Journal of Autism and Developmental Disorders

https://doi.org/10.1007/s10803-024-06241-5

ORIGINAL ARTICLE

Convergent and Discriminant Validity of the Verbal Behavior

Milestones Assessment and Placement Program (VB‑MAPP)
and the Vineland Adaptive Behavior Scales (VABS)
Amin D. Lotfizadeh1 · Brendan Gard4 · Cynthia Rico1 · Alan Poling2 · Kristen R. Choi3

Accepted: 8 January 2024

© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024

Abstract
Behavior analysts frequently use the Verbal Behavior Milestones Assessment and Placement Program (VB-MAPP) to assess
the language and social skills of children with autism in everyday practice and in research. Despite the widespread use of
the VB-MAPP, its psychometric characteristics have not been extensively investigated. To provide information about its
convergent validity, we calculated correlations between scores earned by 235 children with autism on the VB-MAPP and the
Vineland Adaptive Behavior Scale (VABS), a commonly used assessment with good reliability and validity. We obtained
moderate or strong positive correlations between the VB-MAPP Milestones score and VABS Communication, Socialization,
and Daily Living Skills subdomains. There was also a strong positive correlation between the VB-MAPP Milestones score
and VABS overall raw score. These findings suggest that the VB-MAPP Milestones Assessment measures aspects of social
and communicative behavior comparable to those indexed by these VABS subdomains which, like prior findings, supports
the use of the Milestones Assessment. No significant relationship was observed between the VB-MAPP Barriers score and
the VABS Internalizing, Externalizing, or overall Maladaptive Behavior scores. These findings, like prior findings, question
the value of the VB-MAPP’s Barriers Assessment as a measure of maladaptive behavior.

Keywords VB-MAPP · VABS · Convergent validity · Psychometrics · Autism assessment

Autism is characterized by social and communication

deficits with restricted and repetitive behaviors (American
* Amin D. Lotfizadeh Psychiatric Association, 2013). Applied behavior analysis
amin.lotfizadeh@essc.org (ABA) is a general strategy for treating autism that is sup-
Brendan Gard ported by substantial evidence (National Autism Center,
brendan.gard@essc.org 2015). ABA interventions primarily focus on teaching and
Cynthia Rico assessing communication and social skills (e.g., Drash et al.,
cynthia.rico@essc.org 1999; Goldsmith et al., 2007; Ingvarsson & Hollobaugh,
Alan Poling 2010). Applied behavior analysts often use the Verbal
alan.poling@wmich.edu Behavior Milestones Assessment and Placement Program
Kristen R. Choi (VB-MAPP; Sundberg, 2008, 2014) to assess language and
krchoi@ucla.edu social skills in everyday practice and in research. For exam-
1 ple, Padilla (2020) reported that 76% of a worldwide sample
Autism Services, Research Division, Easterseals Southern
California, 27200 Tourney Rd., Ste. 255, Valencia, of 1,428 individuals who primarily provide ABA services
CA 91355, USA use the VB-MAPP, which reportedly can detect moderate
2
Department of Psychology, Western Michigan University, treatment gains (Lotfizadeh et al., 2018).
Kalamazoo, USA The VB-MAPP is based on Skinner’s (1957) analysis
3
Department of Health Policy and Management, University of verbal behavior. It is used to evaluate early social and
of California, Los Angeles, USA language skills that typically emerge during the first four
4
Autism Services Assessment Division, Easterseals Southern years of life. The VB-MAPP comprises three assessment
California, Valencia, USA areas: Milestones Assessment, Barriers Assessment, and

Vol.:(0123456789)
 Journal of Autism and Developmental Disorders

Transition Assessment. It also contains a task analysis that To analyze the content validity of the VB-MAPP, Padilla
is used to guide Individualized Education Program (IEP) and Akers (2021) had 13 experts “evaluate domain rele-
goals and to develop curricula. The skills within each of vance, age appropriateness, method of measurement appro-
the assessment areas are developmentally sequenced and priateness, and domain representation across the three
assessed in such a manner. We will focus on the Milestones levels of the Milestones Assessment, Early Echoic Skills
Assessment and Barriers Assessment, both of which are Assessment (EESA), and Barriers Assessment” (p. 4054).
criterion-referenced and often used in clinical practice to The results suggested that, in general, content validity was
guide curriculum development and assess learner progress. moderate to strong.
The VB-MAPP Milestones Assessment contains 16 Mason et al. (2018) used the VB-MAPP to assess the
domains, each composed of five to 15 specific milestones functional language of five participants with autism across
that are developmentally sequenced into three levels. Level four verbal operants (mands, tacts, intraverbals, echoics).
1 represents skills that typically emerge during the first They also measured performance of these operants using a
18 months of life; Level 2 represents skills that typically functional assessment procedure derived from Lerman et al.
emerge between 18 and 30 months of age; and Level 3 (2005) that Mason et al. termed “verbal operant analysis”
represents skills that typically emerge between 30 and (VOA). They found that the VOA and VB-MAPP assess-
48 months. There are a total of 170 milestones. A trained ments often yielded dissimilar results, with the former
behavior analyst familiar with Skinner’s (1957) verbal characteristically indicating higher levels of performance.
operants (e.g., mand, tact, intraverbal, echoic, mimetic) Therefore, treatment recommendations for children might
directly probes and observes the skills in a developmen- well differ, depending on whether their verbal behavior was
tally sequential order. Each domain takes a functional assessed with the VB-MAPP or the VOA.
approach to assessing the respective skill. For example, Dixon et al. (2015) administered the VB-MAPP, the
the mand domain is concerned with assessing the learn- PEAK Direct Training (PEAK-DT) assessment (Dixon et al.,
er’s ability to request various actions or items vocally or 2014), and the PEAK generalization assessment (PEAK-G)
non-vocally. Early-level mands typically consist of one- to 40 students with autism. They reported statistically sig-
word requests but higher-level mands typically consist nificant correlations between the VB-MAPP and the PEAK
of requests within complex sentences. Upon completion assessments. Dixon et al. concluded that the VB-MAPP
of the assessment, the assessor calculates the cumulative served as a predictor of combined PEAK assessment scores.
number of milestones that the learner exhibits and reports It is not clear, however, if the same individual administered
the total score. The Barriers Assessment is used to assess all the assessments for a student and how closely in time
behaviors that may impede learning. The Barriers Assess- they were administered. Both variables can influence the
ment contains 24 separate questions related to potential concordance between measures. For example, if the second
barriers. A 4-point Likert scale is used to indicate the assessment was administered months after the first one, the
severity of each potential barrier. A cumulative Barriers second measure was likely to be inflated due to learning
Assessment score of 0–96 can be obtained; higher scores or maturity, which would have reduced the concordance
indicate the presence of more obstacles to learning. across the measures. In addition, there were no measures
Despite the popularity of the VB-MAPP in clinical of divergence to ascertain the convergent validity of the
practice and its recent use in autism outcomes research, assessments. Lastly, the PEAK assessment has been used to
its psychometric characteristics have not been extensively a lesser extent than other psychometrically validated assess-
evaluated and little is known about its reliability and validity, ment tools in autism research. To assess convergent validity
which is a weakness of the instrument (Gould et al., 2011; of the VB-MAPP adequately, it should be compared with
Montallana et al., 2019; Padilla et al., 2023). It appears that other, well established, assessments that appear to measure
only four published studies have evaluated its psychometric similar constructs (i.e., aspects of behavior).
properties (Dixon et al., 2015; Mason et al., 2018; Montal- One of the most utilized skills assessment tools in autism
lana et al., 2019; Padilla & Akers, 2021). interventions is the Vineland Adaptive Behavior Scale
Montallana et al. (2019) assessed the inter-rater reliabil- (VABS) (Sparrow & Cicchetti, 1989; Sparrow et al., 2005,
ity of the VB-MAPP Milestones and Barriers Assessments. 2016), which was used by 34% of 1,428 behavior analysts
A pair of independent behavior analysts administered the surveyed by Padilla (2020). The VABS has also appeared in
VB-MAPP to learners within two weeks of one another. numerous peer-reviewed studies that have evaluated adap-
The researchers measured the correlation between the two tive skills for autistic individuals. For example, Ridout and
scores for individual items and concluded that most mile- Eldevik (2023) conducted a literature review on measures
stones had moderate reliability. However, most items in the used in autism outcomes research. Out of 43 articles that met
Barriers Assessment had poor reliability (i.e., low inter-rated their inclusionary criterion, 39 indicated the VABS was the
agreement). measurement instrument for adaptive living skills. Only four
Journal of Autism and Developmental Disorders

articles included a secondary measure of adaptive function- the correlation between expressive language skills and adap-
ing in addition to the VABS. tive living skills on the VABS-II (Soriano et al., 2022), the
The VABS is a norm-referenced adaptive behavior assess- VB-MAPP Milestones scores should also be closely related
ment tool that measures adaptive living skills across the to the milestones that assess adaptive living skills on the
lifespan. The VABS comprises four domains: the Commu- VABS-II. Of the three VABS assessments, the psychometric
nication domain, Daily Living Skills domain, Socialization properties of the VABS-II have been studied most exten-
domain, and Motor Skills domain. There is also a Maladap- sively and it has been used most in longitudinal ABA out-
tive Behavior Index domain that can be used to assess prob- comes research (Ridout & Eldevik, 2023).
lem behaviors. The VABS is administered to an individual Given the favorable psychometric properties of the
familiar with the person being assessed in an interview for- VABS-II and its frequent administration in autism research
mat, with different forms designated for different interview- and practice (Stolte et al., 2016), it is of interest to com-
ees. Each question is scored on a three-point Likert-scale pare VABS-II scores to VB-MAPP scores. The VABS and
with a score of "2" indicating that a skill always occurs, a VB-MAPP assess similar constructs, hence comparing the
score of “1” indicating that it sometimes occurs, and “0” two provides an index of convergent validity. The purpose
indicating that it never or seldom occurs. Each domain that of the present study was to obtain that information. As
is assessed yields a raw score that can be converted into a Padilla et al. (2023) recently pointed out, further research
standard score, which can be compared with the normative is needed to provide strong evidence supporting the use of
samples. An Adaptive Behavior Composite (ABC) score the VB-MAPP.
also is obtained. For the Maladaptive Behavior scale score,
a higher score indicates more maladaptive behaviors.
Several studies have evaluated the psychometric prop- Method
erties of the VABS- I, II, and III and they have proven to
have appropriate psychometric properties (e.g., Perry & Participants
Factor, 1989). The psychometric properties of VABS-I
indicate excellent reliability and validity (Sparrow et al., We utilized archival clinical assessment data for clients who
1984). Similar conclusions have been obtained for the most received ABA assessments at the organization affiliated with
recent version (VABS-III). For example, the VABS-III has the first author. That organization, located in the western
excellent internal consistency, adequate test–retest reliability United States of American, provides a wide range of clinical
(e.g., r = 0.64 to 0.94), and good construct validity (r = 0.83) services for individuals with autism at multiple locations.
(Sparrow et al., 2016). Farmer et al. (2020) assessed the We conducted a power analysis with an effect size of 0.2,
concordance between VABS-III and VABS-II for a sample alpha of 0.05, and 80% power for a correlation based on
of 12.7- and 16.7-year-old participants and found concord- the method described by Kang (2021), which indicated a
ance between the two versions of the VABS. sample size of 150 participants. The inclusionary criteria
A VABS domain of interest for our purposes is the Com- consisted of individuals with a primary or secondary diag-
munication domain. The VABS’s Communication domain nosis of autism (based on the DSM-V categorization) who
milestones are categorized within a non-Skinnerian frame- had received a VABS and VB-MAPP assessment within
work, whereas the VB-MAPP is based on such a framework. 30 days of one another by the same clinician. A total of 235
Although the categorizations within each of the assess- individuals met these criteria and their archival data were
ments are different, both tools assess similar communica- analyzed. The participants’ ages ranged from three years to
tive behaviors. The VABS Communication domain consists nine years (M = 4.2 years, SD = 1.4). This sample consisted
of the Expressive, Receptive, and Written subdomains. In of 77.4% boys (n = 182) and 22.6% girls (n = 53). There were
contrast, the VB-MAPP evaluates specific verbal operants no significant gender differences in VB-MAPP or VABS
that fall within the three VABS Communications subdo- ABC standard scores by gender (p = 0.704 and p = 0.08,
mains. For example, the VB-MAPP does not have domains respectively). The average VB-MAPP Milestones score for
or subdomains labeled as expressive and receptive, but the the sample was 60.0 (SD = 34.1); the average VABS ABC
specific milestones within the mand and tact domains in the standard score was 68.2 (SD = 10.2). Other subdomain score
VB-MAPP are comparable to the milestones listed in the averages are shown in Table 1.
Expressive Communication subdomains of the VABS. The
VB-MAPP also contains a domain titled Writing, which is Materials
comparable to the VABS Written subdomain skills. There-
fore, the VABS Communication domain and the VB-MAPP The assessors used the VABS-II (Sparrow et al., 2005)
assessment index similar constructs and they should serve Comprehensive Interview Form, with the addition of the
as an appropriate index of convergence. In addition, given Maladaptive Behavior Index and Motor Skills domain for
 Journal of Autism and Developmental Disorders

Table 1  Sample description (N = 235) participants’ primary caretakers, who were in daily contact
Demographics
with the participants. The assessments were conducted in the
setting where ABA interventions were implemented, typi-
Age n (%) cally in the participants’ daily living environments.
3–4 years 99 (42.1%)
4–5 years 65 (27.7%) Procedures
5–6 years 35 (14.9%)
6–7 years 16 (6.8%) The researchers obtained the VABS and VB-MAPP data
7–8 years 10 (4.3%) through an internal data retrieval process designed for the
8–9 years 7 (3%) organization and transferred the numbers to a Microsoft
9–10 years 3 (1.3%) Excel® spreadsheet. The archival data were obtained from
Gender assessment reports that were completed between March
Boys 182 (77.1%) 2019 and September 2019. The assessments’ data were
Girls 53 (22.5%) obtained from archival records of clients who had been
VB-MAPP scores M (SD) assessed by Board Certified Behavior Analysts (BCBA®)
Avg. milestones score 60 (34.1) who conducted assessments as part of their regular clinical
Avg. barriers score 32 (19.5) job duties. The same BCBA® was listed on the records as
VABS scores the individual who administered both the VABS and VB-
ABC standard score 68.2 (10.2) MAPP assessment for a particular individual within 30 days
Communication raw score 102.1 (40.4) of one another. If different assessors administered the two
Socialization raw score 74.9 (30.2) tools or one assessor administered them more than 30 days
Daily living skills raw score 62.8 (28) apart, the participant was excluded from the study. The order
Motor raw score 106.4 (16.4) in which the assessments were administered was determined
Maladaptive behavior raw score 11.9 (6.2) by the assessors. The VB-MAPP data were gathered through
VB-MAPP Verbal Behavior Milestones Assessment and Placement direct observations of the client, and the VABS data were
Program, VABS Vineland Adaptive Behavior Scales, ABC Adaptive gathered through caregiver interviews.
Behavior Composite. One participant did not have a specified gender The BCBA® assessors were trained to conduct the VB-
MAPP and VABS assessment as part of the onboarding
training with the organization. The onboarding training con-
some participants who had a need for an assessment in sisted of at least 48 h of didactic training with role-play sce-
these domains. Although there were some VABS-III assess- narios and practice assessments. Two days were dedicated
ments available, we selected the VABS-II as the compari- to the VB-MAPP and VABS assessments. An experienced
son because (a) the clinicians had extensive experience and BCBA® trainer provided the training. The training consisted
training with the VABS-II and less so with the VABS-III of both synchronous and asynchronous online lessons with
when the study was conducted, (b) the psychometric proper- modeling and roleplay scenarios embedded within them.
ties of the VABS-II have been studied more extensively than Training sessions also included administrative procedures
the VABS-III, and (c) given the high correlation of VABS-II related to conducting assessments and a review of each
and VABS-III measures (Farmer et al., 2020), the results of assessment tool. The VB-MAPP specific training focused
any comparisons with the VABS-II should also hold true on the implementation of the Milestones assessment and
for the VABS-III. The assessors also used the VB-MAPP Barriers assessment and incorporating the assessment results
(Sundberg, 2008) assessment protocol. Both assessments into treatment planning. The VABS specific training focused
were conducted on an encrypted laptop computer using the on building rapport, conducting parent interviews, scoring
appropriate software designated for each assessment. The the assessment, using the appropriate software to score the
researchers transferred the assessment results onto a Micro- assessment, and treatment planning. Upon completion of the
soft Excel® spreadsheet. All analyses were performed using training, the assessors worked conjointly with other experi-
the R statistics package, version 4.2.1. enced assessors for at least two assessments. Next, an expe-
rienced mentor assessor observed the trainee for at least two
Setting assessments and provided feedback. All assessors had prior
experience in conducting the VB-MAPP.
Both assessments were conducted as part of regularly sched- This study was reviewed and approved as an exempt
uled annual ABA assessments for individuals referred to study by an appropriate Human Subjects Institutional
the organization by a large insurance company in Southern Review Board (HSIRB) prior to implementation of any of
California. The assessors obtained information from the the procedures. The clients were aware of the use of data
Journal of Autism and Developmental Disorders

for research purposes during intake and had the option to Results
decline to have their data used for research purposes. Only
data for clients who did not opt-out were included. Convergent Validity

We found moderate or strong positive correlations between

Interobserver Agreement (IOA) the VB-MAPP Milestones score and VABS Communica-
tion, Socialization, and Daily Living Skills subdomains
We calculated the reliability of the data that were obtained (p < 0.001) (Table 2). There was also a strong positive
through the automated data retrieval process and a research correlation between the VB-MAPP Milestones score and
assistant entered them in a spreadsheet. This was done by VABS overall raw score (R = 0.72, p < 0.001). No significant
also having a secondary researcher independently obtain relationship was observed between the VB-MAPP Barriers
data for 33.2% of the participants (n = 78). An interob- score and the VABS Internalizing, Externalizing, or overall
server agreement (IOA) coefficient was obtained after- Maladaptive Behavior scores. Significant correlations, their
wards, and a percentage reliability coefficient was cal- strength, and direction are shown in Table 2.
culated. This was done by first examining each pair of
data points that were obtained. We then tallied the total
number of data points that were the same and divided it Discriminant Validity
by the total number of data entry points and multiplied
the result by 100. The VABS IOA coefficient was 99.8% With respect to the VB-MAPP Milestones score, there
(ranging from 95 to 100% across assessment) and the VB- was a weak correlation with the VABS Motor raw score
MAPP IOA coefficient was 99.5% (ranging from 95 to (R = 0.40, p < 0.01). With respect to the VB-MAPP Barriers
100% across assessment categories). The overall IOA for score, there was a moderate but negative correlation with
both was 99.7%. the VABS Communication raw score (R = − 0.61, p < 0.01),
a moderate correlation with the VABS Socialization raw
score (R = − 0.58, p < 0.01), and a weak correlation with the
Analysis VABS Daily Living raw score (R = − 0.38, p < 0.01). There
was a weak correlation between the VB-MAPP Barriers
Descriptive statistics were used to characterize the study score and the VABS Motor raw sore (R = − 0.3, p < 0.01),
sample. We used independent sample t-tests to compare dif- which served as an index of discriminant validity. Overall,
ferences in VB-MAPP and VABS scores by gender. Then, these areas of convergent and discriminant validity were
we calculated Pearson R correlation coefficients for all VB- apparent in the dendrogram and cluster plot (Figs. 1 and 2).
MAPP and VABS overall scores and sub-domain scores to
assess convergent and discriminant validity in a correlation
matrix. We considered R values less than 0.3 to indicate no
correlation or a very weak correlation; 0.3 to 0.49 to indicate Discussion
a weak correlation; 0.5 to 0.69 to indicate a moderate cor-
relation; and greater than 0.7 to indicate a strong correlation Although applied behavior analysts generally are critical
(Hemphill, 2003). of standardized assessments (e.g., Johnston et al., 2020;
Following the development of the correlation matrix, we Poling et al., 1995), they frequently use the VB-MAPP
performed hierarchical cluster analysis on the correlation (Padilla, 2020). Its widespread use suggests that the VB-
coefficients’ distance matrix with agglomerative clustering MAPP has practical utility. But its technical adequacy has
to assess relatedness among score domains and sub-domains only recently garnered attention, with the first relevant pub-
visually (Bridges, 1966). We considered two, three, and lication, by Dixon et al. (2015), appearing seven years after
four-cluster solutions and ultimately selected a three-clus- the VB-MAPP was introduced (Sundberg, 2008). Results
ter solution as the best conceptual fit for related domains, of that study suggested that the VB-MAPP indexes aspects
as domains where we anticipated theoretical convergence/ of behavior (constructs) comparable to those measured by
discrimination were best differentiated with a three-cluster assessments developed by Dixon et al. (2014), namely the
solution. A dendrogram was generated to visualize hierar- PEAK-G and PEAK-GT.
chical clusters. Finally, we developed a cluster plot using To provide additional information relevant to the con-
k-means cluster analysis on the same distance matrix to fur- vergent validity of the VB-MAPP Milestones Assessment,
ther visualize related score domains and sub-domains (Kas- we compared 235 children’s Milestones scores with their
sambara & Mundt, 2017). Three conceptually meaningful VABS overall, communication, social skills, and daily living
clusters were apparent in both analyses. scores, obtaining strong or moderate correlations between
 Journal of Autism and Developmental Disorders

the measures. Because the VB-MAPP Milestones Assess-

internalizing externalizing maladaptive motor skills

ment is intended to index verbal behavior, a key component

− 0.30**

0.54**
0.47**
0.65**
0.26**
0.72**
0.40**

− 0.05
VABS of communication (Lechago, 2020; Sundberg, 2020), it is

0.13
0.05
raw
unsurprising but heartening that there was a strong (0.79)
correlation between Milestones scores and VABS Com-
munication scores. Effective verbal behavior also makes

0.81**
0.83**
− 0.15
VABS

0.13
0.17
0.01
0.16
0.18
0.06 the acquisition and performance of many daily living skills
raw

easier, which may account for the moderate correlation we

obtained between VB-MAPP Milestones and VABS daily
living scores. The moderate (0.68) correlation we obtained
0.17**
0.19**
0.21**
VABS

0.06
0.18
0.34
0.03

between VABS Communication and VABS Daily Living

− 0.1
raw

supports this possibility.

The present findings suggest that the VB-MAPP Mile-

VB-MAPP Verbal Behavior Milestones Assessment and Placement Program, VABS Vineland Adaptive Behavior, ABS Adaptive Behavior Composite
stones Assessment measures aspects of social and commu-
− 0.15*

− 0.05
VABS daily VABS

0.09
0.04
0.09
0.09
0.08

nicative behavior comparable to those indexed by VABS

raw

subdomains designed for this purpose. That is, the instru-

ments have strong convergent validity. Given that the VABS
socialization living raw

is a widely used instrument with good reliability and valid-

− 0.38**
0.44**
0.87**
0.68**
0.63**
0.55**

ity (e.g., Perry & Factor, 1989), this finding supports the
continued use of the VB-MAPP Milestones Assessment.
The moderate to strong interrater agreement obtained for
− 0.58**

items comprised by the instrument (Montallana et al., 2019)

0.64**
0.84**
0.68**
0.55**
VABS

does likewise. Finally, support for the Milestones Assess-

raw

ment comes from our determination of divergent validity.

To assess divergent validity, we determined the correlation
VABS com-
munication

between VB-MAPP Milestones scores and VABS Motor

− 0.61**
0.58**
0.91**
0.79**

scores, which was low. We also determined the correlation

raw

between VB-MAPP Milestones scores and VABS Barriers

scores, which was moderate. Although one would expect an
VABS sum of all raw

inverse relationship between functional communication and

Table 2  Correlation Matrix of VB-MAPP and VABS Overall and Subdomain Scores

problem behavior (Hagopian et al., 1998), we did not antici-

pate such an association between the VABS Communica-
− 0.59**
0.60**
0.72**

tion domain and the VABS Maladaptive Behavior domains

because the former does not assess functional communi-
cation. There was either a weak correlation between these
standard score

two measures or no correlation at all, which provided for a

VABS ABC

suitable index of divergence with the VB-MAPP Milestones

− 0.46**
0.44**

scores. The present results indicate appropriate divergence

between unrelated domains of the two instruments.
In its totality, the available evidence suggests that the
VB-MAPP barriers

Milestones Assessment is defensible. Further research is

needed, however, to determine whether the relative insen-
sitivity of the Milestones Assessment reported by Mason
− 0.65**

et al. (2018) is of practical importance and to assess the kind

of training needed for practitioners to use the instrument
appropriately (see Barnes et al., 2014).
VABS ABC standard score

VABS communication raw

A previous study found that the interrater reliability of

VABS externalizing raw
VABS internalizing raw
VABS socialization raw

VABS maladaptive raw

VABS daily living raw
VB-MAPP milestones

most individual items on the Barriers Assessment of the

VABS sum of all raw
VB-MAPP barriers

P < .05, **P < .01

VB-MAPP was low (Montallana et al., 2019), and its authors

recommended the assessment be used with caution. The 24
behavioral domains are rated from zero (no issues) to four
(a significant barrier to language acquisition). The compos-
ite Barriers Assessment score indicates the probable overall
*
Journal of Autism and Developmental Disorders

Fig. 1  Cluster dendrogram

Fig. 2  Cluster plots

difficulty that a child will encounter in acquiring verbal the absence of verbal operants quantified in the Milestones
behavior and individual scores target specific domains Assessment (e.g., mand). Others are concerned with the
likely to pose problems. The domains differ widely. Some presence of specific challenging responses (e.g., self-stimu-
(e.g., absent, weak, or impaired mand control) simply reflect lation), general patterns of responding (e.g., hyperactivity),
 Journal of Autism and Developmental Disorders

the absence of motivation (e.g., weak or atypical motivat- counterbalance the order in which assessments occur. A
ing operations), or inappropriate learned patterns of behav- second limitation is that despite the assessors' training
ior (e.g., prompt dependency). As a collective they do not and qualifications being consistent with the requirements
form a meaningful construct and there is no published data outlined in the VB-MAPP, it may not have been sufficient
indicating that the ease with which children acquire ver- to administer the Barriers assessment of the VB-MAPP
bal behavior is directly correlated to their overall Barriers appropriately. Future studies can assess this limitation by
Assessment score. reporting the interobserver agreement scores for the Barri-
In the present study, correlations between Barriers scores ers. In addition, future studies should focus on the psycho-
and VABS Maladaptive Behavior raw scores (r = 0.06), metric properties of the Barriers assessment as a function
Externalizing raw scores (r = 0.03), and Internalizing (0.08), of how clinicians are trained to implement that portion of
all measures of maladaptive behavior, were low and not sta- the VB-MAPP. Lastly, given that the VABS-II is outdated,
tistically significant. The absence of convergent validity the present study should be replicated using the VABS-III.
between the VABS measures of maladaptive behavior and Although there is good concordance between the VABS-II
VB-MAPP Barriers scores suggests that the latter should not and -III, the convergence of the VB-MAPP with the VABS-
be construed as an index of maladaptive behavior as gener- III can only be inferred based on these results, therefore,
ally construed. The only promising finding with respect to additional studies are warranted.
the VB-MAPP Barriers score was that there was a statisti-
cally significant inverse correlation between the VB-MAPP
Author Contributions ADL: Study conception and design, analysis,
Barriers score and the VABS Communication and Social- and manuscript writeup, BG: Data collection and manuscript prep-
ization score. Such a relation is expected because higher eration, CR: Data collection and interobserver agreement calculation,
communication and socialization skills are associated with AP: Study design and manuscript writeup, KC: Study design and data
fewer maladaptive behaviors (Charlop‐Christy et al., 2002). analysis.
However, the poor overall reliability of the VB-MAPP Bar-
riers Assessment and the lack of data indicating Barriers
Declarations
scores relate to the acquisition of verbal behavior points to a Conflict of interest There were no conflicts of interest in the present
need for substantial revision and improvement in this section study.
of the VB-MAPP assessment tool. In its present form, the
VB-MAPP Barriers Assessment does not appear sufficiently
valid or reliable to be used in research or in making impor-
tant decisions in everyday practice. References
The present results suggest that the VB-MAPP Mile-
stones Assessment is a valid and reliable tool for assess- American Psychiatric Association. (2013). Diagnostic and statisti-
cal manual of mental disorders (5th ed.). American Psychiatric
ing communication, and a prior study indicates that it can Association.
detect moderate improvements in communication (Lotfiza- Barnes, C. S., Mellor, J. R., & Rehfeldt, R. A. (2014). Implementing
deh et al., 2018), which is another point in its favor. But, like the verbal behavior milestones assessment and placement program
the results of a prior study (Montallana et al., 2019) they (VB-MAPP): Teaching assessment techniques. Analysis of Verbal
Behavior, 30, 36–47. https://​doi.​org/​10.​1007/​s40616-​013-​0004-5
also suggest the VB-MAPP Barriers Assessment is not an Bridges, C. C., Jr. (1966). Hierarchical cluster analysis. Psychological
adequate tool for assessing maladaptive behaviors in clinical Reports, 18(3), 851–854. https://​doi.​org/​10.​2466/​pr0.​1966.​18.3.​
practice or in research. Additional studies are warranted to 851
examine the correlation between the VB-MAPP domains Charlop-Christy, M. H., Carpenter, M., Le, L., LeBlanc, L. A., & Kel-
let, K. (2002). Using the picture exchange communication system
and VABS sub-domains to determine if specific VABS sub- (PECS) with children with autism: Assessment of PECS acquisi-
domains are better predictors of VABS domain scores. tion, speech, social-communicative behavior, and problem behav-
Although these findings offer support for the convergent ior. Journal of Applied Behavior Analysis, 35(3), 213–231. https://​
validity of VABS and VB-MAPP Milestones assessment doi.​org/​10.​1901/​jaba.​2002.​35-​213
del Hoyo Soriano, L., Villarreal, J. C., Sterling, A., Edgin, J., Berry-
but fail to do so for the VB-MAPP Barriers assessment, the Kravis, E., Hamilton, D. R., & Abbeduto, L. (2022). The associa-
study has limitations. The primary limitation was that the tion between expressive language skills and adaptive behavior
order that the two assessments were administered was not in individuals with Down syndrome. Scientific Reports, 12(1),
counterbalanced, but rather determined by the individuals 20014. https://​doi.​org/​10.​1038/​s41598-​022-​24478-x
Dixon, M. R. (2014). The PEAK relational training system: Direct
who administered the assessment. If the VABS was con- training module. Shawnee Scientific.
sistently administered prior to the VB-MAPP, an assessor Dixon, M. R., Belisle, J., Stanley, C., Rowsey, K., Daar, J. H., &
may have been less likely to directly observe participants’ Szekely, S. (2015). Toward a behavior analysis of complex
behaviors when using the latter assessment. Future studies language for children with autism: Evaluating the relationship
between PEAK and the VB-MAPP. Journal of Developmental
assessing the convergent validity of the VB-MAPP should
Journal of Autism and Developmental Disorders

and Physical Disabilities, 27, 223–233. https://​doi.​org/​10.​1007/​ National Autism Center. (2015). Findings and conclusions: National
s10882-​014-​9410-4 standards project (phase 2). Author.
Drash, P. W., High, R. L., & Tudor, R. M. (1999). Using mand training Padilla, K. L. (2020). Global assessment use and practices in applied
to establish an echoic repertoire in young children with autism. behavior analysis: Surveying the field. Research in Autism Spec-
The Analysis of Verbal Behavior, 16(1), 29–44. https://​doi.​org/​ trum Disorders. https://​doi.​org/​10.​1016/j.​rasd.​2020.​101676
10.​1007/​BF033​92945 Padilla, K. L., & Akers, J. S. (2021). Content validity evidence for the
Farmer, C., Adedipe, D., Bal, V. H., Chlebowski, C., & Thurm, A. verbal behavior milestones assessment and placement program.
(2020). Concordance of the vineland adaptive behavior scales, Journal of Autism and Development Disorders, 51, 4054–4066.
second and third editions. Journal of Intellectual Disability https://​doi.​org/​10.​1007/​s10803-​020-​04864-y
Research, 64(1), 18–26. https://​doi.​org/​10.​1111/​jir.​12691 Padilla, K. L., Weston, R., Morgan, G. B., Lively, P., & O’Guinn, N.
Goldsmith, T. R., LeBlanc, L. A., & Sautter, R. A. (2007). Teaching (2023). Validity and reliability evidence for assessments based on
intraverbal behavior to children with autism. Research in Autism applied behavior analysis: A systematic review. Behavior Modi-
Spectrum Disorders, 1(1), 1–13. https://​doi.​org/​10.​1016/j.​rasd.​ fication, 47(1), 247–288. https://​doi.​org/​10.​1177/​01454​45522​
2006.​07.​001 10981​51
Gould, E., Dixon, D. R., Najdowski, A. C., Smith, M. N., & Tarbox, Perry, A., & Factor, D. C. (1989). Psychometric validity and clini-
J. (2011). A review of assessments for determining the content cal usefulness of the Vineland Adaptive Behavior Scales and the
of early intensive behavioral intervention programs for autism AAMD Adaptive Behavior Scale for an autistic sample. Journal
spectrum disorders. Research in Autism Spectrum Disorders, 5(3), of Autism and Developmental Disorders, 19(1), 41–55. https://d​ oi.​
990–1002. https://​doi.​org/​10.​1016/j.​rasd.​2011.​01.​012 org/​10.​1007/​BF022​12717
Hagopian, L. P., Fisher, W. W., Sullivan, M. T., Acquisto, J., & LeB- Poling, A., Methot, L., & LeSage, M. (1995). Fundamentals of behav-
lanc, L. A. (1998). Effectiveness of functional communication ior analytic research. Plenum Press.
training with and without extinction and punishment: A summary Ridout, S., & Eldevik, S. (2023). Measures used to assess treatment
of 21 inpatient cases. Journal of Applied Behavior Analysis, 31(2), outcomes in children with autism receiving early and inten-
211–235. https://​doi.​org/​10.​1901/​jaba.​1998.​31-​211 sive behavioral interventions: A review. Review Journal of
Hemphill, J. F. (2003). Interpreting the magnitudes of correlation coef- Autism and Developmental Disorders. https://​doi.​org/​10.​1007/​
ficients. American Psychologist, 58(1), 78–79. https://​doi.​org/​10.​ s40489-​023-​00355-9
1037/​0003-​066X.​58.1.​78 Skinner, B. F. (1957). Verbal behavior. Appleton-Century-Crofts.
Ingvarsson, E. T., & Hollobaugh, T. (2010). Acquisition of intraverbal Sparrow, S. S., Balla, D., & Cicchetti, D. V. (1984). Vineland adap-
behavior: Teaching children with autism to mand for answers to tive behavior scales (Survey Form). American Guidance Service.
questions. Journal of Applied Behavior Analysis, 43(1), 1–17. Sparrow, S. S., & Cicchetti, D. V. (1989). The Vineland adaptive
https://​doi.​org/​10.​1901/​jaba.​2010.​43-1 behavior scales. Allyn & Bacon.
Johnston, J. M., Pennypacker, H. S., & Green, G. (2020). Strategies and Sparrow, S. S., Cicchetti, D. V., & Balla, D. A. (2005). Vineland-
tactics of behavioral research and practice (4th ed.). Routledge. ii: Vineland adaptive behaviour scales, survey forms manual.
Kang, H. (2021). Sample size determination and power analysis using Pearson.
the G* Power software. Journal of Educational Evaluation for Sparrow, S. S., Cicchetti, D. V., & Saulnier, C. A. (2016). Vineland
Health Professions. https://​doi.​org/​10.​3352/​jeehp.​2021.​18.​17 adaptive behavior scales, Third Edition (Vineland-3). Pearson.
Kassambara, A., & Mundt, F. (2017). Package ‘factoextra.’ Extract and Stolte, M., Hodgetts, S., & Smith, V. (2016). A critical review of out-
Visualize the Results of Multivariate Data Analyses, 76(2), 458. come measures used to evaluate the effectiveness of comprehen-
Lechago, S. A. (2020). Verbal behavior. In S. Hupp & J. D. Jewell sive, community based treatment for young children with ASD.
(Eds.), The encyclopedia of child and adolescent development Research in Autism Spectrum Disorders, 23, 221–234. https://d​ oi.​
(pp. 1–11). Wiley. org/​10.​1016/j.​rasd.​2015.​12.​009
Lerman, D. C., Parten, M., Addison, L. R., Vorndran, C. M., Volk- Sundberg, M. L. (2008). Verbal behavior milestones assessment and
ert, V. M., & Kodak, T. (2005). A methodology for assessing placement program: The VB-MAPP. AVB Press.
the functions of emerging speech in children with developmental Sundberg, M. L. (2014). The verbal behavior milestones assessment
disabilities. Journal of Applied Behavior Analysis, 38, 303–316. and placement program: The VB-MAPP (2nd ed.). AVB Press.
https://​doi.​org/​10.​1901/​jaba.​2005.​106-​04 Sundberg, M. L. (2020). Verbal behavior. In J. O. Cooper, T. E. Heron,
Lotfizadeh, A. D., Kazemi, E., Pompa-Craven, P., & Eldevik, S. (2018). & W. L. Heward (Eds.), Applied behavior analysis (3rd ed., pp.
Moderate effects of low-intensity behavioral intervention. Behav- 412–449). Pearson.
ior Modification, 44(1), 92–113. https://​doi.​org/​10.​1177/​01454​
45518​796204 Publisher's Note Springer Nature remains neutral with regard to
Mason, L. L., Sawchak, A., Scott, C. S., Andrews, A., Arriaga, A., & jurisdictional claims in published maps and institutional affiliations.
Pena, H. (2018). A comparison of outcomes from criterion-refer-
enced and experimental evaluations of verbal behavior. Behavio- Springer Nature or its licensor (e.g. a society or other partner) holds
ral Development, 23(2), 118–129. https://​doi.​org/​10.​1037/​bdb00​ exclusive rights to this article under a publishing agreement with the
00080 author(s) or other rightsholder(s); author self-archiving of the accepted
Montallana, K. L., Gard, B. M., Lotfizadeh, A. D., & Poling, A. manuscript version of this article is solely governed by the terms of
(2019). Inter-rater agreement for the milestones and barriers such publishing agreement and applicable law.
assessments of the verbal behavior milestones assessment and
placement program (VB-MAPP). Journal of Autism and Devel-
opmental Disorders, 49(5), 2015–2023. https://​doi.​org/​10.​1007/​
s10803-​019-​03879-4