 
developed several skill lists per child (consisting of multiple intraverbal targets) that were 
equated for syllable length and then assessed which of the skills in the lists were or were not in 
the child?s repertoire by providing an opportunity to perform each skill in the set (see Appendix 
E for a sample data sheet). These probe sessions took place in a loosely structured one-to-one 
22 
 
teaching context either seated at a table or on the floor, or during play breaks where the child was 
given free access to toys with trials presented intermittently during play. Probe session visits 
lasted between 30 and 60 min in duration and multiple trials were conducted per session. 
 During each probe trial, the experimenter presented the instructions necessary for the 
child to complete each skill target (e.g., asked ?What animal says ?Moo?? for assessing an 
intraverbal skill). If a child was able to independently complete any skill in the skill set (e.g., said 
?cow? when asked ?What animal says ?Moo??) it was removed from the probe list and new skills 
were targeted and probed in the same fashion (e.g., ?What animal says ?Quack??). During each 
probe an instruction was presented and data were collected on the child?s response (correct, 
incorrect, or no response). The consequence provided for both a correct and incorrect response 
was a nonspecific statement redirecting the child to the next trial (e.g., ?Let?s try this one? or 
?Now it?s time to try this?). If the child made no response, the experimenter waited 5 s before 
providing a verbal redirection to the next trial.  
 The skill-probe trials were presented quasi-randomly until 18 skill targets were identified 
as not being in the child?s repertoire (i.e., the child could not answer correctly; required a 
minimum of 54 total trials). The criteria for determining which skills could not already be 
performed by the child were 3 consecutive probes in which the child could not independently 
emit the response or emitted an incorrect response. When all six three-stimulus sets were 
identified, baseline sessions began.  
 Treatment evaluation. After the initial pre-experimental assessments were completed, 
each participant took part in an experimental treatment evaluation to determine whether the type 
of praise would have an effect on the acquisition of language skills.  
23 
 
 Experimental design. To assess the effects of general and descriptive praise on skill 
acquisition, an adapted alternating treatment design (AATD) (Sindelar, Rosenberg, &Wilson, 
1985) embedded within a multiple-probe design (for within-participant replication) was used. An 
AATD requires that two equivalent sets of instructional targets be taught with two different 
instructional strategies rather than applying multiple interventions to the same response class as 
in a traditional alternating treatments design. Target skill sets were equated for difficulty (i.e., 
required verbal responses were equated for syllable length) and each skill set was taught using 
either general or descriptive praise in addition to a least-to-most prompting procedure. 
 The AATD embedded within a multiple-probe design consisted of teaching different 
target skill sets to each participant. All participants had six skill sets chosen for the study (see 
Table 1). Target sets were presented in pairs of two (one for each praise intervention) in the 
AATD with baseline levels established for all pairs of sets concurrently. The introduction of the 
two teaching interventions (general and descriptive praise) began with the first pair of sets while 
baseline sessions continued to be probed for the other existing target sets. After the first set of 
skill targets showed a marked increase in acquisition (as determined by visual inspection of 
graphical data), the teaching interventions were then introduced with the second pair of target 
sets and so on. For all target sets, teaching continued until criteria were met for mastery or failure 
(described below). 
 General session structure. Each session assessed performance for one target skill set at a 
time and included 3 trials for each target skill in the set for a total of 9 trials. Between sessions 
the participant was provided a play break of 1 min during which he or she had access to 
moderately preferred toys and items as identified from his stimulus preference assessment. After 
24 
 
the play break, another session was conducted assessing either the same or a different target set. 
The order of presentation of the target sets was determined quasi-randomly  
 Baseline. For all target sets, baseline sessions consisted of the presentation of trials in 
which an SD was provided (e.g., an instruction such as ?What animal says ?Moo???) followed by 
the presentation of a programmed consequence for the participant?s response. In baseline the 
consequence for responding consisted of a verbal redirection (not a praise statement) to the next 
trial (e.g., ?Let?s move on? or ?Time to try this?). If a participant provided no response 
immediately after the SD was presented, the experimenter waited 5 s before redirecting to the 
next trial. 
 Prompting plus praise. During this condition, two target skill sets were taught using two 
separate prompting plus praise strategies. Sessions were conducted in the same manner as 
described in baseline except for the programmed consequences that followed the participants? 
responses. In the prompting plus praise condition the consequence for correct responding 
(provided immediately) or prompted responding (provided after prompting/error correction) 
consisted of an excited praise statement. When the praise strategy consisted of general praise, the 
praise statement included an enthusiastic statement unrelated to the target response (consisting of 
two words), such as ?Good job!? or ?Great job!? When the praise strategy consisted of 
descriptive praise, the excited praise included a statement related to the target response that 
reiterated what the child did correctly, such as ?Great job saying cow!? or ?Good job saying 
horse!?  
 If in either praise condition a participant provided no response or an incorrect response 
immediately after the SD was presented, the experimenter implemented a least-to-most 
prompting procedure in which the experimenter waited 5 s for a response before providing the SD 
25 
 
again in conjunction with a verbal prompt. The first prompt was a partial-word verbal prompt in 
which part of the correct word was provided as the prompt (e.g., saying ?c-? for cow). If the 
child did not respond within 5 s after the partial-word prompt a full-word verbal prompt was 
provided, which consisted of providing the full correct word as the prompt (e.g., ?cow?). A 
differential reinforcement procedure was included in this procedure in which correct responses 
given after the partial-word prompt resulted in the same general or descriptive praise statement 
as mentioned above and correct responses after full-word prompts resulted in no programmed 
consequence and the start of the next trial.  
 The prompting with praise condition continued until both target sets reached mastery 
criterion (89% correct or better for three consecutive sessions over two visits) or until both target 
sets reach the failure criterion, which was defined as 10 consecutive sessions with no increasing 
trend. In instances where one set reached mastery before the other set, sessions ceased for the 
mastered set but continued with the nonmastered set until that set either reached the mastery or 
failure criterion. If both target sets reached the failure criterion, a new phase was introduced. 
 Prompting with praise plus a highly preferred item (Brett and Tina). If a participant met 
the failure criterion in the first intervention phase, it could have been that social praise was not an 
effective reinforcer. Therefore, a new phase was implemented in which highly preferred toys 
were added to the prompting plus praise intervention (for Brett and Tina). The pairing of highly 
preferred items was used because it is a commonly recommended EIBI strategy (e.g., Lovaas, 
2003; Partington, 2006). In this condition, trials were conducted in the same manner as described 
above for the general and descriptive praise strategies with a few exceptions. Prior to each visit, a 
brief 1-array MSWO preference assessment (Carr et al., 2000) was conducted to identify items 
that were highly preferred for that day?s visit (see Appendix F for a sample data sheet). The 
26 
 
highly preferred items were then provided in conjunction with the praise statements contingent 
upon correct responding. Brett?s common highly preferred items included superhero figurines 
and stuffed animals and Tina?s common highly preferred items were bubbles and a photo book. 
In this phase a differential reinforcement procedure was used, as described above. However, in 
this condition independent correct responses resulted in praise plus the additional item, responses 
following a partial-word verbal prompt resulted in praise alone, and responses following the full-
word verbal prompt resulted in no programmed consequence and the introduction of the next 
trial.    
 Errorless prompting with praise plus highly preferred items (Brett and Tina). In the 
event that a participant met the failure criterion for a target set beyond the introduction of the 
highly preferred item, an additional phase was implemented that consisted of using a most-to-
least prompting procedure. During this phase a brief 1-array MSWO was implemented at the 
beginning of the visit and highly preferred items were paired with the praise statements as 
described above. Additionally, each session began with an errorless training session in which the 
SD was presented (e.g., ?What animal says ?Moo???) immediately followed by the experimenter 
giving the full-word prompt of the correct answer (e.g., ?Cow?). After the full word prompt was 
given, the participant said the answer back to the experimenter. If he or she did not, the 
experimenter waited 5 s, said nothing, and introduced the next target. If the participant responded 
incorrectly to the full-word prompt (e.g., said ?Pig? when given the prompt ?Cow?) the 
experimenter provided one additional full word prompt (e.g., ?Cow?) and the participant said the 
correct word back to the experimenter. If he or she did not respond or responded incorrectly, the 
experimenter waited 5 s, said nothing, and introduced the next target. It should be noted that 
participants never responded incorrectly to the full-word prompt.  This continued with all 9 trials 
27 
 
in the set. After the errorless training session, the experimenter provided a probe of each target in 
the set using the least-to-most prompting procedure described above, except in this phase the 
experimenter provided the SD and followed through with the least-to-most prompting strategy for 
each target only once (as opposed to 3 times each as in previous phases). For these least-to-most 
probes (consisting of 3 trials), the same differential reinforcement procedure was used as 
discussed in the previous phase.  
 Prompting plus neutral statement (Brad only). In the event that there was no clear 
difference between descriptive or general praise (as evidenced in Brad?s evaluation) an 
additional phase was implemented during which a neutral statement was paired with the 
prompting strategy. Sessions were conducted in the same manner as the prompting plus praise 
condition, except that the consequence delivered contingent upon correct responding was a 
neutral statement, such as ?Let?s move on? or ?Let?s try this.? Additionally, the differential 
reinforcement component included the neutral statement for independent correct responses and 
correct responses after the partial word prompt was provided. For responses given after the full 
word prompt, nothing was said and instead the next trial was presented.   
 Follow-up assessment. Three to four weeks after each participant?s completion of the 
study (by reaching mastery or the failure criterion) follow-up assessments were conducted. 
During follow-up, the experimenter conducted one session with the child similar to baseline 
condition, except that the session consisted of only one presentation of each target (3 trials). In 
each trial, the corresponding SD was provided (e.g., an instruction such as ?What animal says 
?Moo???) and data were collected on the child?s response (either correct, incorrect, or no 
response; as defined above). The consequence provided by the experimenter for correct and 
incorrect responses was a neutral statement (e.g., ?Let?s move on? or ?Time to try this?). If a 
28 
 
participant provided no response immediately after the SD was presented, the experimenter 
waited 5 s before beginning the next trial.  
Procedural Integrity 
 To assess procedural integrity, a second, trained observer collected data on the 
experimenter?s behavior to determine whether the experimental procedures were implemented 
correctly. Specifically, procedural data were collected on the following: (a) type of praise given, 
(b) enthusiasm in teaching, and (c) correct implementation of teaching protocol. In assessing the 
correct implementation of the teaching protocol, the trained observer watched video-taped 
sessions and assessed whether the experimenter presented the SD correctly throughout the session 
according to the target set being taught (e.g., household items, body part functions). The observer 
also assessed whether the experimenter provided the least-to-most prompting sequences correctly 
within 5 s of the participants? responses. The observer also assessed whether the correct 
consequences were provided (e.g., neutral statement in baseline, general praise in the general-
praise condition). Additional detail on procedural integrity skills can be found on the a sample 
data sheet in Appendix G.  
 A procedural integrity score was calculated for 36%, 33%, 33%, and 33% of the 
treatment-evaluation sessions for Brad, Shaun, Tina, and Brett respectively. The procedural 
integrity score was calculated as the percentage of correct responses made by the experimenter 
during a session (out of a total of 46 possible responses). The mean and range of integrity scores 
in each participant?s evaluation were as follows: Brad 99% (range, 95-100%), Shaun 100%, Tina 
99% (range, 97-100%), and Brett 100%. 
 In addition, IOA was assessed for 100%, 100%, 38%, and 50% of the integrity data 
collection sessions for Brad, Shaun, Tina, and Brett, respectively. During IOA checks, a trained 
29 
 
observer collected procedural integrity data and compared those data to the integrity data 
collected by the second, trained observer. Data were compared using the point-by-point 
agreement formula. An agreement was defined as the two independent observers agreeing on 
whether the experimenter?s response during an aspect of the trial (e.g., presentation of 
instruction, consequence delivery, etc.) was correct or incorrect. The mean and range of IOA 
scores for procedural integrity measures for each participant were as follows: Brad 98% (range, 
89-100%), Shaun 100%, Brett 100%, and Tina 100%. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
30 
 
 
 
 
 
 
CHAPTER III   
Results 
Reinforcer (Praise) Assessment 
 Results of the pre-experimental praise assessments for all participants are depicted in 
Figure 1. Across participants there was no clear differentiation of effects of the descriptive praise 
(DP), general praise (GP), or no praise (control) conditions on responding. Brad?s average break 
point were 5.3 (range, 5-6) when descriptive praise was provided, 4 responses (range, 1-6) when 
general praise was provided, and 4.6 responses (range, 3-6) when no praise was provided for 
placing stickers on a piece of paper. Shaun engaged in higher rates of responding, but also 
showed no differentiated responding across conditions. His average break points were 9.3 (range, 
7-12) when descriptive praise was provided, 10 (range, 9-12) when general praise was provided, 
and 10 responses (range, 8-12) when no praise was provided for putting blocks in a box. 
Similarly, Tina emitted stable, undifferentiated responding in the assessment. Her average break 
points were 9 (range, 8-10) when descriptive praise was provided, 9.5 (range, 9-10) when general 
praise was provided, and 9 (range, 8-10) when no praise was provided for sticker placement. 
Finally, Brett emitted similar patterns of responding with average break points of 8 (range 8-8) 
when descriptive praise was provided, 8.5 (range, 8-9) when general praise was provided, and 
8.6 (range, 8-10) when no praise was provided for sticker placement. Thus, from these 
assessments it appears that general and descriptive praise were equally ineffective in elevating 
response rates above those obtain in the absence of programmed reinforcement. 
 
31 
 
Treatment Evaluation 
 Brad. The percentage of correct intraverbals emitted by Brad during the treatment 
evaluation is shown in Figure 2. During baseline, Brad emitted none of the answers to the 
targeted intraverbal ?What? questions correctly. Upon implementation of the prompting-plus-
praise condition, both the descriptive and general praise conditions produced acquisition to 
mastery of the body-part and household-item intraverbal sets (i.e., sets 1-4; top 2 panels). 
Moreover, Brad?s responding was comparable under the descriptive and general praise 
conditions, showing no clear differentiation between the two types of praise on the rate of 
acquisition, which was similar to the results of Brad?s praise assessment. Additionally, the 
number of sessions required for him to master sets 1-4 was similar across descriptive and general 
praise conditions. For the comparison of set 1 versus set 2 (top panel; body part functions) he 
mastered set 1 in 10 sessions with general praise and set 2 in 8 sessions with descriptive praise. 
This finding was replicated in the middle panel (household items) with mastery obtained within 8 
sessions with general praise (set 3) and 6 sessions with descriptive praise (set 4). Follow-up data 
showed that his mastery of skills in sets 1, 2, and 3 maintained at 100% correct, while 
performance on set 4 decreased to 66% under descriptive praise. Although he was able to master 
skills slightly more quickly with descriptive praise (M=7 sessions; range, 6-8) than general praise 
(M=9 sessions, range, 8-10), the difference between the effects of each type of praise was 
minimal and this finding was replicated across both evaluations. 
 The bottom panel of Figure 1 shows Brad?s acquisition of the skills in sets 5 and 6 under 
the prompting-plus-neutral-statement intervention, which produced acquisition to mastery 
criteria for both sets, with maintenance at 100% for both sets during the follow-up evaluation. 
Brad mastered set 5 in 14 sessions and set 6 in 23 sessions, which differed greatly from his 
32 
 
performance in the praise interventions when it took an average of 8 sessions for his responding 
to reach mastery (range, 6-10). Thus, in comparison to his performance when praise was 
provided for correct responding, his rate of skill acquisition was much slower. This finding also 
shows the utility of prompting for skill acquisition, as Brad was able to learn the skills with 
prompting and differential neutral consequences, but it appears that praise may be important for 
increasing instructional efficiency. 
 Shaun. The percentage of correct intraverbals emitted by Shaun during the treatment 
evaluation is shown in Figure 3. During baseline, Shaun never responded correctly to the ?What? 
questions presented. Upon implementation of the prompting-plus-praise condition, both 
descriptive and general praise led to mastery for all target sets. In the first comparison (top 
panel), Shaun emitted similar patterns of responding across descriptive and general praise 
conditions. His responding in both sets reached mastery in the same number of sessions (17), 
however only descriptive praise resulted in strong maintenance of skills at follow up (100% 
correct for set 2 [descriptive] versus 33% for set 1 [general]). In the second comparison (middle 
panel), Shaun?s rate of acquisition was faster with descriptive than general praise, and he reached 
mastery of set 4 (descriptive praise) within 12 sessions, more quickly than his mastery of set 3 in 
general praise (20 sessions). Both interventions resulted in high performance at follow up (88% 
and 100% correct for sets 3 and 4, respectively).  
 In the third comparison (bottom panel), Shaun?s performance was comparable across 
praise conditions. His responding reached mastery criteria in 13 sessions with descriptive praise 
(set 6) and 16 sessions with general praise (set 5), and his performance at follow up was 
maintained at mastery level (88%) for both conditions. Although Shaun was able to learn all 
targeted skill sets to mastery, the effects of the type of praise on his acquisition performance 
33 
 
were varied. The first comparison showed equal effects (as was also shown in his praise 
assessment); however, the second comparison showed an advantage of descriptive over general 
praise (a 12-session difference). Finally, the third comparison showed only a slight (3 session) 
advantage of descriptive praise.  
 Tina. Data from Tina?s treatment evaluation are depicted in Figure 4. During baseline, 
Tina emitted zero intraverbal fill-in phrases correctly. In the first comparison (top panel; 
household items) Tina?s responding showed very slow and variable acquisition upon 
implementation of the prompting-plus-praise interventions. Although the descriptive praise 
intervention initially produced higher acquisition than general praise (until session 33), her 
responding eventually met failure criteria in both praise conditions after equal exposure to the 
same number of sessions (21 sessions to failure). After contingent highly preferred items were 
added to the treatment package, Tina?s responding immediately met mastery criteria within 3 
sessions under the descriptive praise condition (set 2), while she emitted slow but steady 
acquisition in the general praise condition (set 1) which eventually reached the failure criterion 
(after 17 sessions). After implementation of the final phase in which errorless training was added 
to the treatment package, Tina quickly mastered the set after additional 6 sessions. Thus, in this 
comparison, it appeared that neither type of praise was effective alone as a reinforcer and 
required the addition of a highly preferred item to facilitate skill acquisition. For set 1, changing 
the type of prompting procedure eventually resulted in final mastery. At follow up, both skill sets 
maintained at 66% correct. 
 In the second comparison (middle panel; body part functions), Tina?s responding under 
both descriptive and general praise interventions showed very similar patterns, often 
overlapping. Although general praise appeared to produce a similar rate of acquisition, Tina 
34 
 
mastered set 3 (general praise; 17 sessions) more quickly than set 4 (descriptive praise; 23 
sessions). At follow up, her responding remained at mastery (100%) for the general praise set 
and dropped to 33% for the descriptive praise set. Finally the third panel served as a control 
condition (baseline only) which showed no skill acquisition without any teaching intervention in 
place.  
 Tina?s treatment evaluation results showed inconsistencies regarding the effects of praise. 
In the first comparison, it appeared that descriptive praise produced faster mastery (with 
additional tangible reinforcement), but in the second comparison general praise produced slightly 
faster mastery. Thus, there did not appear to be clear and consistent beneficial effects of one type 
of praise over the other. 
 Brett. Brett?s treatment evaluation data are presented in Figure 5. During baseline, Brett 
emitted none of the intraverbal math addition fill-in phrases correctly, with the exception of 
session 97 in panel 3, when he answered one phrase correctly (responded ?23? to the phrase 
?20+3 __?). In the first comparison (top panel; ?ones?), prompting and praise failed to produce 
responding much higher than baseline (22% correct for general praise and 11% correct for 
descriptive praise). After the addition of contingent high-preference items, Brett?s responding 
showed very slow acquisition with no differential effects between praise conditions. Both praise 
interventions produced similar trends of responding (often overlapping), eventually meeting the 
failure criterion after 24 sessions in each praise condition. The third and final phase included an 
additional errorless training component.  This change produced additional acquisition, but the 
mastery criterion was not met, even when the target behavior was modified to reduce the 
likelihood of erroneous stimulus control.  Additionally, there was no difference in the effects of 
descriptive and general praise. Brett?s responding eventually met the failure criterion after 21 
35 
 
sessions in each of the praise interventions and his percentage correct scores did not exceed 66% 
in either condition. After this final phase, sessions were discontinued per final failure criteria 
(mentioned previously).  
 Brett?s second praise comparison (middle panel; ?tens?) produced similar findings. 
Making use of the findings from Brett?s first evaluation, the initial treatment package in the 
second evaluation included errorless training and contingent high-preference items, in addition to 
praise. Although there was some increase in responding, the failure criterion was eventually met, 
with percentage correct scores never exceeding 66%.   Overall, Brett?s treatment evaluation 
showed that there was no difference in the effects of descriptive versus general praise on the 
acquisition of intraverbal fill-in phrases, a finding consistent with that of the other participants. 
However, with Brett there was no differential effect between praise conditions even when skills 
were not mastered. 
 Summary. Figure 6 depicts a summary of all treatment-evaluation findings. The number 
of sessions to meet the mastery criterion is displayed in the top panel. Although the effects of the 
specific type of praise on skill acquisition appear to be idiosyncratic, there were minimal, if any, 
advantages of descriptive praise in acquisition efficiency. In 5 out of 9 comparisons (55%), 
participants reached mastery faster with descriptive praise; however, the differences were often 
minor. In 3 out of 9 comparisons (33%), participants met mastery (or failure for Brett) in the 
exact same number of sessions. Only 1 of the 9 comparisons (11%) resulted in faster skill 
acquisition with general praise (Tina), and again, the difference was small. 
 The bottom panel of Figure 6 depicts percentage correct scores during follow-up 
evaluations. Just as it appeared that there was little difference between the effects of descriptive 
and general praise in terms of acquisition efficiency, the follow-up data also showed no clear 
36 
 
difference between the effects of descriptive or general praise on skill maintenance. Of the 7 
comparisons that resulted in mastery (Brett not included), the majority (43%) of comparisons 
resulted in no difference in follow-up performance. Of the remaining comparisons, 2 out of 7 
(28.5%) resulted in higher performance at follow-up with descriptive praise and the other 2 out 
of 7 (28.5%) resulted in greater maintenance of skills at follow-up with general praise. These 
data support the conclusion that there was no reliable benefit of one type of praise over the other 
in terms of skill maintenance.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
37 
 
 
 
 
 
 
CHAPTER IV  
Discussion 
 The results of the present study contribute to the contemporary line of research on the use 
of descriptive praise when teaching skills to young children with ASDs. The use of descriptive 
praise resulted in no reliable benefit over general praise (both within and across participants) and 
when differences were detected they were minimal. More specifically, skills taught with 
descriptive praise were not acquired more quickly nor did they maintain better at follow-up 
compared to skills taught with general praise.  Additionally, there were no clear added benefits 
of descriptive praise when comparisons examined praise alone or when praise was combined 
with highly preferred items. These findings appear to have been predicted by the results of the 
reinforcer (praise) assessments that showed no differential effect of the type of praise on 
responding. 
 The findings of the present study are consistent with recent empirical reports. For 
example, Stevens et al. (2010) evaluated the effects of descriptive and general praise in 
combination with contingent token delivery on the acquisition of tacts of two children ASDs and 
found no between-condition differences. Similarly, Sellers and Higbee (2009) found no 
difference in the rate of skill acquisition of receptive actions (i.e., auditory discriminations) 
across descriptive and general praise conditions with three children with ASDs. Of the 3 
children, 2 exhibited very little learning with either type of praise (as was also seen with Brett in 
the current study). Additionally, in the Sellers and Higbee investigation, the children did not 
exhibit faster acquisition when praise was used alone 
38 
 
or when praise was paired with an edible item, providing further evidence (along with Stevens et 
al. and the present investigation) that pairing praise with additional reinforcers may not increase 
the effectiveness of the type of praise used. Finally, the current study investigated praise with 
more skilled learners (i.e., with strong listener repertoires) than those in the Stephens et al. and 
Sellers and Higbee studies, and still no differential benefit of descriptive praise was found. Thus, 
the utility of descriptive praise may not be specifically related to receptive language skills, at 
least for those children on the autism spectrum. 
 From the present findings, it appears that the numerous recommendations favoring 
descriptive over general praise commonly found within the scientific (e.g., Brophy, 1981) and 
practice literatures (e.g., Leaf & McEachin, 1999; Maurice, Green, & Luce, 1996) have not yet 
been supported. It is possible, however, that praise might be an integral educational component 
as all children in the current study learned to some extent with the implementation of the 
prompting-plus-praise. On the other hand, praise appeared to have idiosyncratic effects on skill 
acquisition. Two of the 4 participants required the addition of highly preferred items with praise 
to increase acquisition. Moreover, one participant (Brad) was able to acquire skills without praise 
(a neutral statement alone was sufficient). 
 Since it does appear that the utility of praise (descriptive or general) could be highly 
individualized, investigating predictors of such outcomes would be useful to increase the 
efficiency of skill acquisition program development and implementation. In the current study, we 
conducted reinforcer (praise) assessments prior to intervention to determine if they would predict 
the results of the praise evaluations. The results of these assessments showed no differentiation 
of responding in either descriptive or general praise conditions, suggesting that the assessments 
did serve as a predictor for the utility of the type of praise (that there would not be a difference). 
39 
 
However, it is possible that the participants? undifferentiated responding was a result of 
compliance to the initial instruction to ?Do this? from their instructional history. In addition, it is 
also possible that the tasks that were chosen for the assessments were automatically reinforcing. 
This does not appear to be a strong possibility, because different tasks were probed prior to the 
assessments and neutral materials were used (e.g., beige blocks, white stickers).  Thus, to the 
extent that responding in the reinforcer assessment was influenced by instructional control or the 
automatically reinforcing properties of the preparation, the results of the assessments would not 
be predictive of subsequent learning in the treatment evaluations. 
 Another interesting finding from this study was Brad?s skill acquisition when only a 
neutral statement (e.g., ?let?s move on?) was provided as a consequence for responding, 
suggesting that learning can be produced by simply pairing prompting with differential 
consequences. This particular evaluation is further evidence that young children with ASDs can 
learn without the explicit use of praise; however, it may take longer to acquire skills under such 
conditions (as was found with Brad). This finding is particularly interesting because research has 
suggested that the use of praise by teachers can be infrequent in general education classrooms 
(ranging 0.02 to 4.4 per hour, Brophy, 1981; Shores, Gunter, Ellis, DeBriere, & Wehby, 1993; 
Wehby, Symons, & Shores, 1995). The paucity of praise has lead to efforts to increase teachers? 
use of praise (e.g., Alber et al., 1999; Craft et al., 1998). Given the relative lack of praise that is 
possible in general education classrooms, the finding that a child diagnosed with autism (Brad) 
was able to acquire skills without praise suggests that he may have been responding more like a 
?typical? child, since typical children can still succeed in general education classrooms 
regardless of the density of praise (as evidenced by skill acquisition of children in general 
40 
 
education classrooms that include little praise). Moreover, it is possible that this finding is a 
direct benefit of Brad?s 2 years experience with EIBI prior to the study. 
 Further research is needed to evaluate the necessity of praise in teaching children with 
ASDs. Perhaps very early learners require dense praise which can be thinned out over time.  
However, research suggests that praise can be difficult to establish as a conditioned reinforcer. 
Dozier (2006) evaluated strategies for establishing praise as a reinforcer for adults with 
developmental disabilities (e.g., intellectual disabilities) whose behavior was not initially 
sensitive to praise as a reinforcer (a common finding also in children with ASDs; Lovaas, 2003). 
Three different approaches to conditioning praise were evaluated (all of which paired previously 
neutral statements with preferred edibles) and the results of the experiments showed that the 
effects of praise as a conditioned reinforcer were inconsistent and idiosyncratic at best. Thus, 
praise alone may not be an effective reinforcer for individuals with developmental disabilities, 
even after multiple pairings with known reinforcers. Additionally, there may be other stimulus 
factors critical to the efficacy of praise as a reinforcer, such as facial expressions, tone of voice, 
enthusiasm, etc. Further research could isolate the particular stimulus components of praise to 
determine if the verbal statement is integral or if a large smile, widened eyes, or excited tone of 
voice are more relevant for establishing praise as a reinforcer. 
 Given the ubiquitous recommendations for descriptive praise in EIBI (e.g., Anderson et 
al., 1996; Leaf & McEachin, 1999), the question remains whether the recommendations should 
be tempered or discontinued in light of the emerging empirical evidence on the topic. 
Additionally, it may be more beneficial to focus on what current evidence suggests as the most 
effective for increasing responding in skill acquisition programming versus focusing efforts on 
the use of praise. Until further support is provided for the use of praise, practitioners may 
41 
 
consider pairing either type of praise (general or descriptive) with highly preferred items 
contingent upon correct responding in teaching language skills to children with ASDs. There is a 
large body of literature on the use of stimulus preference assessment technology to identify 
preferred items that consequently serve as reinforcers (e.g., Carr et al., 2000; Fisher et al., 1992). 
Thus, while there is lacking support for one type of praise over the other, practitioners should 
still see benefits in skill acquisition by focusing their clinical efforts on pairing praise with 
known preferred items when teaching language and language-related skills.  Of course, the 
recommendation to use praise in conjunction with highly preferred tangible items begs the 
question, ?Is praise necessary?? 
 In the current investigation, it is possible that descriptive praise could have produced 
additional benefits over general praise, but that the effects were simply undetected. The current 
study taught individual responses (i.e., intraverbals) in a direct instructional context where 
correct answers were prompted. It is possible that descriptive praise may be particularly 
beneficial when teaching skills in indirect teaching formats, such as teaching multiple responses 
in a more complicated and social context. For example, descriptive praise may have an 
advantage over general praise when teaching a social skill such as appropriate peer play. In such 
training, a practitioner could contrive a teaching scenario with a peer and observe the child, then 
contingent upon correct peer play provide descriptive praise immediately. In such a situation the 
correct response may not be salient to the child, as the appropriate skill is part of a more complex 
social environment. It is possible that the added verbal feedback in descriptive praise may serve 
additional benefits to the learner by allowing them to tact the correct response they had engaged 
(as per the praise provided). Future research could evaluate the use of descriptive versus general 
praise in teaching social skills. 
42 
 
 Another potential benefit of descriptive praise might be in the area of emergent behavior. 
For example, it may be possible that teaching a skill such as motor imitation with descriptive 
praise could result in the emergence of a corresponding tact. To teach a motor imitative skill, the 
practitioner would provide the instruction ?Do this? and then model the correct response, such as 
touching her nose. Upon correct imitation, if the clinician uses descriptive praise, she would say 
?Great job touching your nose!? Upon acquisition of this skill with descriptive praise, it is 
theoretically possible that the child might also acquire the tact for the item or body part involved 
in the motor imitation skill.  This could feasibly occur if the child echoed (overtly or covertly) 
the therapist?s praise statement (?nose?) while looking at her nose.  Similar emergent relations 
might be evaluated with other programs that include a nonverbal stimulus and the therapist?s tact 
of the stimulus during descriptive praise (e.g., matching, receptive identification).  Future 
research on such emergent relations would be necessary to justify the use of descriptive praise 
for such purposes.  
 It is also important to consider other environmental conditions under which descriptive 
praise could potentially be more useful. Although direct evaluations of descriptive versus general 
praise are few, studies have showed some utility of using descriptive praise in a number of 
educational contexts (Chalk & Bizo, 2004; Novak & Hammond, 2001; Sutherland et al., 2001), 
mostly with typically developing children (with and without emotional or behavioral disorders). 
It is possible that a stronger verbal repertoire is necessary to reap the benefits of descriptive 
praise, and it may be that individuals must be able to verbalize rules either privately or publicly 
in order to gain benefits from descriptive praise. For example, telling a child ?Thank you for 
putting your toys away? may not be a meaningful statement to a very young learner with ASD, 
but with a greater history of reinforcement for rule following, a child with stronger verbal 
43 
 
abilities may acquire a rule ?If I put my toys away I will get attention.? While the child?s 
?putting toys away? behavior may or may not be continuously reinforced by the mother, the 
verbal rule statement could eventually be paired with public reinforcement to exert stimulus 
control over the behavior in question (putting toys away). If a child does not have the ability to 
make such rules, the descriptive nature of praise may not serve as a meaningful stimulus.  
The results of the present study should be evaluated in the context of several limitations. 
First, the results of the reinforcer (praise) assessments for all participants produced no change in 
responding across descriptive, general, and no-praise conditions. It is possible that the tasks used 
in the study were inherently reinforcing despite efforts to select activities that were neutral to the 
children (i.e., through the use of plain beige-colored blocks and white circle stickers). Another 
explanation for this finding is that the initial prompt and model provided at the beginning of the 
session (i.e., saying ?Do this? and modeling the response) exerted stimulus control over 
consistent responding. Thus, perhaps the participants? responding was under the control of the 
antecedent as opposed to the scheduled consequences. These possibilities do not appear to 
undermine the results of the study, although they do question the utility of the praise evaluation?s 
contribution to the final project. Future research could further investigate the utility of praise 
assessments as predictors of success in praise-based interventions. 
Second, it is possible that the skills targeted for the study were not as equated for 
difficulty as initially predicted. While each skill set included single-syllable words, in the third 
comparison conducted with Brad (prompting plus neutral statement) he mastered set 5 more 
quickly than set 6 (in 14 versus 23 sessions). In this comparison, there were not two separate 
interventions being compared, and instead the same intervention was implemented across sets. If 
the two were equally difficult he should have mastered them in the same number of sessions. In 
44 
 
the 3-word set 6, he emitted more incorrect answers to the question ?What animal purrs (cat)?? 
compared to ?What animal swims (fish)?? and ?What animal has a curly tail (pig)?? Instead of 
answering ?Cat,? he would answer ?Cat-ten? (possibly a combination of cat and kitten) and then 
answer ?Cat? during the error-correction procedure. Although the data suggest that it was 
possible that the sets were not equated for difficulty, this is likely not an issue regarding the other 
sets because the other participants? trends of skill acquisition were often very similar (or 
identical). Additionally, when Brad would incorrectly answer ?Cat-ten? he often did so while 
giggling, which suggests additional sources of reinforcement for the response. 
Third, Tina?s variability in responding in both comparisons could have been due to 
problems in articulation of the words in the sets. She would sometimes engage in response chains 
by repeating back the entire intraverbal phrase, resulting in variable incorrect responding. Thus, 
it is possible that some words were more difficult phonetically. For example, it took her longer to 
master set 1, during which she provided incorrect answers for only the phrase ?You cook soup in 
a___ (pot)?. She would commonly answer ?Po? and not the full phrase ?Pot? suggesting the ?-t? 
sound may have been more difficult to produce. However the most-to-least prompting strategy 
eradicated the acquisition problem and comparisons were still able to be made across the types of 
praise (although in the context of additional intervention components). 
A fourth limitation to the study was the final failure to teach math intraverbals to Brett. It 
may have been that the skill was too difficult and that pre-requisite mathematic skills (addition) 
were necessary to be able to complete the intraverbal phrases. For example, a child may need to 
master addition through other methods prior to teaching the phrase ?1+1= __? (e.g., show one 
penny, then show another penny, and put them together and ask the child to count how many 
there are in total). However, it could also be possible that when teaching math skills, pairing 
45 
 
picture prompts (e.g., showing a flashcard with the numbers 2+3=___) with the introduction of 
the intraverbal phrases may be facilitative. Although he did not acquire the skills in the study, his 
performance is particularly interesting because it further demonstrated the nondifferential effects 
of descriptive praise on performance. 
In the current investigation it appears that recommendations for using descriptive praise 
in teaching language to children with ASDs exceed the evidence to support them. In the 9 
comparisons conducted in this study, descriptive praise did not produce any reliable beneficial 
effects across and within participants. Thus, it cannot be confidently concluded that descriptive 
praise provides a greater benefit than general praise for skill acquisition in children with ASDs, 
but rather the effectiveness of descriptive praise (and praise, in general) as a teaching strategy 
seems to be idiosyncratic. The present study suggests the need to proceed with caution when 
considering behavioral strategies in EIBI, even if the strategy is commonly recommended. 
Furthermore, practitioners should seek evidence in support of EIBI practices they are using or 
considering. This is imperative in the field of ABA where behavior analysts are ethically bound 
to provide effective, evidence-based behavioral services and to make data-based decisions 
(Behavior Analyst Certification Board, 2010).  
 
 
 
 
 
 
 
46 
 
 
 
 
 
 
References 
Alber, S. R., Heward, W. L., & Hippler, B. J. (1999). Teaching middle school children with 
learning disabilities to recruit positive teacher attention. Exceptional Children, 65, 253-
270. 
American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders 
(4th ed.) Text revision. Washington, DC: Author. 
Anderson, S. R., Taras, M., & Cannon, B. O. (1996). Teaching new skills to young children with 
autism. In C. Maurice, G. Green, & S. C. Luce (Eds.). Behavioral intervention for young 
children with autism: A manual for parents and professionals (pp. 181-194). Austin, TX: 
PRO-ED. 
Behavior Analysis Association of Michigan (n. d.). BAAM behavioral essentials: Behavior 
analysis Q & A. Retrieved January 25, 2011, from: 
http://www.baam.emich.edu/BAAMMainpages/BAAMfaq.htm 
Behavior Analyst Certification Board (2010). Behavior analyst certification board guidelines for 
responsible conduct. Retrieved February 7, 2011 from: 
http://www.bacb.com/index.php?page=57 
Broden, M., Bruce, C., Mitchell, M. A., Carter, V., & Hall, R. V. (1970). Effects of teacher 
attention of attending behavior of two boys at adjacent desks. Journal of Applied 
Behavior Analysis, 3, 199-203. 
Brophy, J. (1981). Teacher praise: A functional analysis. Review of Educational Research, 51, 5-
32.
47 
 
Brown, J. L., Krantz, P. J., McClannahan, L. E., & Poulson, C. L. (2008). Using script fading to 
promote natural environment stimulus control of verbal interactions among youths with 
autism. Research in Autism Spectrum Disorders, 2, 480-497. 
Carr, J. E., Nicolson, A. C., & Higbee, T. S. (2000). Evaluation of a brief multiple-stimulus 
preference assessment in a naturalistic context. Journal of Applied Behavior Analysis, 33, 
353-357. 
Centers for Disease Control and Prevention (2004, October). Developmental disabilities. 
Retrieved January 25, 2011, from http://www.cdc.gov/ncbddd/dd/dd1.htm  
Centers for Disease Control and Prevention. (2006). Parental report of diagnosed autism in 
children aged 4-17 years-United States, 2003-2004. Morbidity and Mortality Weekly 
Report, 55, 481-486. 
Centers for Disease Control and Prevention. (2007). Prevalence of autism spectrum disorders-
autism, developmental disabilities monitoring (ADDM) network, 14 sites, United States, 
2002. Morbidity and Mortality Weekly Report, 56, 12-28. 
Chalk, K., & Bizo, L. A. (2004). Specific praise improves on-task behaviour and numeracy 
enjoyment: A study of year four pupils engaged in the numeracy hour. Educational 
Psychology in Practice, 20, 335-351. 
Craft, M. A., Alber, S. R., & Heward, W. L. (1998). Teaching elementary students with 
developmental disabilities to recruit teacher attention in a general education classroom: 
Effects on teacher praise and academic productivity. Journal of Applied Behavior 
Analysis, 31, 399-415. 
48 
 
Dozier, C. L. (2006). An evaluation of methods to establish praise as a reinforcer. Unpublished 
doctoral dissertation, University of Florida, Gainesville. 
Eikeseth, S. (2009). Outcome of comprehensive psycho-educational interventions for young 
children with autism. Research in Developmental Disabilities, 30, 158-178. 
Eikeseth, S., Smith, T., Jahr, E., & Eldevik, S. (2002). Intensive behavioral treatment at school 
for 4-7 year-old children with autism: A 1-year comparison controlled study. Behavior 
Modification, 26, 49-68. 
Eikeseth, S., Smith, T., Jahr, E., & Eldevik, S. (2007). Outcome for children with autism who 
began intensive behavioral treatment between age four and seven: A comparison 
controlled study. Behavior Modification, 31, 264-278 
Eldevik, S., Eikeseth, S., Jahr, E., & Smith, T. (2006). The effects of low-intensive behavioral 
treatment of children with autism and mental retardation. Journal of Autism and 
Developmental Disorders, 36, 211-224. 
Eldevik, S., Hastings, R. P., Hughes, J. C., Jahr, E., Eikeseth, S., & Cross, S. (2009). Meta-
analysis of early intensive behavioral intervention for children with autism. Journal of 
Clinical Child & Adolescent Psychiatry, 38, 439-450. 
Faja, S., & Dawson, G. (2006). Early intervention for autism. In J. L. Luby (Ed.), Handbook of 
preschool mental health: Development, disorders, and treatment (pp. 388-416). New 
York: Guilford Press. 
Fenske, E., Zalenski, S., Krantz, P., & McClannahan, L. (1985). Age at intervention and 
treatment outcome for autistic children in a comprehensive intervention program. 
Analysis and Intervention for Developmental Disabilities, 5, 49-58. 
49 
 
Ferguson, E., & Houghton, S., (1992). The effects of contingent teacher praise, as specified by 
Canter?s Assertive Discipline Programme, on children?s on-task behaviour. Educational 
Studies, 18, 83-93. 
Ferster, C. B., & DeMyer, M. K. (1961). The development of performances in autistic children in 
an automatically controlled environment. Journal of Chronic Diseases, 13, 312-345. 
Fisher, W., Piazza, C. C., Bowman, L. G., Hagopian, L. P., Owens, J. C., & Slevin, I. (1992). A 
comparison of two approaches for identifying reinforcers for persons with severe and 
profound disabilities. Journal of Applied Behavior Analysis, 25, 491-498. 
Foxx, R. M., McMorrow, M. J., & Schloss, C. N. (1983). Stacking the deck: Teaching social 
skills to retarded adults with a modified table game. Journal of Applied Behavior 
Analysis, 16, 157-170. 
Good, C. E., Eller, B. F., Spangler, R. S., & Stone, J. E. (1981). The effect of an operant 
intervention program on attending and other academic behavior with emotionally 
disturbed children. Journal of Instructional Psychology, 9, 25-33. 
Green, G., Brennan, L. C., & Fein, D. (2002). Intensive behavioral treatment for a toddler at high 
risk for autism. Behavior Modification, 26, 69-102. 
Hall, R. V., Lund, D., & Jackson, D. (1968). Effects of teacher attention on study behavior. 
Journal of Applied Behavior Analysis, 1, 1-12. 
Hanson, E., Kalish, L. A., Bunce, E., Curtis, C., McDaniel, S., Ware, J., et al. (2007). Use of 
complementary and alternative medicine among children diagnosed with autism spectrum 
disorder. Journal of Autism and Developmental Disorders, 37, 628-636. 
50 
 
Harris, S., & Handleman, J. (2000). Age and IQ at intake as predictors of placement for young 
children with autism: A four- to six-year follow-up. Journal of Autism and 
Developmental Disorders, 30, 137-142. 
Harris, S., Handleman, J., Gordon, R., Kristoff, B., & Fuentes, F. (1991). Changes in cognitive 
and language functioning of preschool children with autism. Journal of Autism and 
Developmental Disorders, 21, 281-290. 
Heflin, L. J., & Alberto, P. A. (2001). Establishing a behavioral context for learning for students 
with autism. Focus on Autism and Other Developmental Disabilities, 16, 93-101. 
Keller, C. L., Brady, M. P., & Taylor, R. L. (2005). Using self evaluation to improve student 
teacher interns? use of specific praise. Education and Training in Developmental 
Disabilities, 40, 368-376. 
Leaf, R., & McEachin, J. (Eds.) (1999). A work in progress: Behavior management strategies 
and a curriculum for intensive behavioral treatment of autism. New York: DRL Books. 
Lovaas, O. I. (1987). Behavioral treatment and normal educational and intellectual functioning in 
young autistic children. Journal of Consulting and Clinical Psychology, 55, 3-9. 
Lovaas (2003). Teaching individuals with developmental delays: Basic intervention techniques. 
Austin, TX: Pro-Ed.  
Lovaas, O. I., Berberich, J. P., Perloff, B. F., & Schaeffer, B. (1966). Acquisition of imitative 
speech by schizophrenic children. Science, 151, 705-707. 
Love, J. R., Carr, J. E., Almason, S. M., & Petursdottir, A. I. (2009). Early and intensive 
behavioral intervention for autism: A survey of clinical practices. Research in Autism 
Spectrum Disorders, 3, 421-428. 
51 
 
Maurice, C., Green, G., & Luce, S. C. (1996). Preface. In C. Maurice, G. Green, & S. C. Luce 
(Eds.). Behavioral intervention for young children with autism: A manual for parents and 
professionals (pp. xiii-xiv). Austin, TX: PRO-ED. 
McClannahan, L. E., & Krantz, P. J. (2004). Some guidelines for selecting behavioral 
intervention programs for children with autism. In H. E. Briggs & T. L. Rzepnicki (Eds.), 
Using evidence in social work practice: Behavioral perspectives (pp. 92-103). Chicago: 
Lyceum Books. 
McGee, G. G., Krantz, P. J., Mason, D., & McClannahan, L. E. (1983). A modified incidental 
teaching procedure for autistic youth: Acquisition and generalization of receptive object 
labels. Journal of Applied Behavior Analysis, 16, 329-338. 
Nebel-Schwalm, M. S., & Matson, J. L. (2008). Differential diagnosis. In J. Matson (Ed.), 
Clinical assessment and intervention for autism spectrum disorders (pp. 91-129). 
Burlington, MA: Academic Press.  
Novak, G., & Hammond, J. M. (2001). Self-reinforcement and descriptive praise in maintaining 
token economy reading performance. Journal of Educational Research, 76, 186-189. 
Odom, S. L., Horner, R. H., Snell, M. E., & Blacher, J. (2007). The construct of developmental 
disabilities. In S. L. Odom, R. H. Horner, M. E. Snell, & J. Blacher (Eds.), Handbook of 
developmental disabilities (pp. 3-15). New York: Guilford Press.   
Partington, J. W. (2006). The assessment of basic language and learning skills- Revised. Pleasant 
Hill, CA: Behavior Analysts.  
Phillips, R. H. (1984). Increasing positive self-referent statements to improve self-esteem in low-
income elementary school children. Journal of School Psychology, 22, 155-163. 
52 
 
Reinke, W. M., Lewis-Palmer, T., & Martin, E. (2007). The effect of visual feedback 
performance feedback on teacher use of behavior-specific praise. Behavior Modification, 
31, 247-263. 
Remington, B., Hastings, R. P., Kovshoff, H., degli Espinosa, F., Jahr, W., Brown, T., et al. 
(2007). A field effectiveness study of early intensive behavioral intervention: Outcomes 
for children with autism and their parents after two years. American Journal on Mental 
Retardation, 112, 418-438. 
Rigsby-Elderidge, M., & McLaughlin, T. F. (1992). The effects of modeling and praise on self-
initiated behavior across settings with two adolescent students with autism. Journal of 
Developmental and Physical Disabilities, 4, 205-218. 
Romanczyk, R. G., Gillis, J. M., White, S., & Digennaro, F. (2008). Comprehensive treatment 
packages for ASD: Perceived vs. proven effectiveness. In J. Matson (Ed.). Clinical 
assessment and intervention for autism spectrum disorders (pp. 351-381). New York: 
Academic Press. 
Rosenwasser, B., & Axelrod, S. (2001). The contributions of applied behavior analysis to the 
education of people with autism. Behavior Modification, 25, 671-677. 
Scheer, R. R. (1977). The relative effects of general versus descriptive praise on a card sorting 
task (Doctoral dissertation, University of Maryland, 1977). Dissertation Abstracts 
International, 37, 7049. 
Schreibman, L. (2000). Intensive behavioral/psychoeducational treatments for autism: Research 
needs and future directions. Journal of Autism and Developmental Disorders, 30, 373-
378.  
53 
 
Schreibman, L., & Anderson, A. (2001). Focus on integration: The future of the behavioral 
treatment of autism. Behavior Therapy, 32, 619-632. 
Sellers, T. P., & Higbee, T. S. (2009). The effects of general praise statements versus behavior 
specific praise statements on skill acquisition in young children with autism. Poster 
presented at the 35th Annual Convention of the Association for Behavior Analysis 
International. Phoenix, AZ.  
Sherer, M. R., & Schreibman, L. (2005). Individual behavioral profiles and predictors of 
treatment effectiveness for children with autism. Journal of Consulting and Clinical 
Psychology, 73, 525-538. 
Shores, R. E., Jack, S. L., Gunter, P. L., Ellis, D. N., DeBriere, T. J., & Wehby, J. H. (1999). 
Classroom interactions of children with behavior disorders. Journal of Emotional and 
Behavioral Disorders, 1, 27-39. 
Shute, V. J. (2008). Focus on formative feedback. Review of Educational Research, 78, 153-189. 
Sigafoos, J., Doss, S., & Reichle, J. (1989). Developing mand and tact repertoires in persons with 
severe developmental disabilities using graphic symbols. Research in Developmental 
Disabilities, 10, 183-200. 
Simonsen, B., Fairbanks, S., Briesch, A., Myers, D., & Sugai, G. (2008). Evidence-based 
practices in classroom management: Considerations from research to practice. Education 
and Treatment of Children, 21, 351-380. 
Sindelar, P. T., Rosenberg, M. S., & Wilson, R. J. (1985). An adapted alternating treatment 
design for instructional research. Education & Treatment of Children, 8, 67-76. 
Skinner, B. F. (1957). Verbal behavior. Acton, MA: Copley Publishing. 
54 
 
Stevens, C., Sidener, T. M., Reeve, S. A., & Sidener, D. W. (2011). Effects of behavior-specific 
and general praise on acquisition of tacts in children with pervasive developmental 
disorders. Research in Autism Spectrum Disorders, 5, 666-669. 
Sundberg, M. L. (2008). VB-MAPP: Verbal behavior milestones assessment and placement 
program. Concord, CA: AVB Press. 
Sundberg, M. L., & Partington, J. W. (1998). Teaching language to children with autism or other 
developmental disorders. Pleasant Hill, CA: Behavior Analysts. 
Sutherland, K. S., Wehby, J. H., & Copeland, S. R. (2000). Effect of varying rates of behavior-
specific praise on the on-task behavior of students with EBD. Journal of Emotional and 
Behavioral Disorders, 8, 2-8. 
Steege, M. W., Mace, F. C., Perry, L., & Longenecker, H. (2007). Applied behavior analysis: 
Beyond discrete trial teaching. Psychology in the Schools, 44, 91-99. 
Wehby, J. H., Symons, F. J., & Shores, R. E. (1995). A descriptive analysis of aggressive 
behavior in classrooms for children with emotional and behavioral disorders. Behavioral 
Disorders, 20, 87-105. 
Wolf, M. M., Risley, T. R., & Mees, H. L. (1964). Application of operant conditioning 
procedures to the behavior problems of an autistic child. Behaviour Research and 
Therapy, 1, 305-312. 
Wright, R. A. (2008). An examination of the good behavior game and behavior specific praise 
statements on student and teacher behavior. Unpublished doctoral dissertation, Louisiana 
State University, Baton Rouge. 
55 
 
Zahler, N. S. (1975). The effects of descriptive vs. general praise on a simple monotonous motor 
task (Doctoral dissertation, University of Maryland, 1975). Dissertation Abstracts 
International, 36, 3533. 
 
56 
 
Footnote 
 
 1 From this point forward, the term descriptive praise will encompass other references to 
the same strategy, such as behavior-specific praise, approval with description, behavior-
descriptive praise, and process praise. 
 
 
57 
 
Table 1 
Intraverbal Skill Targets used for Instruction 
 Participants 
 Brad Shaun Tina Brett 
Intraverbal 
Skill Area  
Answering 
?What?? 
questions 
Answering 
?What?? 
questions 
Answering ?Fill-in? 
phrases 
Answering 
?Fill-in? 
phrases 
Set 1  
 
What do you: 
Eat with? 
Chew with? 
Kick with? 
What animal: 
Barks? 
Has feathers? 
Has whiskers? 
You cook soup in a _ 
You sleep in a _ 
You drink out of a _ 
2+ 1 equals _ 
3+ 2 equals _ 
4+5 equals _ 
 
 
Set 2  
 
 
What do you: 
Clap with? 
Blink with? 
Taste with? 
 
What animal: 
Is pink? 
Do you get milk 
from? 
Eats hay? 
 
You cut food with a _ 
You take a bath in a _ 
You sit on a _ 
 
3+1 equals _ 
4+2 equals _ 
5+3 equals _ 
 
 
Set 3  
 
What do you: 
Cook soup in? 
Take a bath in? 
Sweep the floor 
with? 
 
What do you: 
See with? 
Taste with? 
Walk with? 
 
You see with your _ 
You smell with your _ 
You clap with your _ 
 
9+4 equals _ 
7+8 equals_ 
10+9 equals_ 
 
 
Set 4  
 
What do you: 
Eat ice cream 
with? 
Cut food with? 
Eat food off of? 
 
What do you: 
Eat with? 
Smile with? 
Chew with? 
 
You kick with your _ 
You eat with your _ 
You hear with your _ 
 
6+8 equals _ 
10+6 equals _ 
9+9 equals _ 
 
 
Set 5  
 
What animal: 
Has feathers? 
Do you get milk 
from? 
Barks? 
 
What do you: 
Sweep the floor 
with? 
Eat food off of? 
Take a bath in? 
 
Animal that barks is a _ 
Animal that has a curly 
tail is a _ 
Animal that swims is a _ 
 
19+2 equals _ 
18+5 equals _ 
17+8 equals_ 
 
 
Set 6  
 
What animal: 
Swims? 
Purrs? 
Has a curly tail? 
 
What do you: 
Cook soup in? 
Cut food with? 
Eat ice cream 
with? 
 
Animal that has feathers 
is a _ 
Animal that purrs is a _ 
Animal you get milk 
from is a _ 
 
20+2 equals_ 
21+3 equals _ 
22+4 equals_ 
 
58 
 
 
Figure 1. Reinforcer (praise) assessment results for Brad, Shaun, Tina, and Brett. Note: DP = 
descriptive praise, GP= general praise.  
59 
 
 
Figure 2. Percentage correct during baseline, treatment, and follow-up conditions for Brad.  
Note: during the Prompting+Neutral Statement phase no praise was given. 
60 
 
 
Figure 3. Percentage correct during baseline, treatment, and follow-up conditions for Shaun.  
Note: BL = baseline. 
61 
 
 
Figure 4. Percentage correct during baseline, treatment, and follow-up conditions for Tina.  
Note: Errorless = errorless training, HPI = highly preferred item. 
62 
 
  Figure 5. Percentage correct during baseline and treatment conditions for Brett. Note: At session 
149 the target behavior in errorless training trials changed from a single word (e.g., ?Nine?) to 
the full phrase (i.e., ?Four plus five equals nine.?); Errorless = errorless training, HPI = highly 
preferred item. 
63 
 
  
 
 
Figure 6. Number of sessions to meet mastery criteria during interventions consisting of either 
descriptive or general praise (top panel). Percentage correct during follow up data collection 
(bottom panel). *Brett did not meet mastery criteria and instead his data show the total number 
of sessions to meet final failure criteria. 
 
 
64 
 
 
Appendix A 
 
Sample Experimental Visit Data Collection Sheet 
 
Participant: _______________   Date:__________  Visit Time Start: _____ End:_____ 
Directions: For each trial circle if the child made a correct response (+), incorrect response (-), or 
no response (NR). Then circle 1 or 2 according to the prompt level given. 
 
Session # Session # Session # 
Condition: 
 
Condition: 
 
Condition: 
 
Target Set: SET 1 Target Set: SET 2 Target Set: SET 1 
SD:  Response SD : Response SD : Response 
You cook soup in a 
_______( Pot) 
 
+    ?    NR 
         1    2 
You cut food with 
___( Knife) 
 
+    ?    NR 
         1    2 
You sleep in a 
_______(bed)  
 
+    ?    NR 
         1    2 
You sleep in a 
_______(bed)  
 
+    ?    NR 
         1    2 
You take a bath in 
a ____(tub) 
 
+    ?    NR 
         1    2 
You drink out of a 
_______(cup) 
 
+    ?    NR 
         1    2 
You drink out of a 
_______(cup) 
 
+    ?    NR 
         1    2 
You sit on a 
_____(chair) 
 
+    ?    NR 
         1    2 
You cook soup in a 
_______( Pot) 
 
+    ?    NR 
         1    2 
You cook soup in a 
_______( Pot) 
 
+    ?    NR 
         1    2 
You cut food with 
___( Knife) 
 
+    ?    NR 
         1    2 
You sleep in a 
_______(bed)  
 
+    ?    NR 
         1    2 
You sleep in a 
_______(bed)  
 
+    ?    NR 
         1    2 
You take a bath in 
a ____(tub) 
 
+    ?    NR 
         1    2 
You drink out of a 
_______(cup) 
 
+    ?    NR 
         1    2 
You drink out of a 
_______(cup) 
 
+    ?    NR 
         1    2 
You sit on a 
_____(chair) 
 
+    ?    NR 
         1    2 
You cook soup in a 
_______( Pot) 
 
+    ?    NR 
         1    2 
You cook soup in a 
_______( Pot) 
 
+    ?    NR 
         1    2 
You cut food with 
___( Knife) 
 
+    ?    NR 
         1    2 
You sleep in a 
_______(bed)  
 
+    ?    NR 
         1    2 
You sleep in a 
_______(bed)  
 
+    ?    NR 
         1    2 
You take a bath in 
a ____(tub) 
 
+    ?    NR 
         1    2 
You drink out of a 
_______(cup) 
 
+    ?    NR 
         1    2 
You drink out of a 
_______(cup) 
 
+    ?    NR 
         1    2 
You sit on a 
_____(chair) 
 
+    ?    NR 
         1    2 
You cook soup in a 
_______( Pot) 
 
+    ?    NR 
         1    2 
% Correct 
 
 
Session Duration: 
 
% Correct 
 
 
Session Duration: 
 
% Correct 
 
 
Session Duration: 
 
Comments: 
 
 
 
 
65 
 
 
 
Appendix B 
 
Sample IOA Data Sheet 
IOA Data Collection Sheet 
Directions: For each trial after the Sd was given (e.g., What do you eat with?) circle if the child 
made an independent correct response (+), incorrect response (made a sound that was not the full 
correct word) (-), or no response (said absolutely nothing when given the Sd) (NR). Also for 
each trial, circle the prompt that was given by Amy. Circle PP if she gave a partial word prompt 
(she restated the Sd and said part of the word) and circle FW if she restated the Sd and said the 
full word. Both PP and FW may be circled during a single trial, but only ONE of the +, -, or NR 
should be circled. 
 
Participant: 
 
 
Participant: 
 
 
Participant: 
 
 
Participant: 
 
 
Date of video: 
 
Date of video: 
 
Date of video: 
 
Date of video: 
 
Session#: 
 
Condition (circle): 
BSL   DP   GP  NS 
 
Session#: 
 
Condition (circle): 
BSL   DP   GP  NS 
 
Session#: 
 
Condition (circle): 
BSL   DP   GP  NS 
 
Session#: 
 
Condition (circle): 
BSL   DP   GP  NS 
 
Trial Response Trial Response Trial Response Trial Response 
1 +    ?    NR 
PP      FW 
1 +    ?    NR 
PP      FW 
1 +    ?    NR 
PP      FW 
1 +    ?    NR 
PP      FW 
2 +    ?    NR 
PP      FW 
2 +    ?    NR 
PP      FW 
2 +    ?    NR 
PP      FW 
2 +    ?    NR 
PP      FW 
3 +    ?    NR 
PP      FW 
3 +    ?    NR 
PP      FW 
3 +    ?    NR 
PP      FW 
3 +    ?    NR 
PP      FW 
4 +    ?    NR 
PP      FW 
4 +    ?    NR 
PP      FW 
4 +    ?    NR 
PP      FW 
4 +    ?    NR 
PP      FW 
5 +    ?    NR 
PP      FW 
5 +    ?    NR 
PP      FW 
5 +    ?    NR 
PP      FW 
5 +    ?    NR 
PP      FW 
6 +    ?    NR 
PP      FW 
6 +    ?    NR 
PP      FW 
6 +    ?    NR 
PP      FW 
6 +    ?    NR 
PP      FW 
7 +    ?    NR 
PP      FW 
7 +    ?    NR 
PP      FW 
7 +    ?    NR 
PP      FW 
7 +    ?    NR 
PP      FW 
8 +    ?    NR 
PP      FW 
8 +    ?    NR 
PP      FW 
8 +    ?    NR 
PP      FW 
8 +    ?    NR 
PP      FW 
9 +    ?    NR 
PP      FW 
9 +    ?    NR 
PP      FW 
9 +    ?    NR 
PP      FW 
9 +    ?    NR 
PP      FW 
% 
Correct 
 % 
Correct 
 % 
Correct 
 % 
Correct 
 
 
COMMENTS: 
66 
 
 
 
Appendix C 
 
MSWO Data Sheet 
Participant #:         Date:     
Observer:      
Directions: 
1.  Write the names of the 8 items from the RAISD in the gray boxes. 
2.  Arrange the items in a semi-circle so that all of the items are equi-distant. 
3.  Instruct child to ?Pick one.? 
4.  Block any attempts to select more than one item. 
5.  Once item selected, allow access for 30 s. Fill in the order that the items are selected (e.g., 
write a 1 in the box for the first item selected) 
6.  After the 30 seconds are up, remove the item. 
7.  Rearrange the array and repeat steps until all items have been selected or the child refuses to 
select an item. 
8.  Do three full MSWO arrays, then calculate preference percentage. 
   
ITEMS  
 
 
 
       
1         
2         
3         
TOTAL         
Percentage         
Overall 
Rank 
        
 
Preference Percentage = (Number of times selected / Number in the ?Total? box) X 100%  
 
COMMENTS: 
 
 
67 
 
 
Appendix D 
 
Sample Reinforcer (Praise) Assessment Data Sheet 
 
Participant: ______________________  Date: _____________________ 
 
 Session#_____Condition ________  Session# ______Condition _____ 
Session Begin/End: Session Begin/End: 
TIME Response 
Tally 
Crit Crit 
Met? 
TIME Response 
Tally 
Crit Crit Met? 
 Say ? Do This? And Model Bx  Say ? Do This? And Model Bx 
  FR-1 Y  N   FR-1 Y  N 
  1 Y  N   1 Y  N 
  2 Y  N   2 Y  N 
  3 Y  N   3 Y  N 
  4 Y  N   4 Y  N 
  5 Y  N   5 Y  N 
  6 Y  N   6 Y  N 
  7 Y  N   7 Y  N 
  8 Y  N   8 Y  N 
  9 Y  N   9 Y  N 
  10 Y  N   10 Y  N 
  11 Y  N   11 Y  N 
  12 Y  N   12 Y  N 
  13 Y  N   13 Y  N 
  14 Y  N   14 Y  N 
  15 Y  N   15 Y  N 
  16 Y  N   16 Y  N 
  17 Y  N   17 Y  N 
  18 Y  N   18 Y  N 
  19 Y  N   19 Y  N 
  20 Y  N   20 Y  N 
 Break Point:    Break Point:  
*RUN UNTIL 5 MINUTES 
* RUN UNTIL NO RESPONSE FOR 2 MINUTES 
*# ATTEMPTS TO LEAVE________________ #ATTEMPTS TO 
LEAVE_________________ #ATTEMPTS TO LEAVE___________ 
*ON 4TH ATTEMPT, END SESSION 
 
 
 
 
68 
 
 
Appendix E 
 
Intraverbal Fill ins: TINA    Date: ______________________________ 
Directions: Play with Tina and while playing pause and state the following phrases. Record + if she gives 
correct answer, - if she says something else (and write what she answers) and NR if she doesn?t answer. 
 If she answers (+) on the first trial, no need to ask that phrase again. 
 If she is incorrect or gives no answer, do not prompt the correct answer, give neutral statement 
Household Items Functions: 
Fill in phrase: Trial 1 Trial 2 Trial 3 Comments 
You clean the floor with a ______ (mop) +  -  NR +  -  NR +  -  NR  
You sleep in a ___________ (bed) +  -  NR +  -  NR +  -  NR  
You drink out of a ___________ (cup) +  -  NR +  -  NR +  -  NR  
You sit on a __________ (chair) +  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
 
You cook soup in a ___________ (pot) +  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
 
You take a bath in a ___________ (tub) +  -  NR +  -  NR +  -  NR  
You cut food with a _________ (knife) +  -  NR +  -  NR +  -  NR  
You eat ice cream with a_______ (spoon) +  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
 
You eat food off of a __________ (plate) +  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
 
 +  -  NR +  -  NR +  -  NR  
Body part functions: 
Fill in phrase: Trial 1 Trial 2 Trial 3 Comments 
You see with your ______________ +  -  NR +  -  NR +  -  NR  
You smell with your ____________ +  -  NR +  -  NR +  -  NR  
You hear with your ___________ +  -  NR +  -  NR +  -  NR  
You chew with your _____________ +  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
 
You clap with your _____________ +  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
+  -  NR 
 
You taste with your ____________ +  -  NR +  -  NR +  -  NR  
You kick with your ___________ +  -  NR +  -  NR +  -  NR  
 
 
69 
 
Appendix F 
1-Array MSWO Data Sheet 
Participant:         Date:     
Observer:      
Directions: 
1.  Write the names of the 8 potentially preferred items in the gray boxes 
2.  Arrange the items in a semi-circle so that all of the items are equi-distant. 
3.  Instruct child to ?Pick one.? 
4.  Block any attempts to select more than one item. 
5.  Once item selected, allow access for 30 s. Fill in the order that the items are selected (e.g., 
write a 1 in the box for the first item selected) 
6.  After the 30 seconds are up, remove the item. 
7.  Rearrange the array and repeat steps until all items have been selected or the child refuses to 
select an item. 
8.  List the top 3 items 
 
ITEMS  
 
 
 
       
1         
 
Top 3 preferred items (LIST): 
 
 
 
 
 
 
 
 
  
70 
 
Appendix G 
 
Sample Procedural Integrity Data Collection Sheet 
 
Participant Name: ____________    
Date/Session #_______________________  Observer 1: ____________________ 
Experimenter: __________________  Observer 2 (IOA):  ______________ 
Directions: For each videotaped session, please observe the experimenter and record ?yes? or 
?no? for each trial that the experimenter runs. There should be nine trials per session. 
Experimental Condition (circle) 
(Consequences of condition) 
Baseline        General Praise (GP)         Descriptive Praise (DP)        GP+HP item          DP+ HP item  
(neutral statement)   (general praise statement)  (descriptive praise statement)           (general + access)       (descriptive + access) 
 
If Highly preferred condition, what was the highly preferred item? 
Trial 
# 
Provide 
correct 
SD 
 
Wait 5s 
before 1st 
prompt/ or 
give when 
incorrect 
Wait 5s 
before 2nd 
prompt/ or 
give when 
incorrect 
Provide 
correct 
consequence 
for response 
(see above) 
Enthusiasm Rating 
(Please rate the 
experimenter?s enthusiasm 
by circling 1, 2, 3, 4) 
1= negative tone/mean 
2= no enthusiasm/ neutral  
3= enthusiastic/excited 
4= overly enthusiastic 
Comments 
1 Y     N Y  N  N/A Y  N  N/A Y     N 1   2   3   4   N/A  
2 Y     N Y  N  N/A Y  N  N/A Y     N 1   2   3   4   N/A  
3 Y     N Y  N  N/A Y  N  N/A Y     N 1   2   3   4   N/A  
4 Y     N Y  N  N/A Y  N  N/A Y     N 1   2   3   4   N/A  
5 Y     N Y  N  N/A Y  N  N/A Y     N 1   2   3   4   N/A  
6 Y     N Y  N  N/A Y  N  N/A Y     N 1   2   3   4   N/A  
7 Y     N Y  N  N/A Y  N  N/A Y     N 1   2   3   4   N/A  
8 Y     N Y  N  N/A Y  N  N/A Y     N 1   2   3   4   N/A  
9 Y     N Y  N  N/A Y  N  N/A Y     N 1   2   3   4   N/A  
Were there 9 trials in this session?    Y      N 
Other comments:      /46 = _____TI% 
      IOA: _____________   = IOA% _______  
          46 
 
  
 
 
 
 
 
 
 
 
 
 
 
IOA=       # Agreements 
 # Agree + Disagree (46)