COMPARING FUNCTIONAL ASSESSMENT METHODOLOGIES:
A QUANTITATIVE SYNTHESIS
by
CAITLIN V. HERZINGER
(Under the Direction of Jonathan M. Campbell)
ABSTRACT
Problem behaviors are a commonly associated feature of people diagnosed with autism.
Functional assessment has become the norm for evaluating these behaviors. There has been
much research concerning functional assessment over the past twenty years, but several
important research considerations have yet to be explained. One is the comparison of different
types of functional assessment (e.g., experimental functional analysis and behavioral functional
assessment). Controversy exists regarding the types of functional assessments and their validity.
The current study aims to compare the different methodologies of functional assessment and
their effectiveness in ascribing function to a target behavior and in the treatment selection that
follows such an assessment. Quantitative synthesis data were used to answer questions regarding
behavioral function, assessment type, and treatment effectiveness. Results indicate that
assessment type can impact treatment effectiveness and that there is a relationship between type
of behavior and ascribed behavioral function.
INDEX WORDS:
Autism, Problem behaviors, Functional assessment, Functional analysis
COMPARING FUNCTIONAL ASSESSMENT METHODOLOGIES:
A QUANTITATIVE SYNTHESIS
by
CAITLIN V. HERZINGER
B.S., The University of Florida, 2001
A Thesis Submitted to the Graduate Faculty of The University of Georgia in Partial Fulfillment
of the Requirements for the Degree
MASTER OF ARTS
ATHENS, GEORGIA
2005
© 2005
Caitlin V. Herzinger
All Rights Reserved
COMPARING FUNCTIONAL ASSESSMENT METHODOLOGIES:
A QUANTITATIVE SYNTHESIS
by
CAITLIN V. HERZINGER
Electronic Version Approved:
Maureen Grasso
Dean of the Graduate School
The University of Georgia
May 2005
Major Professor:
Jonathan M. Campbell
Committee:
David L. Gast
A. Michele Lease
TABLE OF CONTENTS
Page
LIST OF TABLES......................................................................................................................... vi
LIST OF FIGURES ...................................................................................................................... vii
CHAPTER
1
Introduction....................................................................................................................1
Autism Definition and Prevalence Rates...................................................................1
Approaches to Assessing and Treating Problem Behaviors......................................2
Purpose of Current Study ........................................................................................13
2
Method .........................................................................................................................15
Study Identification and Selection ..........................................................................15
Estimating Effects of Behavioral Interventions ......................................................16
Statistical Analyses..................................................................................................20
3
Results..........................................................................................................................21
Characteristics of Participants .................................................................................21
Characteristics of Studies ........................................................................................21
Characteristics of Functional Assessment, Behavioral Intervention, and
Experimental Quality ........................................................................................22
4
Discussion ....................................................................................................................40
Summary of Findings ..............................................................................................40
Implications for Clinicians ......................................................................................41
iv
Limitations of the Current Study.............................................................................42
Directions for Future Research................................................................................43
REFERENCES ..............................................................................................................................45
APPENDICES ...............................................................................................................................53
A
Coding Form used in Meta-Analysis ...........................................................................53
B
Articles used in Meta-Analysis....................................................................................62
C
Interrater Reliability Data ............................................................................................70
v
LIST OF TABLES
Page
Table 3.1: Participant Characteristics ............................................................................................26
Table 3.2: Description of Study Characteristics ............................................................................28
Table 3.3: Assessment, Intervention, and Experimental Characteristics.......................................29
Table 3.4: Descriptive Statistics for Three Effect Sizes ................................................................34
Table 3.5: Descriptive Statistics for Assessment Type and Function............................................35
Table 3.6: Treatment Effect Sizes by Function .............................................................................36
Table 3.7: χ2 for Function by Behavior Type ...............................................................................38
Table 3.8: χ2 for Behavior Type by Function ...............................................................................39
vi
LIST OF FIGURES
Page
Figure 1.1: Hypothetical EFA graph results ....................................................................................9
vii
CHAPTER 1: INTRODUCTION
Autism Definition and Prevalence Rates
Autism is a pervasive developmental disorder that is characterized by qualitative social
impairment, communication delays, and restricted patterns of behavior. For persons with autism,
social impairments are often typified by marked impairment in using and understanding
nonverbal social cues, such as facial expressions and body posture. Communication delays
noted in persons with autism can range from a total lack of developmentally appropriate verbal
or nonverbal language to the inability to initiate or sustain a conversation with others. The
restricted patterns of behavior associated with autism include those behaviors that are abnormal
in intensity or focus as well as an inflexible adherence to nonfunctional routines and rituals. The
onset of autism is prior to 3 years of age. The prevalence rate of autism is currently estimated at
2-20 cases per 10,000 and a male to female ratio of 4:1 (Diagnostic and Statistical Manual IVTR, American Psychiatric Association [APA], 2000). Autism is often associated with Mental
Retardation, which can range from mild to profound. Often the cognitive profile of an individual
with autism is skewed with verbal skills being weaker than nonverbal skills. In addition to the
core features of autism, individuals with autism often show behavioral symptoms such as deficits
in attention, hyperactivity, sensory integration difficulties, and temper tantrums (APA, 2000).
One of the most consistently noted concerns for individuals with autism is their propensity to
engage in problem behaviors, such as aggression towards self and others.
Individuals with autism may demonstrate a wide range of problematic behaviors, such as
self-injurious behavior [SIB] (e.g., self-hitting, head banging, eye poking), stereotypic behaviors
1
(e.g., body rocking, hand flapping), property destruction, aggression towards others, and severe
disruptions (e.g., tantrums). Epidemiological studies suggest that 13-30% of children with
autism engage in problematic behaviors so severe that intervention is warranted (Horner et al.,
2002). According to Wicks-Nelson and Israel (1999), 15% of developmentally disabled children
engage in SIB. Emerson et al. (2001) reported that “challenging behaviors are shown by 10-15%
of people with mental retardation who are in contact with educational, health, or social care
services” (p. 77). These data, although not comprehensive, highlights the importance of
assessing and treating these maladaptive behaviors. Aside from tissue damage and property
destruction, problem behaviors can interfere with new learning, prevent acquisition of adaptive
behavior and compete with socially acceptable behaviors thus having a negative impact on social
relationships, academic performance, and overall personal growth.
Approaches to Assessing and Treating Problem Behaviors
Psychoanalytic Approach
Once the primary method for treating individuals with autism, the psychoanalytic
approach has since fallen out of favor with most researchers and clinicians in the area. This
perspective typically focuses on the relationship between parent and child when treating children
with autism and their problematic behaviors. Psychodynamic treatments are based on the
assumption that autism is caused by emotional trauma or parental mishandling and treatments are
focused on procedures that attempt to “reduce the core hypothesized inner conflict the children
[were] thought to be exhibiting” (Koegel, Koegel, & McNerney, 2001, p. 19). Bettelheim
(1967), perhaps the most well-known proponent of the psychoanalytic model, suggested that
children with autism in a supportive environment will gradually develop trust and no longer feel
threatened by their parents.
2
Gentle therapy and holding therapy are two of the more common psychoanalytic
treatments that attempt to create a supportive environment for the individual with autism.
Holding therapy is described as forcibly holding the child “so as to cause the autistic defense . . .
to crumble” (Welch, 1987, p. 48). The theoretical assumptions, treatment procedures, and
outcome claims made by Bettelheim and others have been challenged by researchers (Charlop,
Schreibman, & Kurtz, 1990). Studies comparing children receiving psychoanalytic treatment to
those receiving other forms of treatment and no treatment at all, found the effectiveness of the
psychoanalytic treatment to be inadequate in comparison (Levitt, 1963; Bartak & Rutter, 1973).
For further reading on the psychoanalytic approach to treating children with autism, see
Psychoanalysis with Children: History, Theory, and Practice (Rodriguez, 1999).
Psychopharmacological Approach
The psychopharmacological approach to the assessment and treatment of maladaptive
behaviors in people with autism targets specific behaviors for treatment, such as “resistance to
change, ritualistic/compulsive behaviours, hyperactivity, aggressive behaviours and sleep
problems” (Gringas, 2000, p. 229). Overactive dopaminergic activity is thought to be the cause
of several problem behaviors, notably stereotypies and general overactivity, and is often treated
with dopamine receptor blockers such as haloperidol, a traditional antipsychotic (Volkmar,
2001). There is also research regarding the role of serotoninergic dysfunction in the etiology of
autism (Schultz & Anderson, 2004). Like dopamine, altering the levels of serotonin in the body
have been shown to improve some symptoms of autism. Schultz and Anderson report that
serotonin reuptake inhibitors have had a positive affect on reducing autistic symptoms. They
also note that symptoms of autism have been exacerbated by depletion of tryptophan, a precursor
to serotonin production. Stimulants, antidepressants, and melatonin have also been used to treat
3
problem behaviors. According to Bryson, Rogers, and Fombonne, (2003) “There is no curative
treatment for autism, and psychotropic drugs have only a minimal role to play in its
management. As a rule, drugs should be used sparingly and only when other strategies to reduce
maladaptive behaviours have been properly tried and have failed to bring about the desired
changes” (p. 512). In any case, careful clinical and laboratory monitoring is recommended when
psychotropic drugs are part of the treatment package.
Sensory Integration Approach
Dr. A. Jean Ayres is the founder of Sensory Integration therapy (SI), a sensory-motor
treatment based upon theories developed over the last 30 years. Advocates of this theory
maintain that sensory integration is an innate neurobiological process (Hatch-Rasmussen, 1995).
Proponents suggest that individuals with autism and other developmental disabilities experience
dysfunction in which sensory input is not integrated or organized appropriately by the brain. SI
therapy is viewed as a direct intervention that can improve nervous system function. This is
done by providing the child with enhanced levels of sensory information gleaned during physical
activities that are meaningful to the child, and that elicit adaptive behaviors (Koomar & Bundy,
1991).
There are three forms of stimulation often used in conjunction with one another during SI
treatment: a) vestibular; b) proprioceptive; and c) tactile (Fisher et al., 1991). Vestibular
treatments refer to engaging the individual in specific movements to provide information about
where the body is in space (e.g., use of suspended swings). Proprioceptive input refers to touch
or pressure applied to the individual to provide general body awareness (e.g., use of weighted
vests, hug machines). Tactile input is used to provide information about objects in the
environment (e.g., use of textured toys). Thus far, little research has supported the anecdotal
4
claims that SI is successful at reducing problem behaviors in children or adults with autism. In
fact, in one study, SI was shown to increase problem behaviors (Mason & Iwata, 1990). Despite
the lack of supportive evidence, the SI approach is often used in conjunction with other
treatments as a way to treat individuals with autism. The interested reader is referred to Dawson
and Wattling (2000) and Baranek (2002) for more in depth reviews of the literature regarding SI
treatment for individuals with autism.
Behavioral Approaches
The behavioral approach to treating maladaptive behaviors has consistently focused on
identifying antecedents and consequences of problem behavior and typically uses the operant
conditioning approach during treatment. Prior to the 1980’s, problem behaviors were often
treated with punishment, without concern for the contingencies maintaining the behavior.
Punishment was a successful treatment strategy (i.e., reduced problem behaviors) because “its
effectiveness [did] not depend on its ability to alter a reinforcement contingency” (Iwata, 1982,
p. 5). Interventions were often implemented arbitrarily without regard to the contingencies that
maintained the problem behavior (Kahng, Iwata, & Lewin, 2001).
With the growth of research defining and describing functional assessment of problem
behavior, the behavior analytic perspective is quickly becoming the “gold standard” for the
assessment and treatment of problem behaviors. Carr (1977) and Johnson and Baumeister
(1978) were some of the first to document the usefulness of identifying the environmental events
that maintain problem behaviors. Behavior analysts hold the view that all behavior responses
have a function and are observable demonstrations of cause-effect relationships (Skinner, 1953).
To date, a few functions of problem behavior have been identified: a) escape from academic
tasks or disliked events; b) attention from peers, parents, or teachers; c) tangible–characterized
5
as leading to a physical or food reinforcer (e.g., toy or cracker); or d) automatic function, in
which it is assumed that the behavior is nonsocially mediated or maintained independent of
social consequences (i.e., self stimulation). The adoption of at least part of the behavioral
approach by the federal government with the amendments to the Individuals with Disabilities
Education Act [IDEA] (1997) emphasizes the importance of the behavioral approach to
assessment and treatment of problematic behaviors (Gresham et al., 2004).
Functional Assessment
Functional Assessment (FA) is an umbrella term used to describe any methodology in
which the goal is to document the “function” of a behavior. Function can be defined as the
“purpose” of any behavior and describes the contingencies that are antecedent and consequential
to the behavior. Once function is determined, the results of the FA are to be used in treatment
selection for reducing maladaptive behaviors, rather than selecting treatments arbitrarily. For
example, if the FA indicates that an individual is engaging in a problem behavior (e.g.,
aggression) to gain attention, the treatment may include the introduction of an alternative way to
gain attention (e.g., saying “Help”) and extinction to decrease the likelihood that the aggressive
behaviors are reinforced with attention. Kahng, Iwata, and Lewin (2002) compared the
effectiveness of treatments based on FAs to arbitrarily chosen ones and found treatments based
on FAs to be more effective.
FA encompasses indirect assessments (e.g., interviews, questionnaires, rating scales
such as the MAS and QABF); descriptive assessments (e.g., A-B-C sheets, direct observation
with no variable/environment manipulation); and experimental functional analyses (EFA), also
known as “true” or traditional FAs (e.g., analogue conditions in which variables are
manipulated). “The primary benefit of functional assessments has been to increase the precision
6
and resulting effectiveness of reinforcement-based interventions” (Kahng et al., 2002, p. 130).
Functional assessment has become foundational in the study of aberrant behaviors. For example,
Kahng et al. report that there has been a dramatic increase in the number of published data sets
investigating functional assessments since the early 1980’s. This increase coincides with and is
presumed to be a result of the increase of data sets that included an EFA.
Based on the ability of FAs to ascribe function to aberrant behaviors and thus improve
treatment effectiveness of those behaviors through treatment selection (Kahng at al., 2002;
Campbell, 2003), FAs have become an integral part of the entire treatment process. In 1997, the
amendments to IDEA required the use of functional behavioral assessment and positive
behavioral interventions based on these assessments when student placement may be changed
due to problem behavior. Prior to this, these assessments were considered “best practices” but
were not mandated by federal law (Horner & Carr, 1997; Sugai, Horner, & Sprague, 1999).
IDEA does not, however, define what constitutes a valid behavioral functional assessment.
Because of this lack of detailed description, school districts and other professionals are using less
specific models of assessing function. The current review of the literature aims to delineate the
differences in accuracy and effectiveness of different functional assessment methodologies.
Experimental Functional Analysis (EFA) and Behavioral Functional Assessment (BFA)
EFA represents simulation of natural environments and is the primary tool for
demonstrating causal relationships (Carr, Langdon, & Yarborough, 1999). With EFA, the goal is
to isolate the function or purpose of the problem behavior. This allows the therapist to directly
and specifically alter the social and physical environment in a way that is most likely to alter the
problem behavior. Iwata et al. (1982) defined the first standardized and comprehensive method
for conducting functional analyses. As outlined by Iwata et al. there are four different conditions
7
in an EFA: social disapproval, academic demand, unstructured play, and alone. In the social
disapproval condition, the client receives therapist attention contingent on problem behavior.
The social disapproval condition is thought to mimic naturalistic occurrences where individuals
may respond with attention when the problem behavior is exhibited. This “attention” includes
one or more of the following: emotional behavior, physical contact, verbal reprimand, and
consolatory attention. All other behaviors, even those that are socially acceptable, are ignored.
In the academic demand condition, the client is presented several tasks and permitted escape
contingent on targeted behavior. The unstructured play condition serves as a control for the
experimenter’s presence and is conducted in a room filled with toys and other possibly
stimulating materials. In the alone condition, the therapist is assessing the possibility that the
target behavior is self-reinforcing.
Through each condition of the EFA, frequency data are collected. Typically, conditions
are presented more than once and in a random order to prevent order bias. The frequency data
are graphed and rate of behavior comparisons are made across condition types. If the rate of
behavior is significantly higher in one condition, then it is assumed that the problem behavior is
a function of that condition. For example, if the problem behavior occurs at much higher rates in
the tangible condition, it is assumed that the purpose of the behavior is to gain access to tangible
items. See Figure 1.1 on the following page for a visual representation of this example. This
information would be used as an integral part in the treatment selection process. Following the
assessment, treatment selection might include some type of differential reinforcement of
alternative behavior (DRA) such as having access to items contingent on pointing to a picture of
the wanted item.
8
Figure 1.1
EFA Results
Hypothetical results of an EFA
18
attention
escape
tangible
alone
16
Aggressions per minute
14
12
10
8
6
4
2
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Sessions
BFA represents the other two categories mentioned under the umbrella term FA: indirect
and descriptive assessments. These assessments range from brief interviews with parents in
which they are asked questions such as, “Why do you think the problem behavior is occurring?”
to long, extended observations in which the subject is observed in a variety of settings. The
Motivation Assessment Scale (MAS; Durand & Crimmins, 1988), Questions About Behavioral
Function (QABF; Matson & Vollmer, 1995) and Motivation Analysis Rating Scale (MARS;
Wieseler et al., 1985) are three recognized indirect methods of assessing behavioral function.
The Functional Assessment Observation Form (FOAF; O’Neill et al., 1997) is a direct
observation method of behavioral functional assessment. Some research has been reported
comparing the validity of these methods of assessment to EFA. For example, the MAS ratings
9
have been correlated with EFA analogue data and were found to be highly significant (r = .99, p
< .001) (Durand & Crimmins). Thus, “the teacher ratings on the MAS predicted their student's
behavior in the experimental condition” (Durand & Crimmins, p. 112). In almost all published
accounts of comparison data, the EFA represented the gold standard for validity tests of other
types of assessment.
Because all BFAs are not administered in the same way (e.g., rating scales, observations,
interviews), interpretations of the assessments differ. Some BFAs are simply used to gain
qualitative information that is later combined with other sources of information to identify the
function of maladaptive behavior. Other BFAs, specifically the rating scales, use cut scores or
relative rankings to identify the functions of the targeted behavior. For example, the authors of
the MAS, a 16-item questionnaire, note that if one condition (e.g., attention) has clearly received
the highest score it is assumed that this is the function of the behavior. The relative rankings of
conditions are used to determine the most important influence on the behavior (Durand &
Crimmins, 1992). On the MAS, like all other measures of behavioral function, more than one
possible function may be identified. The authors provide a guide to interpret most outcome
results, including decision rules when two functions are scored within .25 to .50 points of each
other. In this scenario, both categories would be considered influences that may be causing the
problem behavior to continue (Durand & Crimmins). No matter the method of interpretation, the
results of BFAs are used in the same manner as those from EFAs. The qualitative data from
interviews, rating scale results, and direct observations are often combined and if possible a
single function of behavior is identified. Treatment selection is then based on the outcome of the
gathered information.
10
Benefits of EFA when compared to BFA. Much research has been published regarding
the effectiveness of EFA. EFA has provided methods to test the notion that each behavior has a
function that can be observed. EFA is able to identify causal and maintaining factors of problem
behavior and has led to the development of effective treatments for severe problem behaviors.
EFA is a specific and structured assessment method, in which the analog conditions and the
contingencies set in place for each condition are defined a priori. The structure and conditional
quality of the EFA makes replication possible. EFA allows the clinician/ practitioner to base
treatment selection on an observable, recorded (not simply reported) function of behavior. The
primary benefit of EFA is that this methodology allows for causal inferences to be made
regarding the purpose of the targeted maladaptive behavior.
In contrast, BFA does not allow for systematic manipulation of environmental variables
(antecedents or consequences) and involves a recording of what is seen (or noticed) in the
interaction. Therefore, BFA methodologies are correlational in nature. BFA is less thorough
and is more susceptible to data recorder biases. For example, a teacher completing a functional
rating scale about her own observations may be more likely to recall instances in which her
attention was drawn from a target activity to a problem behavior that an instance in which a child
engaged in problem behaviors in order to get a break from difficult academic tasks.
In 1994, Horner wrote that although different situations may require different functional
assessment procedures, EFA “will remain the expectation within published research” (p. 402).
Although other procedures for functional assessment should be created, considered and
researched, the clinical standard should include the following four standards: a) problem
behaviors are operationally defined, b) antecedent behaviors are identified, c) hypotheses are
11
developed concerning variables maintaining problem behaviors, and d) direct observation data
are collected (Horner).
Benefits of BFA when compared to EFA. According to research by Sturmey (1995), the
experimental design of EFA presents a number of procedural and psychometric problems. EFAs
are time consuming, require extensive training, and are not cost-effective. Matson et al. (1999)
reports that a single EFA can require several hours daily over a two- to three-week-period. Iwata
et al. (1994) summarized data from over 150 EFAs and reported that the mean length of
assessment was 26 sessions, which represented approximately 6.5 hours of direct observation,
not to mention the time spent preparing for each condition, analyzing data, and creating a visual
representation in the form of graphs. When assessing a severe behavior problem that is life
threatening (e.g., head banging), this time constraint may have vital consequences and immediate
attention may be necessary.
Thorough training is necessary for the service provider conducting the EFA sessions as
well as the behavior data specialists who record the antecedents, behaviors, and consequences of
the interactions during the analogue sessions. Limitations in mental health work funding have
exposed the monetary constraints on using EFA as an automatic prelude to behavioral treatment
(Applegate, Matson, & Cherry, 1999). Martin, Gaffan, and Williams, (1999) also noted
problems with poor test-retest reliability of analogue conditions. Martin and colleagues (1999)
further identified the potential problems with “imposing a contingency on a response that is
likely to occur may result in the response coming under the control of that contingency, even if it
was not maintained previously by such a contingency in the person’s natural environment” (p.
126). Other researchers have noted several other limitations of EFA procedures, including non-
12
naturalistic environment and intentional increases in life threatening problem behaviors
(Dawson, Matson, & Cherry, 1998; Axelrod, 1987).
BFAs, in contrast, do not require significant amounts of time and are easier to administer.
They are dependent on the non-compulsory training of the raters, observers, and other
informants. Also, BFAs do not include environmental manipulations that intentionally increase
the rates of problem behaviors. BFAs are naturalistic and do not require careful experimental
design.
Purpose of Current Study
The focus of the current study is to compare different types of FA methodologies,
including analogue conditions, interviews, observations, rating scales, and combinations of these
methods. Similar to Toogood and Timlin (1996), the study aims to compare the different
methodologies of FA in a) similarity of ascribing function to a target behavior, b) treatment
selection that follows such an assessment, and c) differences in treatment efficacy depending on
type of FA methodology. The purpose of the following quantitative synthesis is to answer the
following questions regarding the assessment and treatment of problem behaviors for people
with autism:
1)
Is treatment more effective when following an experimental functional analysis
(EFA) or a behavioral functional assessment (BFA)? I hypothesize that
treatment based on the more standardized and thorough EFA will be more
effective in delineating the specific purpose of aberrant behavior. The EFA is
experimental in nature and controls for most observer bias, unlike the BFA.
13
2)
Do different types of FAs lead to different treatment selections? This question
was noted by Pelios, Morren, Tesch, & Axelrod (1999) as an important
research consideration. This question is exploratory in nature.
3)
Is there a predominant observed function based on the type of assessment? This
question is exploratory in nature.
4)
Does the observed function of the behavior, regardless of FA method used,
have an impact on the effectiveness of treatment? I hypothesize that behaviors
identified as having an automatic function will be more difficult to treat. This
notion is based on the previous research of Vollmer (1994), Piazza, Hanley,
and Fisher (1996), and many others who found that treatment development for
behaviors maintained by automatic reinforcement is more difficult because the
reinforcer may be unknown or not easily manipulated.
5)
Is there a relationship between the type of maladaptive behavior observed and
the obtained function for that behavior? For example, is self injury more often
attributed as having an escape component? I hypothesize that external
maladaptive behaviors (e.g., aggression and property destruction) will be often
identified as having an attention and/or tangible function. This hypothesis has
been suggested by other researchers (Dawson, Matson & Cherry, 1998).
14
CHAPTER 2: METHOD
Study Identification and Selection
For the years 1998 through 2003, published functional assessments of problem behavior
for people with autistic disorder were identified through searches of PsycLit, ERIC, and
MedLine databases using appropriate search terms, such as subject descriptions (e.g., autism,
autistic disorder), target behaviors (e.g., self-injurious behaviors, aggression, problem behaviors),
and assessment type (e.g., applied behavior analysis, functional assessment, functional analysis).
Published studies were also identified by issue-by-issue hand searches of relevant journals
including Journal of Applied Behavior Analysis, Research in Developmental Disabilities, and
Behavioral Interventions (see Table 3.2 for complete list). Also, timely references (i.e., citations
between 1998 and 2003) from each article found through the literature search were reviewed for
possible inclusion.
Studies were selected for inclusion if the following criteria were satisfied. First, studies
were selected if they were published in peer reviewed journals between January 1998 and
December 2003. Second, single case studies were included only if a participant was diagnosed
with autistic disorder, described as meeting the criteria for autistic disorder, or described as
autistic or autistic-like. Third, a functional assessment had to be conducted and results reported,
with problem behaviors as the target behaviors of treatment. For those articles used in the
treatment effectiveness analyses, the studies were included if: a) data points, not just mean
scores, were reported; b) baseline data and treatment data were reported; and c) if the
intervention procedures targeted reduction of stereotyped, self-stimulatory, self-injurious,
15
destructive, disruptive, or aggressive behaviors. If an article included multiple participants or
studies that only partially met inclusionary criteria, only those components that met criteria were
included in the review.
Estimating Effects of Behavioral Interventions
Effect Size Calculations
There are several methods for assessing effectiveness data using both regression and nonregression approaches. Frequently reported summary methods have involved the calculation of
Mean Baseline Reduction (MBLR), Percentage of Non-overlapping Data (PND), and Percentage
of Zero Data (PZD) (Campbell, 2003). In the current study, three effect sizes based on
nonregression approaches were calculated per intervention: MBLR, PND, and PZD. The
MBLR was calculated by subtracting the mean of treatment observations from the mean of
baseline observations then dividing by the mean of baseline observations and multiplying by 100
(Campbell, 2003; Lundervold & Bourland, 1988; O’Brien & Repp, 1990). The PND statistic
was calculated as the percentage of treatment data that did not overlap with baseline data points
(Scruggs, Mastropieri, & Casto, 1987). If a baseline phase reported one or more data points of
zero, then the same number of data points was excluded in the treatment phase prior to
calculation of the PND (Didden et al., 1997). The PND can range from 0 to 100%. According to
Scruggs, Mastropieri, Cook, and Escobar (1986) a PND greater than 90% reflects a highly
effective treatment, a PND of 70-90% is considered a fair treatment outcome, and a PND of less
than 50% indicates unreliable/ineffective intervention. The PZD statistic was calculated by
locating the first intervention data point that reached zero and computing the percentage of data
points that reached zero subsequent to this first zero (Scotti et al., 1991). The PZD score is
considered a more stringent efficacy indicator as it requires target behaviors to reach and stay at
16
zero levels throughout treatment to be considered effective. Campbell (2004) noted that the PZD
score represents a "degree of behavior suppression versus degree of behavior reduction" (p. 235).
PND and PZD scores have been found to be independent indicators of treatment outcome
(Campbell, 2003).
Handling Multiple Outcomes, Participants, Assessment types, and Experimental phases
Several rules were established for the coding of assessment type. Functional assessment
type was coded as either: a) EFA (strictly adhering to guidelines set forth in Iwata, 1985), b)
modified EFA, c) ABC sheet, d) rating scales (e.g., MAS, QABF), e) informal assessment or f)
other. Under the modified EFA code, brief EFA sessions, (i.e., shorter sessions than those
described by Iwata) such as those described by Northup et al. (1991) and summarized in Derby
et al. (1992) were coded. Later, this group (modified EFA) was consolidated with the EFA
group. Thus, articles were coded as EFA if they included analogue conditions and manipulated
variables in the environment in each condition.
If two different types of FAs (e.g., EFA and MAS) were used with a participant, the
methods were coded separately with the possibility of two functions and different treatments
identified. If a participant's problem behavior was assessed using multiple BFA methods (e.g.,
MAS, parent interview, and observation) the assessments were coded as a combination. In such
a case, the coding resulted in one effect size unless BFAs yielded different functions for each
method. In the current data set, the BFA methods did not identify contradicting functions, thus
results never required separate coding.
Studies that reported on multiple outcomes or multiple participants required separate
effect size calculations for each outcome for each participant. When more than one problem
behavior was targeted for a participant and separate data points were reported, individual effect
17
sizes were calculated per problem behavior per participant. For example, a hypothetical study
reported data for two different participants using two different assessment methodologies for
each and targeting two behaviors for participant A and one behavior for participant B. For
participant A, a teacher interview and EFA were conducted both ascribing function to escape
from academic tasks. Only one treatment (FCT plus extinction) was implemented. For
participant A, 10 baseline data points and 20 treatment data points were reported and all three
nonregression effect sizes were computed and reported. Identical effect sizes were coded for
both EFA and BFA. For participant B, a functional rating scale assessment was conducted. The
data suggested that the behavior was maintained by different functions in different settings (i.e.,
escape at school; attention at home) and two different treatment packages were implemented.
The baseline and treatment data points from each setting specific observation were reported
separately, thus for participant B three effect sizes were calculated for behavior at school and
three effect sizes were calculated for behavior at home.
Single case designs vary (e.g., A-B; A-B-A-B) and effect sizes can be calculated from
varied contrasts (Allison & Gorman, 1993). In the present study, the effect sizes were calculated
between the first nontreatment phase and the last treatment phase, per Faith’s et al.’s (1996)
recommendations and implemented in Campbell (2003). In designs that compared multiple
treatments (A-B-A-C), the final treatment phase was coded.
Data extraction, Variables coded, and Reliability check
For the necessary analyses in the present study, the graphs provided by the articles were
transformed into raw data via a ruler. The distance between each point and the abscissa was
calculated in millimeters and rounded to the nearest .05. The data conversion procedure has
18
been used by Allison, Faith, and Franklin (1995) and Campbell (2003) with a high degree of
interrater reliability.
The following participant information was coded when available: participant’s age,
gender, intellectual functioning, secondary diagnoses, years since autism diagnosis prior to
study, and years of prior treatment. The following assessment/pre-intervention data was coded:
target behavior, type of FA (e.g., EFA, informal assessment), function(s) of behavior, type of
intervention used (e.g., reinforcement only, combination of extinction and punishment), length of
session, treatment setting, and type of therapist. Targeted behaviors were coded as: aggression,
property destruction, disruptive behaviors (e.g., spitting), vocalizations, SIB, and stereotyped
behaviors. If relevant, specific types of SIB were also recorded. These subtypes included: head
banging, face slapping, eye gouging, biting, and scratching.
The following intervention data was coded: type of intervention, type of experimental
design, interrater reliability, number of baseline data points, number of final phase treatment
points, and attempt to generalize treatments. The types of intervention coded included:
reinforcement, punishment, timeout, extinction, sensory extinction, FCT, combined treatments,
and other interventions. These categories were later consolidated into six categories: a)
reinforcement only, b) punishment only, c) extinction only, d) reinforcement and punishment, e)
extinction plus reinforcement or punishment, and f) other. Refer to Appendix A for copy of the
coding form used.
Twelve articles were randomly selected for independent coding by an advanced graduate
student in School Psychology, who had experience working with individuals with autism, and
interrater agreement was established. The 12 articles (20.69% of all articles) included 17
separate assessments (16.04% of all assessments) and 13 different participants (16.05% of all
19
participants). Interrater agreement, with a mean of 99.06% and a range of 82.35% to 100%
across all coded variables, was determined by the percent agreement method (# of agreements / #
of agreements + # of disagreements X 100). See Appendix C for a detailed reliability summary.
Statistical Analyses
Descriptive analysis of participants, studies, interventions, and experimental features
Participants and articles were examined through descriptive statistics. Characteristics of
the articles and studies located included: the number of articles located; the number of studies
extracted from each article; and the journals contributing to the review. Participant
characteristics were examined next. Gender, age, race/ethnicity, level of mental retardation and
other variables were summarized. This analysis gave an indication of the representativeness of
the sample. Assessment, intervention, and experimental characteristics of the studies were
examined as well (e.g., percentage of studies that used EFA methodology; percentage of studies
who targeted aggression). One-way ANOVAs were used to examine research question 1 (a
comparison of treatment effectiveness for EFA and BFA) and research question 4 (treatment
effectiveness as impacted by function). Two variable chi-square tests of non-independence of
categorical variables were used to assess research question 2 (FA type’s impact on treatment
selection), research question 3 (possible bias in assessment outcomes based on FA type), and
research question 5 (relationship between behavior type and behavioral function).
20
CHAPTER 3: RESULTS
Characteristics of Participants
This review included 58 articles reporting on 81 participants with a total of 106 separate
FAs. Similar to prior reviews documenting the higher prevalence of autism in males, the
majority of participants included in the review were male. However, the ratio of males to
females in this study (2.5 : 1) was somewhat lower than gender ratios reported in the autism
literature, typically 4 : 1. Consistent with prior reviews documenting the prevalence of mental
retardation in individuals with autism, the majority of the participants (77.7%) included in the
present review functioned in the range of mental retardation or were considered “untestable” via
standardized, formal intelligence testing. In most cases (84%), the criteria used to diagnose
autism were not reported; however, when the criteria were reported, they were most commonly a
version of those presented in the DSM (9.8%). In the great majority of cases, information
regarding a client’s time since diagnosis, time in prior treatment, or current use of prescription
medication was excluded. Detailed information about the characteristics of participants is
presented in Table 3.1.
Characteristics of Studies
Studies included in the research synthesis were collected from a total of six journals, with
the Journal of Applied Behavior Analysis contributing the highest percentage of articles (56%).
Studies most often included one participant and rarely included more than two. Detailed
information about the characteristics of studies included in the meta-analysis is presented in
Table 3.2.
21
Characteristics of Functional Assessment, Behavioral Intervention,
and Experimental Quality
Studies employed both experimental (75.47%) and non-experimental (24.53%) methods
of FA. Under the EFA umbrella, the majority of assessments were Modified EFA (45%), based
on the analogue conditions of Iwata and colleagues, but tailored in terms of time or specific
conditions used to the particular needs of the researcher, clinician, and/or participant. The type
of BFAs most often reported were described as informal assessment (53.8%).
The studies included in the quantitative synthesis targeted the reduction of more than one
problematic behavior more often than any single maladaptive behavior. When assessed
individually, aggression (20%) and SIB (18%) were the most commonly targeted behaviors. In
combination with other maladaptive behaviors, aggression (28%) and disruption (18%) were the
most commonly reported among the participants. Reinforcement only (20%) was the most
commonly reported intervention type. Functional communication training (FCT) was the most
commonly reported form of treatment included under the “reinforcement” coding. FCT
constituted 43% of the treatments coded as reinforcement. A combination of extinction and
either reinforcement or punishment (16%) was another common intervention technique. In 16%
of the studies that claimed treating the targeted behaviors assessed, no specific intervention
information was reported. In a majority of cases (65.6%), behavioral therapist could not be
coded due to lack of information reported by the authors. Inpatient hospital setting was the
treatment setting that occurred with the highest frequency (23.4%).
The simple AB experimental design was the most commonly used, reported in 37.5% of
the studies. Studies in the meta-analysis omitted follow up data collection (67.2%) more often
than included follow-up data collection (25%). When follow up data was collected, the intervals
22
from treatment to follow up data collection ranged from two weeks to two years with an average
interval of approximately 8.6 months. The most commonly reported follow up interval was 6
months, reported in 6.3% of studies. In 5 of 64 cases, there was no clear indication of whether
follow-up data was collected or not. Generalization data were omitted from the studies more
often than reported. When reported, the data indicated that generalization to new situations and
settings was the most common form of generalization data reported (50%). Inter-rater reliability
for FA sessions was reported in 80.2% of articles. The reliability ranged from 84.0 to 100.0 with
a mean of 94.6 and a standard deviation of 3.7. For treatment sessions, inter-rater reliability data
was reported in 93.8% of articles. The reliability ranged from 87.0 to 100.0 with a mean of 95.3
and standard deviation of 3.3. Detailed information about functional assessments, behavioral
interventions, and experimental quality is presented in Table 3.3.
Treatment Effectiveness Based on FA Type
Three one-way ANOVAs showed that when comparing EFA and BFA, there is a
significant difference in treatment effectiveness as measured by the PZD statistic, F (1, 60) =
7.58, p < .01). EFA and BFA did not differ when treatment effectiveness was measured with
MBLR, F (1, 60) = 1.35, n.s.) or PND, F (1, 60) = .99, n.s.). Therefore, statistical analyses
indicate that treatment is more effective when based on the results of EFA as compared to BFA,
when the PZD statistic is used to assess effectiveness. See Table 3.4 for means and standard
deviations. In summary of this finding, it appears that when behavior suppression (as measured
by the PZD statistic), rather than behavior reduction is the goal of treatment, intervention
packages based on the results of EFAs are more effective than those based on results of BFAs.
23
Is Treatment Selection Impacted by Function?
A two variable χ2 test of non-independence shows that the type of treatment selected
(i.e., reinforcement only, combination of extinction and punishment) is independent from
ascribed behavioral function of the behavior, χ2(Ν = 25, 62) 32.94, n.s. This finding highlights
that although the results of FAs are intended to guide intervention selection, there is not a
predominant intervention that results from a FA. A one variable χ2 test was used to assess the
distribution of interventions that included FCT, a commonly reported treatment, across ascribed
functions. No significant differences exist in the distribution of the FCT intervention packages
in comparison to the non-FCT packages. Treatments for problem behavior are tailored to the
identified function of behavior, but no specific intervention types (e.g., reinforcement only) or
specific treatments (e.g., FCT) were found to have a significant relationship with a particular
identified function.
Is Ascribed Function Related to FA Methodology?
The data show that there is not a relationship between the type of methodology (e.g.,
EFA, BFA) used in the assessment and the result of that assessment, χ2 (N = 6, 106) 8.01, n.s.
This finding supports the notion that ascribed behavioral function is not a product of the type of
assessment employed. This is positive support for both EFA and BFA as independent
assessment methodologies (see Table 3.5 for frequency information).
Does Ascribed Function Impact Treatment Effectiveness?
The results from three one-way ANOVAs indicate that there is no statistical significance
for the three calculated effect sizes. Treatment effectiveness, as assessed by the a) MBLR
statistic, F (5, 56) = .90, n.s., b) PND statistic, F (5, 56) = 1.57, n.s., and c) PZD statistic, F (5,
56) = .90, n.s. was not significantly affected by the function of the problem behavior. Table 3.6
24
includes data regarding the ranges, means, and standard deviations for the three calculated effect
sizes: MBLR, PND, and PZD.
Is There a Relationship Between Problem Behavior Type and Identified Function?
The two variable χ2 test of non-independence indicates that there is a statistically
significant relationship between the type of problem behavior (i.e., internal versus external) and
the behavioral function, χ2 (Ν = 10, 106) 43.37, p < .001. The goodness of fit, one variable
χ2 test was then identified for each reported function. Internal behaviors were more likely
identified as having automatic functions. External behaviors were more often identified as
having escape and tangible functions. Combinations of internal and external behaviors were
identified as having combination functions (two functions). The attention, tangible, automatic,
and undifferentiated (3 or more functions) functions were significant when tested to determine
the specific relationships with types of behavior. This data is reported in Tables 3.7 and 3.8.
25
Table 3.1.
Participant Characteristics
__________________________________________________________________
Characteristic
n
%
r
M
SD
__________________________________________________________________
Gender
Male
58
71.6
Female
23
28.4
Age (in years)
3.42-49.00
Race/Ethnicity
African-American
Not reported
1
80
Level of Mental Retardation (IQ range)
Severe (<39)
38
46.9
Not reported/Other 17
21.0
Mild (70 - 55)
12
14.8
Moderate (54 - 40) 12
14.8
None (IQ > 70)
1
1.2
Untestable
1
1.2
Diagnostic criteria used
Not reported
68
84.0
DSM-IV
7
8.6
“Autistic-like”
5
6.2
DSM-III
1
1.2
26
13.53
9.90
Table 3.1 continued
__________________________________________________________________
%
r
M
SD
Characteristic
n
__________________________________________________________________
Months since diagnosis
100% Not reported
Months of prior treatment
(7 - 168)
Medication administered
Not reported
No
70
86.4
6
7.4
Yes
5
6.2
______________________________________________________________________________
Note. n = number of participants; % = percentage of participants; r = range; M = mean;
SD = standard deviation.
27
Table 3.2.
Description of Study Characteristics
__________________________________________________________________
Characteristic
n
%
__________________________________________________________________
Journal of Applied Behavior Analysis
56
52.8
Research in Developmental Disabilities
24
22.6
Behavioral Interventions
12
11.3
Journal of Autism and Developmental Disorders
8
7.5
Behavior Therapy
4
3.8
Journal of Clinical Psychopharmacology
2
1.9
106
Total N
Number of participants per study
1
42
72.41
2
11
18.97
3
4
6.90
5
1
1.72
58
Total N
__________________________________________________________________
Note. n = number of studies; % = percentage of studies.
28
Table 3.3
Assessment, Intervention, and Experimental Characteristics
__________________________________________________________________
%
Characteristic
n
__________________________________________________________________
Assessment and Intervention characteristics
Behavior targeted for reduction
Aggression
20
18.87
Self-injurious Behavior
18
16.98
Self-stimulation/stereotypic behavior
16
15.09
Disruptive Behavior
6
5.66
Pica
5
4.72
Property Destruction
3
2.83
Vocalizations
3
2.83
Other
1
.94
34
32.08
Aggression
28
82.40
Disruption
18
52.90
Property Destruction
13
38.20
SIB
7
20.60
Self stimulatory
4
11.80
Vocal
3
8.80
Combinations
Combination Breakdown:
29
Table 3.3 continued
__________________________________________________________________
Characteristic
n
%
__________________________________________________________________
Type of functional assessment
Experimental
80
Modified EFA
36
EFA
30
Brief EFA
10
Partial Experimental
75.47
4
Non-experimental
26
Informal Assessment
14
ABC sheet
8
Rating Scale
2
Descriptive Assessment
2
Type of intervention
Reinforcement only
20
Extinction and Reinforcement or Punishment
16
Other/Not reported
16
Reinforcement and Punishment
6
Punishment only
2
Extinction only
2
Total N
62
30
24.53
Table 3.3 continued
__________________________________________________________________
Characteristic
n
%
__________________________________________________________________
Behavioral Therapist
Not reported or insufficient information
42
65.6
Other Professional
9
14.1
Teacher/Special Education teacher
9
14.1
Student
2
3.2
Parent
1
1.5
Paraprofessional
1
1.5
Inpatient Hospital
15
23.4
Other
13
20.3
Public School
12
18.7
Not Reported
8
12.5
Outpatient Clinic
6
9.4
Inpatient Classroom
5
7.8
Home
4
6.3
Residential Treatment Setting
1
1.6
Treatment Setting
31
Table 3.3 continued
__________________________________________________________________
Characteristic
n
%
__________________________________________________________________
Experimental characteristics
Experimental design
Simple AB
24
22.6
Reversal
21
19.8
Alternating Treatments
16
15.1
Multiple Baseline
2
1.9
Not reported
1
.9
No
43
67.2
Yes
16
25.0
5
7.8
No
36
56.3
Yes
28
43.7
14
13.2
Combination
8
7.5
New people only
5
4.7
Other
1
.9
Follow-up data collected
Not Reported
Follow-up interval (in months) ( M = 8.65, SD = 6.61)
Attempt to generalize behavior
Attempt to generalize to:
New situation or setting only
32
Table 3.3 continued
__________________________________________________________________
M
SD
Characteristic
Range
__________________________________________________________________
Reliability of observations
Inter-rater reliability
FA
84.0-100.0
94.6
3.7
Treatment
87.0-100
95.3
3.3
______________________________________________________________________________
Note. n = number of studies; % = percentage of studies; M = mean; SD = standard deviation.
33
Table 3.4
Descriptive Statistics for Three Effect Sizes
M
SD
Min
Max
MBLR
83.89
22.43
-.96
100.00
PND
81.00
31.23
.00
100.00
PZD
62.55
31.87
.00
100.00
MBLR
77.21
16.63
45.33
100.00
PND
89.02
23.73
10.33
100.00
EFA
BFA
PZD
35.61
42.86
.00
100.00
______________________________________________________________________________
Note. EFA = experimental functional analysis; BFA = behavioral functional assessment; MBLR
= mean baseline reduction; PND = percentage of non-overlapping data; PZD = percentage of
zero data; M = mean; SD = standard deviation; Min = minimum value; Max = maximum value.
Descriptive statistics are presented for 62 treatment outcomes.
34
Table 3.5.
Descriptives for Assessment Type and Function
Function
ATT
ESC
TAN
AUT
UND
COM
Assessment Type
EFA
9
14
9
16
19
13
BFA
4
9
2
4
1
6
Total
13
23
11
20
20
19
______________________________________________________________________________
Note. EFA = experimental functional analysis; BFA = behavioral functional assessment; ATT =
attention; ESC = escape; TAN = tangible; AUT = automatic; UND = undifferentiated; COM =
combination.
35
Table 3.6.
Treatment Effect Sizes by Function
Effect Size
M
SD
Min
Max
MBLR
88.77
13.31
60.22
98.81
PND
96.88
8.84
75.00
100.00
PZD
52.66
43.36
.00
100.00
MBLR
86.27
13.90
46.72
100.00
PND
86.78
29.06
.00
100.00
PZD
56.86
39.51
.00
100.00
MBLR
69.40
33.28
33.97
100.00
PND
94.05
10.31
82.14
100.00
PZD
80.30
34.12
40.91
100.00
MBLR
75.20
28.01
-.96
100.00
PND
87.57
27.56
.00
100.00
PZD
41.55
36.75
.00
100.00
Function
Attention
Escape
Tangible
Automatic
36
Table 3.6 continued
Treatment Effect Sizes by Function
M
SD
Min
Max
MBLR
76.27
27.07
39.03
97.34
PND
72.29
32.31
39.14
100.00
PZD
40.24
30.18
.00
66.50
MBLR
82.49
20.80
31.47
100.00
PND
66.67
36.21
.00
100.00
Undifferentiated
Combination
PZD
63.39
33.56
.00
100.00
______________________________________________________________________________
Note. M = mean; SD = standard deviation; Min = minimum value; Max = maximum value;
MBLR = mean baseline reduction; PND = percentage on non-overlapping data; PZD =
percentage of zero data.
37
Table 3.7.
χ2 for Function by Behavior Type
Function
χ2
df
Asymp. Sig
Attention
8.00
2
.018*
Escape
5.57
2
.062
Tangible
7.36
1
.007*
19.36
2
.000*
6.70
2
.035*
Automatic
Undifferentiated
Combination
1.37
2
.504
______________________________________________________________________________
Note. df = degrees of freedom; * = p < .05
38
Table 3.8.
χ2 for Behavior Type by Function
Behavior Type
χ2
df
Internal
13.16
4
.011*
External
10.55
5
.061
Asymp. Sig
Internal/External
12.90
5
.024*
______________________________________________________________________________
Note. df = degrees of freedom; * = p < .05
39
CHAPTER 4: DISCUSSION
Summary of Findings
In light of the debate regarding the relative benefits of EFA and BFA methodology, the
present review focused on a series on questions. First, is treatment effectiveness influenced by
the type of FA? Second, do different FAs lead to different treatment selections? Third, does FA
type affect the behavioral function identified? Fourth, does observed function impact the
effectiveness of treatment? Lastly, is there a relationship between type of behavior (e.g.,
internal, external) and the identified function? The main findings of the quantitative research
synthesis are summarized as follows. Treatment based on the results of both EFAs and BFAs
were compared and EFA-based treatments were found to be more effective than BFA-based
treatments when the PZD statistic was used to determine effectiveness. This finding indicates
that treatments based on EFAs are more successful at suppressing problematic behaviors when
compared to those based on BFAs. Behavioral function was not found to be a significant
influence in the type of treatment selected. Reinforcement alone and extinction plus
reinforcement or extinction were the most commonly reported treatments.
The type of FA used was not related to the outcome of the assessment. The finding
indicates that neither EFAs nor BFAs result in predictable outcomes because identified
behavioral functions were equally distributed based on type of assessment. The functions of
target behavior, in and of themselves, have no significant impact on treatment effectiveness.
Although previous research (e.g., Vollmer, 1994; Piazza, Hanley, & Fisher, 1996), has indicated
that behaviors maintained by some functions are more difficult to treat, the results of this study
40
do not support that notion. A significant relationship was observed between the type of problem
behavior and the identified function of that behavior. Internal behaviors were more often
identified as having automatic functions while external behaviors were more often identified as
having escape and tangible functions. When both internal and external behaviors were targeted,
the most common result of FAs was a combination of two functions.
Implications for clinicians
The results of the quantitative review are relevant to practicing clinicians regarding the
assessment and treatment of severe problem behaviors exhibited by individuals within the autism
population. The data shows that there is a significant relationship between behavioral function
and behavioral topography (see Tables 3.7 and 3.8). The relationship may help to identify the
function and later possible treatment options for maladaptive behaviors. For example, if an
external behavior (e.g., aggression) is targeted it is more likely that the purpose of that behavior
is to gain access to tangible items or to gain escape from presented activities. Initial treatment
selection could be based on this hypothesis.
When the goal of treatment is suppression of behavior, our findings suggest that
treatments based on the outcome of EFAs are more effective than those based on BFAs. For
severe problem behaviors that are high in intensity, severity, and frequency or those that lead to
tissue damage or other life threatening injury, complete suppression of the targeted behavior is
even more important. The treatments based on the functions identified by both EFAs and BFAs
were shown to be effective in reducing the rate of problem behavior; however, the results of
EFAs were found to be significantly more useful when the treatments were assessed by the PZD
statistic.
41
Limitations of the current study
Conclusions of the review must be considered within the context of its limitations. The
main limitation of this research synthesis is the exclusion of unpublished studies, including
unpublished theses and dissertations. It is possible that the studies included represent a skewed
portion of the population and are not representative of the whole. For example, studies that
report poor treatment effectiveness may go unpublished and thus the average effect sizes
reported within this review represent overestimates. Also, FAs that have undifferentiated results
and are not further assessed may not be published and therefore not included in the current
dataset. Finally, it is possible that published articles that met inclusion criteria may have been
unintentionally excluded and not included in the review.
Another limitation of the current study rests within the primary articles used. Many
articles did not include potentially useful information about the characteristics of the participants.
Basic demographic data such as race, age, and level of mental retardation were often not reported
in the primary articles. Location of assessment and treatment sessions was also excluded from
most articles. The lack of information provided and the possible effects of this exclusion have
been reported by others. However encouraged, many researchers are still excluding important
information about participants and methodological design from their studies. Also, data common
to multiple investigations may have unintentionally been coded more than once in a quantitative
synthesis such as this if not noted by the primary article author. In some cases, articles did not
meet inclusion criteria because a diagnosis of autism or claim that participant was “autistic-like”
was not explicitly stated. The lack of information presented could affect not only the results of
the analyses but also attempts to generalize the findings.
42
Subgroups of both assessment and treatment types were combined throughout the
analyses. For example, FAs reported as “modified” and “brief” analogue sessions were included
under the EFA category, along with traditional EFAs described by Iwata et al. Categories were
combined in order to assess the effectiveness of experimental versus non-experimental
assessments rather than specific subtypes of assessment. Coding intervention groups into six
categories, including three groups comprised of multiple components, may not capture the
differences between specific types of treatment (e.g., verbal reinforcement, tangible
reinforcement).
Treatment effectiveness was summarized by examining the first baseline and last
treatment phase reported in the primary studies. The choice to use these phases was necessary
for legitimate comparison of non-regression-based effect sizes. For example, in some studies,
several different treatments were assessed and reported in an ABCD design. In this case, the rate
of behavior reported in phase D was compared to the baseline data reported in phase A to
determine effectiveness of treatment. However, this choice resulted in a loss of information
available in published reports that may have altered effect sizes in unknown ways.
Directions for future research
Future quantitative reviews examining functional assessment and comparing different
methodologies should include unpublished studies such as theses and dissertations. This would
increase the sample size of the included articles. Also, including studies from a larger time span
(i.e., greater than five years) would increase sample size and possibly improve the statistical
analyses. Bearing in mind that one commonly reported negative aspect of the EFA is the length
of time needed to complete an assessment, an evaluation of “brief” EFAs and BFAs may be a
more appropriate comparison. Although the current study included only the assessment and
43
treatment of individuals with autism, many other clinical populations engage in severe problem
behaviors. Articles that report problem behavior assessment and intervention for children
diagnosed with mental retardation and individuals without a formal diagnosis could be included
in future reviews.
44
REFERENCES
Allison, D.B., Faith, M.S., & Franklin, R.D. (1995). Antecedent exercise in the treatment of
disruptive behavior: A meta-analytic review. Clinical Psychology: Science and Practice,
2, 279-303.
Allison, D.B. & Gorman, B.S. (1993). Calculating effect sizes for meta-analysis: The case of the
single case. Behaviour Research and Therapy, 31, 621-631.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental
disorders: Fourth Edition Text Revision. Washington, DC: American Psychiatric
Association.
Applegate, H., Matson, J.L., & Cherry, K.E. (1999). An evaluation of functional variables
affecting severe problem behaviors in adults with mental retardation by using the
Questions about Behavioral Function scale (QABF). Research in Developmental
Disabilities, 20, 229-237.
Axelrod, S. (1987). Functional and structural analyses of behavior: Approaches leading to
reduced use of punishment procedures? Research in Developmental Disabilities, 8, 165178.
Baranek, G.T. (2002). Efficacy of sensory and motor interventions for children with autism.
Journal of Autism and Developmental Disorders, 32, 397-422.
Bartak, L., & Rutter, M. (1973). Special educational treatment of autistic children: A
comparative study. Journal of Child Psychology and Psychiatry, 14, 151-179.
Bettelheim, B. (1967). The Empty Fortress. New York: Free Press.
Bryson, S.E., Rogers, S.J., & Fombonne, E. (2003). Autism spectrum disorders: Early detection,
intervention, education, and psychopharmacological management. The Canadian Journal
45
of Psychiatry, 48, 506-516.
Campbell, J.M. (2003). Efficacy of behavioral interventions for reducing problem behavior in
persons with autism: A quantitative synthesis of single-subject research. Research in
Developmental Disabilities, 24, 120-138.
Campbell, J.M. (2004). Statistical comparison of four effect sizes for single-subject designs.
Behavior Modification, 28, 234-246.
Carr E.G., (1977). The motivation of self-injurious behavior: A review of some hypotheses.
Psychological Bulletin, 84, 800-816.
Carr, E.G., Langdon, N.A., & Yarborough, S.C., (1999). Hypothesis-based intervention for
severe problem behavior. In A.C. Repp and R.H. Horner (Eds), Functional Analysis of
Problem Behavior. Belmont, CA: Wadsworth Publishing Company, (9-31).
Charlop, M.H., Schreibman, L., & Kurtz, P.F. (1990). Childhood autism. In T.R. Kratchowill
and R.J. Morris (Eds.), The Practice of Child Therapy. Needham Heights, MA: Allyn and
Bacon, (257-297).
Dawson, G., & Wattling, R. (2000). Interventions to facilitate auditory, visual, and motor
integration in autism: A review of the evidence. Journal of Autism and Developmental
Disorders, 30, 415-421.
Dawson, J.E., Matson, J.L., & Cherry, K.E. (1998). An analysis of maladaptive behaviors in
persons with autism, PDD-NOS, and mental retardation. Research in Developmental
Disabilities, 19, 439-448.
Derby, K.M., Wacker, D.P., Sasso, G., Steege, M., Northup, J., Cigrand, K., & Asmus, J. (1992).
Brief functional assessment techniques to evaluate aberrant behavior in an outpatient
setting: A summary of 79 cases. Journal of Applied Behavior Analysis, 25, 713-721.
46
Didden, R., Duler, P.C., &Korzilius, H. (1997). Meta-analytic study on treatment effectivenss for
problem behaviors with individuals that have mental retardation. American Journal on
Mental Retardation, 101, 387-399.
Durand, V.M. & Crimmins, D.B. (1988). Identifying the variables maintaining self-injurious
behavior. Journal of Autism and Developmental Disorders, 18, 99-117.
Durand, V.M., & Crimmins, D.B. (1992). The Motivation Assessment Scale (MAS)
administration guide. Topeka, KS: Monaco and Associates.
Emerson, E., Kiernan, C., Alborz, A., Reeves, D., Mason, H., Swarbrick, R., Mason, L., &
Hatton, C. (2001). The prevalence of challenging behaviors: A total population study.
Research in Developmental Disabilities, 22, 77-93.
Faith, M.S., Allison, D.B., & Gorman, B.S. (1996). Meta-analysis of single-case research. In
R.D. Franklin, Allison, D.B., & Gorman, B.S. (Eds.), Design and analysis of single-case
research (pp. 245-277). Hillsdale, NJ: Lawrence Erlbaum.
Fisher, A.G., Murray, E.A., & Bundy, A.C. (1991). Sensory integration: Theory and practice.
Philadelphia, PA: F.A. Davis.
Gresham, F.M., McIntyre, L.L., Olson-Tinker, H., Dolstra, L., McLaughlin, V., & Van, M.
(2004). Relevance of functional behavioral assessment research for school-based
interventions and positive behavioral support. Research in Developmental Disabilities,
25, 19-37.
Gringas, P. (2000). Practical paediatric psychopharmacological prescribing in autism: The
potential and the pitfalls. Autism, 4, 229-247.
Hatch-Rasmussen, C. (1995). Sensory Integration. Center for the Study of Autism at
www.autism.org/si.html
47
Horner, R.H. (1994). Functional assessment: Contributions and future directions. Journal of
Applied Behavior Analysis, 27, 401-404.
Horner, R.H., Carr, E.G., Strain, P.S., Todd, A.W. &Reed, H.K. (2002). Problem behavior
interventions for young children with autism: A research synthesis. Journal of Autism
and Developmental Disorders, 32, 423-446.
Horner, R.H., & Carr, E.G. (1997). Behavioral support for students with severe disabilities:
Functional assessment and comprehensive intervention. Journal of Special Education, 31,
84-104.
Iwata, B.A., Dorsey, M.F., Slifer, K.J., Bauman, K.E. & Richman, G.S. (1982). Toward a
functional analysis of self-injury. Analysis and Intervention in Developmental
Disabilities, 2, 3-20.
Iwata, B.A., Pace, G.M., Dorsey, M.F., Zarcone, J.R., Vollmer, T.R., Smith, R.G., Rodgers,
T.A., Lerman, D.C., Shore, D.A., Mazaleski, J.L., Goh, H., Cowdery, G.E., Kalsher,
M.J., McCosh, K.C., & Willis, K.D. (1994). The functional analysis of self-injurious
behavior: An experimental-epidemiological analysis. Journal of Applied Behavior
Analysis, 27, 215-240.
Johnson, W.L., & Baumeister, A. (1978). Self-injurious behavior: A review and analysis of
methodological details of published studies. Behavior Modification, 2, 465-484.
Kahng, S., Iwata, B.A., & Lewin, A.B. (2002). Behavioral treatment of self-injury, 1964-2000.
American Journal on Mental Retardation, 107, 212-221.
Kahng, S., Iwata, B.A., & Lewin, A.B. (2002). The impact of functional assessment on the
48
treatment of self-injurious behavior. In S.R. Schroeder, M.L. Oster-Granite, &
T.Thompson (Eds.), Self-injurious Behavior: Gene-brain-behavior Relationships (p. 119131). Washington, DC: American Psychological Association.
Koegel, R.L., Koegel, L.K., & McNerney, E.K. (2001). Pivotal Areas in Intervention for Autism.
Journal of Clinical Child Psychology, 30, 19-32.
Koomar, J.A., & Bundy, A.C. (1991). The art and science of creating direct intervention from
theory. In A.G. Fisher, E.A. Murray, & A.C. Bundy (Eds.), Sensory Integration Theory
and Practice (p.252). Philadelphia: F.A. Davis Company.
Levitt, E.E. (1963). Psychotherapy with children: A further evaluation. Behaviour Research and
Therapy, 1, 45-51.
Lundervold, D., & Bourland, G. (1988). Quantitative analysis of treatment of aggression, selfinjury, and property destruction. Behavior Modification, 12, 590-617.
Martin, N.T., Gaffan, E.A., & Williams, T. (1999). Experimental functional analyses for
challenging behavior: a study of validity and reliability. Research in Developmental
Disabilities, 20, 125-146.
Mason, S.A., & Iwata, B.A. (1990). Artifactual effects of sensory-integration therapy on self –
injurious behavior. Journal of Applied Behavioral Analysis, 23, 361-370.
Matson, J.L., & Vollmer, T.R. (1995). User’s Guide: Questions About Behavioral Function
(QABF). Baton Rouge, LA: Scientific Publishers, Inc.
Matson, J.L., Bamburg, J.W., Cherry, K.E., & Paclawskyj, T.R. (1999). A validity study on the
Questions about Behavioral Function (QABF) scale: predicting treatment success for
self-injury, aggression, and stereotypies. Research in Developmental Disabilities, 20,
163-176.
49
Northup, J., Wacker, D., Sasso, G., Steege, M., Cigrand, K., Cook, J., & Deraad, A. (1991). A
brief functional analysis of aggressive and alternative behavior in an outclinic setting.
Journal of Applied Behavior Analysis, 24, 509-522.
O’Brien, S., & Repp, A.C., (1990). Reinforcement-based reductive procedures: A review of 20
years of their use with persons with severe to profound mental retardation. Journal of the
Association for Persons with Severe Handicaps, 15,148-159.
O’Neill, R.E., Horner, R.H., Albin, R.W., Storey, K., Sprague, J.R., & Newton, J.S. (1997).
Functional assessment of problem behavior: A practical assessment guide. Pacific Grove,
CA: Brooks/Cole.
Pelios, L., Morren, J., Tesch, D. & Axelrod, S. (1999). The impact of functional analysis
methodology on treatment choice for self-injurious and aggressive behavior. Journal of
Applied Behavioral Analysis, 32, 185-195.
Piazza, C.C., Hanley, G.P., & Fisher, W.W. (1996). Functional analysis of cigarette pica. Journal
of Applied Behavioral Analysis, 29, 437-439.
Rodriguez, L.S. (1999). Psychoanalysis with children: History, theory and practice. Free
Association Press.
Schultz, R.T., & Anderson, G.M. (2004). The neurobiology of autism and the pervasive
developmental disorderes. In D.S. Charmey & E.J. Nestler (Eds.) Neurobiology of Mental
Illness 2nd edition. Oxford University Press.
Scotti, J. R., Evans, I. M., Meyer, L. H., & Walker, P. (1991). A meta-analysis of intervention
research with problem behavior: Treatment validity and standards of practice. American
Journal on Mental Retardation, 96, 233-256
Scruggs, T.E., Mastropieri, M.A., Cook, S.B., & Escobar, C. (1986). Early intervention for
50
children with conduct disorders: A quantitative synthesis of single-subject research.
Behavioral Disorders, 11, 260-271.
Scruggs, T.E., Mastropieri, M.A., & Casto, G. (1987). The quantitative synthesis of singlesubject research: Issues and applications. Behavior Modification, 22, 221-242.
Skinner, B.F. (1953). Science and Human Behavior. New York: Free Press.
Sturmey, P. (1995). Analog baselines: A critical review of the methodology. Research in
Developmental Disabilities, 16, 269-284.
Sugai, G., Horner, R.H., & Sprague, J.R. (1999). Functional assessment-based behavior support
planning: Research to practice to research. Behavioral Disorders, 24, 223-227.
Toogood S., & Timlin, K., (1996). The functional assessment of challenging behavior: A
comparison of informant-based, experimental and descriptive methods. Journal of
Applied Research in Intellectual Disabilities, 9, 206-222.
Volkmar, F.R., (2001). Pharmacological Interventions in Autism: Theoretical and Practical
Issues. Journal of Clinical Child Psychology, 30, 80-87.
Vollmer, T.R. (1994). The concept of automatic reinforcement: implications for behavioral
research in developmental disabilities. Research in Developmental Disabilities, 15, 187207.
Welch, M.G., 1987. Toward prevention of a developmental disorder. Pennsylvania Medicine, 90,
47-52.
Wieseler, N.A., Hanzel, T.E., Chamberlain, T.P., & Thompson, T. (1985). Functional taxonomy
of stereotypic and self-injurious behavior. Mental Retardation, 23, 230-234.
51
Wicks-Nelson, R., & Israel, A. C. (1999). Autism and schizophrenia. In Behavior
Disorders of Early Childhood (pp. 330-363). Upper Saddle River, N.J.: Prentice
Hall.
52
APPENDIX A
Coding Form used in Meta-analysis
Study ID#: _____
Article ID#: _____
Subject ID#: _____
Authors: _____________________________________________________________________________
Title: _______________________________________________________________________________
______________________________________________________________________________
Published? _____ Yes (1)
_____ No (2)
Publication Year: __________
Journal: ______________________________________________________________________________
Number of subjects in study that meet criteria: _____
Participant Information
Age (at intervention):
_____ Years, _____ Months
Gender:
_____ Male (1) _____ Female (2)
Race:
_____ Caucasian (1)
_____ Black/African American (2)
_____ Other (4)
_____ Not reported (9)
_____ Hispanic/Latino (3)
Enter IQ/Adaptive behavior Score beside test given:
_____ WISC Full Scale (edition:__________) (01)
_____ WISC Verbal (edition:__________) (02)
_____ WISC Performance (edition:__________) (03)
_____ Bayley Mental Developmental Index (04)
_____ Leiter International (05)
_____ Merrill-Palmer Scales (06)
_____ Cattell Infant Intelligence (07)
_____ McCarthy Scales (08)
_____ Stanford-Binet (edition:__________) (09)
_____ Vineland Scales (10)
_____ Other:_______________________________________________________________ (88)
_____ Not reported (99)
53
Level of Mental Retardation (based on test data above or desribed in report):
_____ None (1)
_____ Mild (70-55) (2)
_____ Moderate (54-40) (3)
_____ Severe/Profound (39 and below) (4)
_____ Untestable (5)
_____ Other:________________________________________________________________(8)
_____ Not reported (9)
Level of verbal communication ability:
_____ Average language skills (1)
_____ Minimally verbal; some functional language (2)
_____ Nonverbal; mute (3)
_____ Echolalic (4)
_____ Other: ________________________________________________________________(8)
_____ Not reported (9)
Secondary diagnoses:
_____________________________________________
_____________________________________________
_____________________________________________
Diagnostic criteria used:
_____ National society for Autistic Children (NSAC; 1978) (1)
_____ Rutter’s (1978) Criteria (2)
_____ DSM-III
(3)
_____ DSM-IIIR (4)
_____ DSM-IV
(5)
_____ Subject described as “autistic”, “autistic-like”, or having “autistic features” (6)
_____ Other: ________________________________________________________________(8)
_____ Not reported (9)
Time since autism diagnosis (in months):
__________
_____ Not reported (999)
Time in prior treatment (in months):
__________
_____Not reported (999)
54
Intervention Information
Target behavior
As defined in report: ___________________________________________________________________
_____ Aggression (1)
_____ Self-injurious Behavior (4)
_____ Property destruction (2)
_____ Self-stimulation/stereotypies (5)
_____ Disruptive behavior (3)
_____ Combination:_________________________________________________________(10)
_____ Other: ________________________________________________________________(8)
_____ Not reported (9)
*If SIB, what type?
_____ Head banging (1)
_____ Face slapping (3)
_____ Eye gouging (2)
_____ Biting (4)
_____ Combination:_________________________________________________(10)
_____ Other: ________________________________________________________(8)
Functional assessment conducted?:
_____ Yes (1)
_____ No (2)
_____ Other: ________________________________________________________________(8)
*If yes, complete below:
_____ Experimental Functional Analysis (EFA) (1)
_____ Partial experimental design (2)
_____ Antecedent-Behavior-Consequence sheet (ABC) (3)
_____ Rating Scale (e.g., MAS, QABF) (4)
_____ Informal assessment (5)
_____ Brief EFA (6)
_____ Modified EFA (7)
_____ Other: _______________________________________________________(8)
Function of behavior?:
_____ Attention (1)
_____ Escape (2)
_____ Access to tangibles (3)
55
_____ Communication (4)
_____ Self stimulation /Automatic (5)
_____ Undifferentiated (6)
_____ Combination: _________________________________________________________(10)
_____ Other: _______________________________________________________________(8)
_____ Not reported (9)
Type of intervention used:
_____ Reinforcement only (01)
_____ DRI only (05)
_____ Timeout only (02)
_____ Punishment only (06)
_____ Extinction (03)
_____ Overcorrection (07)
_____ Sensory extinction (04)
_____ Antecedent exercise (08)
_____ Combination: _________________________________________________________(10)
_____ Other: _______________________________________________________________(88)
_____ Not reported (99)
*If reinforcement used in intervention, complete below:
Type:
_____ Food (1)
_____ Token/Money (2)
_____ Sensory stimulation (3)
_____ Social (4)
_____ Toy; play time (5)
_____ Stereotypies (6)
_____ Combination: ___________________________________________(7)
_____ Other: _________________________________________________(8)
_____ Not reported (9)
Schedule:
_____ Continuous (1)
_____ Fixed ratio (every _____nth response) (2)
_____ Variable ratio (every _____nth response on average) (3)
_____ Other: _________________________________________________(8)
_____ Not reported (9)
*If punishment used in intervention, complete below:
_____ Slap/spank (01)
_____ Social disapproval (e.g. “No!”) (05)
_____ Shock (02)
_____ Water mist (06)
56
_____ Restraint (03)
_____ Physical exercise (07)
_____ Visual screening (04)
_____ Ammonia (08)
_____ Combination: __________________________________________(10)
_____ Other: ________________________________________________(88)
_____ Not reported (99)
Duration (in seconds): __________
_____ Not reported (9999)
Length of exposure to intervention / sessions (in minutes):
__________
_____ Not reported (999)
Medication administered during intervention?:
_____ Yes (1)
_____ No (2)
_____ Not specified; not reported (9)
Treatment setting:
_____ Home (01)
_____ Public School (05)
_____ Laboratory (02)
_____ Inpatient classroom (06)
_____ Inpatient ward (03)
_____ Outpatient clinic (07)
_____ Residential treatment setting (04)
_____ Center for Dev. Disorders (08)
_____ Combination: _________________________________________________________(10)
_____ Other: _______________________________________________________________(88)
_____ Not reported (99)
Dimensions of treatment setting (specify feet, meters, yards, etc):
__________
by
__________
(in__________)
_____ Not reported (99)
Behavioral therapist:
_____ Parent (01)
_____ Paraprofessional (02)
_____ Student research assistants (03)
_____ Teacher (04)
_____ Special Education Teacher (05)
_____ Psychologist (06)
_____ Peers (07)
_____ Self (08)
_____ Other professional (09)
_____ Combination: _________________________________________________________(10)
57
_____ Other: _______________________________________________________________(88)
_____ Not reported (99)
Parental involvement in treatment:
_____ Yes, as therapist (1)
_____ Yes, in other capacity (2)
_____ No (3)
_____ Not reported; unclear (9)
Type of experimental design:
_____ Simple A-B (1)
_____ Multiple baseline (Behavior, Setting, Subject) (2)
_____ Alternating treatments (e.g., A-B-A-C) (3)
_____ Reversal (e.g., A-B-A) (4)
_____ Randomization (5)
_____ Combination: _________________________________________________________(10)
_____ Other design: _________________________________________________________(8)
_____ Not reported (99)
Number of data points in first baseline phase: __________
Number of data points in final treatment phase: __________
Effort to generalize?:
_____ Yes
_____ No
_____ Other: ________________________________________________________________(8)
*If yes, to:
_____ New situation; new setting; new context (1)
_____ New people (2)
_____ New behavior (3)
_____ New object (4)
_____ Combination: _________________________________________________(10)
_____ Other: _______________________________________________________(8)
_____ Not reported (99)
Follow up data collected?:
_____ Yes (1) Interval after treatment, in months: __________
_____ No (2)
_____ Unclear (9)
58
Experimental Quality
Scale used to measure target behavior:
_____ Frequency count (length of interval: __________) (1)
_____ Percentage (length of interval: __________) (2)
_____ Rating on an interval scale (length of interval: __________) (3)
_____ Other: ________________________________________________________________(8)
_____ Not reported (9)
Units of measurement on X axis:
_____ Days (1)
_____ Sessions (4)
_____ Trials (2)
_____ Weeks (5)
_____ Months (3)
_____ Other: _________________(8)
_____ Not reported (9)
For functional assessment:
Inter-rater reliability for target behavior:
__________
_____ None reported (999)
Type of interobserver agreement:
_____ Kappa (1)
_____ Phi coefficient (2)
_____ Percent agreement (e.g. agree / agree + disagree x 100 ) (3)
_____ Other: ________________________________________________________________(8)
_____ Not reported (9)
Interobserver agreement based on:
_____ Partial interval recording system (specify in seconds: __________) (1)
_____ Other: ________________________________________________________________(8)
_____ Not reported (9)
For treatment:
Inter-rater reliability for target behavior:
__________
_____ None reported (999)
Type of interobserver agreement:
_____ Kappa (1)
_____ Phi coefficient (2)
_____ Percent agreement (e.g. agree / agree + disagree x 100 ) (3)
59
_____ Other: ________________________________________________________________(8)
_____ Not reported (9)
Interobserver agreement based on:
_____ Partial interval recording system (specify in seconds: __________) (1)
_____ Other: ________________________________________________________________(8)
_____ Not reported (9)
Source of observations:
_____ Parent (01)
_____ Psychologist (05)
_____ Paraprofessional (02)
_____ Peers (06)
_____ Student research assistants (03)
_____ Self (07)
_____ Teacher (04)
_____ Combination: _________________________________________________________(10)
_____ Other: _______________________________________________________________(88)
_____ Not reported (99)
Effect Size
Mean % reduction from baseline: _____
PND (using Didden et al.’s (1997) modification:
_____ 1st baseline versus 1st treatment
_____ 2nd baseline versus 2nd treatment
_____ 1st baseline versus 2nd treatment
_____ Total (average of 1st baseline/1st treatment and 2nd baseline/2nd treatment)
PZD:
_____ 1st treatment phase
_____ 2nd treatment phase
_____ Total (average of 1st and 2nd treatment phases)
60
(NOTE: Data points equal distance from center of data point to X-axis, rounded to closest 0.5 mm IF Xaxis is labeled zero OR if no clearly defined data points exists. If a clear zero data point lies above the Xaxis, draw line perpendicular to Y-axis through midpoint of those data points clearly designated as zero in
text of manuscript.)
Phase
1._____
Phase
1._____
Phase
1._____
Phase
1._____
2._____
2._____
2._____
2._____
3._____
3._____
3._____
3._____
4._____
4._____
4._____
4._____
5._____
5._____
5._____
5._____
6._____
6._____
6._____
6._____
7._____
7._____
7._____
7._____
8._____
8._____
8._____
8._____
9._____
9._____
9._____
9._____
10._____
10._____
10._____
10._____
11._____
11._____
11._____
11._____
12._____
12._____
12._____
12._____
13._____
13._____
13._____
13._____
14._____
14._____
14._____
14._____
15._____
15._____
15._____
15._____
16._____
16._____
16._____
16._____
17._____
17._____
17._____
17._____
18._____
18._____
18._____
18._____
19._____
19._____
19._____
19._____
20._____
20._____
20._____
20._____
61
APPENDIX B
Studies used in Meta-analysis
Adelinis, J.D., Piazza, C.C. &Goh, H.L. (2001). Treatment of multiply controlled destructive
behavior with food reinforcement. Journal of Applied Behavior Analysis, 34. 97-100.
Adelinis, J.D. & Hagopian, L.P. (1999). The use of symmetrical “do” and “don’t” requests to
interrupt ongoing activities. Journal of Applied Behavior Analysis, 32. 519-523.
Anderson, C.M. & Long, E.S. (2002). Use of a structured descriptive assessment methodology to
identify variables affecting problem behavior. Journal of Applied Behavior Analysis, 35,
137-154.
Borrero, J.C. & Vollmer, T.R. (2002). An application of the matching law to severe problem
behavior. Journal of Applied Behavior Analysis, 35. 13-27.
Braithwaite, K.L. & Richdale, A.L. (2000). Functional communication training to replace
challenging behaviors across two behavioral outcomes. Behavioral Interventions, 15, 2136.
Brown, K.A., Wacker, D.P., Derby, K.M., Peck, S. M., Richman, D.M., Sasso, G.M., Knutson,
C.L. & Harding, J.W. (2000). Evaluating the effects of functional communication
training in the presence and absence of establishing operations. Journal of Applied
Behavior Analysis, 33. 53-71.
Carr, J.C., Dozier, C.L., Patel, M.R., Adams, A. N. & Martin, N. (2002). Treatment of
automatically reinforced object mouthing with noncontingent reinforcement and response
blocking: Experimental analysis and social validation. Research in Developmental
Disabilities, 23. 37-44.
Coleman, C.L. & Holmes, P. A. (1998). The use of noncontingent escape to reduce the disruptive
62
behaviors in children with speech delays. Journal of Applied Behavior Analysis, 31. 687690.
DeLeon, I.G., Anders, B.M., Rodriguez-Catter, V. & Neidert, P.L. (2000). The effects of
noncontingent access to single versus multiple stimulus sets on self-injurious behavior.
Journal of Applied Behavior Analysis, 33. 623-626.
DeLeon, I.G., Neidert, P.L., Anders, B.M. & Rodriguez-Carter, V. (2001). Choices between
positive and negative reinforcement during treatment for escape-maintained behavior.
Journal of Applied Behavior Analysis, 34. 521-525.
Durand, V.M. (1999). Functional communication training using assistive devices: Recruiting
natural communities of reinforcement. Journal of Applied Behavior Analysis, 32. 247267.
Feldman, M.A., Condillac, R.A., Tough, S., Hunt, S. & Griffiths, D. (2002). Effectiveness of
community positive behavioral intervention for persons with developmental disabilities
and severe behavior disorders. Behavior Therapy, 33. 377-398.
Fisher, W.W., Kuhn, D.E. & Thompson, R.H. (1998). Establishing discriminative control of
responding using functional and alternative reinforcers during functional communication
training. Journal of Applied Behavior Analysis, 31. 543-560.
Fisher, W.W., Lindauer, S.E., Alterson, C.J. & Thompson, R.H. (1998). Assessment and
treatment of destructive behavior maintained by stereotypic object manipulation. Journal
of Applied Behavior Analysis, 31. 513-527.
Galiatsatos, G.T. & Graff, R.B. (2003). Combining descriptive and functional analyses to assess
and treat screaming. Behavioral Interventions, 18. 123-138.
Graff, R.B., Lineman, G.T., Libby, M.E. & Ahearn, W.H. (1999). Functional analysis and
63
treatment of screaming in a young girl with severe disabilities. Behavioral Interventions,
14. 233-239.
Hagopian, L.P., Crockett, J.L., van Stone, M., DeLeon, I.G. &Bowman, L.G. (2000). Effects of
noncontingent reinforcement on problem behavior and stimulus engagement: The role of
satiation, extinction, and alternative reinforcement. Journal of Applied Behavior Analysis,
33. 433-449.
Hagopian, L.P., Wilson, D.M. & Wilder, D.A. (2001). Assessment and treatment of problem
behavior maintained by escape from attention and access to tangible items. Journal of
Applied Behavior Analysis, 34. 229-232.
Hanley, G.P., Iwata, B.A., Thompson, R.H. & Lindberg, J.S. (2000). A component analysis of
“stereotypy as reinforcement” for alternative behavior. Journal of Applied Behavior
Analysis, 33. 285-297.
Harding, J.W., Wacker, D.P., Berg, W.K., Cooper, L.J., Asmus, J., Mlela, K. & Muller, J.
(1999). An analysis of choice making in the assessment of young children with severe
behavior problems. Journal of Applied Behavior Analysis, 32. 63-82.
Healey, J.J., Ahearn, W.H., Graff, R.B. &Libby, M.E. (2001). Extended analysis and treatment
of self-injurious behavior. Behavioral Interventions, 16. 181-195.
Hoch, H., McComas, J.J., Thompson, A.L. & Paone, D. (2002). Concurrent reinforcement
schedules: Behavior change and maintenance without extinction. Journal of Applied
Behavior Analysis, 35.155-169.
Kelley, M.E., Lerman, D.C. & Van Camp,C.M. (2002). The effects of competing reinforcement
schedules on the acquisition of functional communication. Journal of Applied Behavior
Analysis, 35. 59-63.
64
Kennedy, C.H., Meyer, K.A., Knowles, T. & Shukla, S. (2000). Analyzing the multiple functions
of stereotypical behavior for students with autism: Implications for assessment and
treatment. Journal of Applied Behavior Analysis, 33. 559-571.
Kuhn, D.E., DeLeon, I.G., Fisher, W.W. & Wilke, A.E. (1999). Clarifying an ambiguous
functional analysis with matched and mismatched extinction procedures. Journal of
Applied Behavior Analysis, 32. 99-102.
Lalli, J.S., Vollmer, T.R., Progar, P.R., Wright, C., Borrero, J., Daniel, D., Barthold, C.H.,
Tocco, K. & May, W. (1999). Competition between positive and negative reinforcement
in the treatment of escape behavior. Journal of Applied Behavior Analysis, 32. 285-296.
Lindauer, S.E., Zarcone, J.R., Richman, D.M. & Schroeder, S.R. (2002). A comparison of
multiple reinforcer assessments to identify the function of maladaptive behavior. Journal
of Applied Behavior Analysis, 35. 299-303.
Linscheid, T.R. & Reichenbach, H. (2002). Multiple factors in the long-term effectiveness of
contingent electric shock treatment for self-injurious behavior: A case example. Research
in Developmental Disabilities, 23. 161-177.
Mace, A.B., Shapiro, E.S. & Mace, F.C. (1998). Effects of warning stimuli for reinforcer
withdrawal and task onset on self-injury. Journal of Applied Behavior Analysis, 31. 679682.
Mancina, C., Tankersley, M., Kamps, D., Kravitz, T. & Parrett, J. (2000). Brief Report:
Reducation of inappropriate vocalizations for a child with autism using a selfmanagement treatment program. Journal of Autism and Developmental Disabilities, 30,
599-606.
McComas, J., Hoch, H., Poane, D. & El-Roy, D. (2000). Escape behavior during academic tasks:
65
A preliminary analysis of idiosyncratic establishing operations. Journal of Applied
Behavior Analysis, 33. 479-493.
McCord, B.E., Iwata, B.A., Galensky, T.L., Ellington, S.E. & Thompson, R.J. (2001). Functional
analysis and treatment of problem behavior evoked by noise. Journal of Applied
Behavior Analysis, 34. 447-462.
McKerchar, T.L., Kahng, S., Casioppo, E. &Wilson, D. (2001). Brief report-Functional Analysis
of self-injury maintained by automatic reinforcement: exposing masked social functions.
Behavioral Interventions, 16. 59-63.
Mueller, M.M., Wilcynzynski, S.M., Moore, J.W., Fusilier, I. & Trahant, D. (2001). Antecedent
manipulations in a tangible condition effects of stimulus preference on aggression.
Journal of Applied Behavior Analysis, 34. 237-240.
Mullins, J.L. & Christian, L. (2001). The effects of progressive relaxation training on the
disruptive behavior of a boy with autism. Research in Developmental Disabilities, 22.
449-462.
O’Neill, R.E. & Sweetland-Baker, M. (2001). Brief Report: An assessment of stimulus
generalization and contingency effects in functional communication training with two
students with autism. Journal of Autism and Developmental Disabilities, 31. 235-240.
Patel, M. R., Carr, J.E., Kim, C., Robles, A. & Eastridge, D. (2000). Functional analysis of
aberrant behavior maintained by automatic reinforcement: Assessments of specific
sensory reinforcers. Research in Developmental Disabilities, 21. 393-407.
Piazza, C.P., Bowman, L.G., Contrucci, S.A., Delia, M.D., Adelinis, J.D. & Goh, Han-Leong.
(1999). An evaluation of the properties of attention as a reinforcement for destructive and
appropriate behavior. Journal of Applied Behavior Analysis, 32. 437-449.
66
Piazza, C.P., Fisher, W.W., Hanley, G.P., LeBlanc, L.A., Worsdell, A.S., Lindauer, S.E. &
Keeney, K.M. (1998). Treatment of pica through multiple analyses of its reinforcing
functions. Journal of Applied Behavior Analysis, 31. 165-189.
Piazza, C.P., Roane, H.S., Keeney, K.M., Boney, B.R. & Abt, K.A. (2002). Varying response
effort in the treatment of pica maintained by automatic reinforcement. Journal of Applied
Behavior Analysis, 35. 233-246.
Progar, P.R., North, S.T., Bruce, S.S., DiNovi, B.J., Nau, P.A., Eberman, E.M., Bailey, J.R. &
Nussbaum, C.N. (2001). Putative behavioral history effects and aggression maintained by
escape from therapists. Journal of Applied Behavior Analysis, 34. 69-72.
Rapp, J.T., Dozier, C.L. &Carr, J.E. (2001). Functional assessment and treatment of pica: A
single case experiment. Behavioral Interventions, 16. 111-125.
Rapp, J.T., Dozier, C.L., Carr, J.L., Patel, M.R. & Enloe, K.A. (2000). Functional analysis of
hair manipulation: replication and extension. Behavioral Interventions, 15. 121-133.
Reese, R.M., Sherman, J.A. & Sheldon, J.B. (1998). Reducing disruptive behavior of a group
home resident with autism and mental retardation. Journal of Autism and Developmental
Disabilities, 28. 159-165.
Richman, D.M. & Hagopian, L.P. (1999). On the effects of “quality” of attention in the
functional analysis of destructive behavior. Research in Developmental Disabilities, 20.
51-62.
Richman, D.M., Wacker, D.P., Asmus, J.M. & Casey, S.D. (1998). Functional analysis and
extinction of different behavior problems exhibited by the same individual. Journal of
Applied Behavior Analysis, 31. 475-478.
Richman, D.M., Wacker, D.P., Asmus, J.M. & Casey, S.D. & Andelman, M. (1999). Further
67
analysis of problem behavior in response class hierarchies. Journal of Applied Behavior
Analysis, 32. 269-283.
Ringdahl, J.E. & Sellers, J.A. (2000). The effects of different adults as therapists during
functional analyses. Journal of Applied Behavior Analysis, 33. 247-250.
Roane, H.S., Lerman, D.C., Kelley, M.C. &Van Camp, C.M. (1999). Within-session patterns of
responding during functional analyses: The role of establishing operations in clarifying
behavioral function. Research in Developmental Disabilities, 20. 73-89.
Roane, H.S., Lerman, D.C. & Vorndran, C.M. (2001). Assessing reinforcers under progressive
schedule requirements. Journal of Applied Behavior Analysis, 34. 145-167.
Rush, K.S., Crockett, J.L. & Hagopian, L.P. (2001). An analysis of the selective effects of NCR
with punishment targeting problem behavior associated with positive affect. Behavioral
Interventions, 16. 127-135.
Sigafoos, J. &Tucker, M. (2000). Brief assessment and treatment of multiple challenging
behaviors. Behavioral Interventions, 15. 53-70.
Tang, J.C., Patterson, T.G. & Kennedy, C.H. (2003). Identifying specific sensory modalities
maintaining the stereotypy of students with multiple profound disabilities. Research in
Developmental Disabilities, 24. 433-451.
Tang, J.C., Kennedy, C.H., Koppekin, A. & Caruso, M. (2002). Functional analysis of
stereotypical ear covering in a child with autism. Journal of Applied Behavior Analysis,
35. 95-98.
Tincani, M.J., Castrogiavanni, A. & Axelrod, S. (1999). A comparison of the effectiveness of
brief versus traditional functional analyses. Research in Developmental Disabilities, 20.
327-338.
68
Valvodinos, M. G., Napolitano, D.A., Zarcone, J.R., Hellings, J.A., Williams, D.C. & Schroeder,
S.R. (2002). Multimodal evaluation of risperidone for destructive behavior: Functional
analysis, direct observations, rating scales, and psychiatric impressions. Experimental
and Clinical Psychopharmacology, 10, 268-275.
Van Camp, C.M., Vollmer, T.R. & Daniel, D. (2001). A systematic evaluation of stimulus
preference, response effort, and stimulus control in the treatment of automatically
reinforced self-injury. Behavior Therapy, 32, 603-613.
Vollmer, T., Borrero, J.C., Lalli, J.S. & Daniel, D. (1999). Evaluating self-control and
impulsivity in children with severe behavior disorders. Journal of Applied Behavior
Analysis, 32, 451-466.
69
APPENDIX C
Interrater Reliability Data
Variable
Age
Gender
Race
IQ/Adaptive behavior score
Level of Mental Retardation
Level of verbal communication
Secondary diagnoses
Diagnostic criteria used
Time since autism diagnosis
Time in prior treatment
Target behavior
SIB type
FA conducted?
FA type
Function of behavior
Type of intervention
Type of reinforcement
Reinforcement schedule
Type of punishment
Punishment duration
Length of intervention
Medication administered?
Treatment setting
Dimensions of setting
Behavioral therapist
Parental involvement
Experimental Design
# of data points in first baseline
# of data points in final treatment
Effort to generalize?
Generalization to what?
Follow up data
Scale used to measure target behavior
Units on X axis
FA: interrater reliability
FA: type of agreement
FA: agreement recording system
TX: interrater reliability
TX: type of agreement
TX: agreement recording system
Source of Observations
MBLR
PND
PZD
Mean = 99.06, Range = 82.35 – 100.00
Percent Agreement
100.00
100.00
100.00
100.00
88.24
82.35
100
100.00
100.00
100.00
100.00
100.00
100.00
100.00
94.12
100.00
100.00
100.00
100.00
100.00
100.00
94.12
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
70