Research in Autism Spectrum Disorders 7 (2013) 1196–1203

Contents lists available at ScienceDirect

Research in Autism Spectrum Disorders

Journal homepage: http://ees.elsevier.com/RASD/default.asp

Parent and teacher ratings of adaptive and challenging

behaviours in young children with autism spectrum disorders
Ben R. Lane a,*, Jessica Paynter b, Rachael Sharman a
a
University of the Sunshine Coast, Locked Bag 4, Maroochydore DC, Queensland 4558, Australia
b
AEIOU Foundation, PO Box 226, Brisbane, Queensland 4105, Australia

A R T I C L E I N F O A B S T R A C T

Article history: This study examined parent and teacher rating correspondence of adaptive and
Received 31 May 2013 challenging behaviours for children with autism spectrum disorders (ASDs) enrolled in
Accepted 9 July 2013 an early intervention program. Data were collected on the Behaviour Assessment System
for Children, second edition (BASC-2; n = 22), and Vineland Adaptive Behaviour Scales,
Keywords: second edition (VABS-II; n = 28). Adaptive behaviour ratings generally demonstrated high
Autism spectrum disorders
parent–teacher correlations, while challenging behaviour ratings demonstrated relatively
VABS-II
low correlations. Only adaptive skills on the BASC-2 showed significant mean parent–
BASC-2
Adaptive behaviour teacher differences, with parent ratings suggesting greater impairment. Results suggest
Challenging behaviour that clinicians should consider gaining both parent and teacher perspectives on a child’s
challenging behaviour, but that a single informant measure of adaptive behaviour, either
parent or teacher, may be sufficient after initial assessments are completed.
ß 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Early intervention is a promising approach for improving the outcomes of individuals with autism spectrum disorders
(ASDs; Magiati, Tay, & Howlin, 2012; Matson & Jang, 2013; Prior, Roberts, Rodger, Williams, & Sutherland, 2011). ASDs are
lifelong conditions, so the main aims of treatment are to minimise symptoms and maximise the individual’s ability to
function effectively in their environment (Ozonoff, Goodlin-Jones, & Solomon, 2005). The skills to achieve this effective
functioning are adaptive behaviours (Paul et al., 2004). Challenging behaviours (e.g., hyperactivity, aggression, attention
problems) are a similarly important outcome variable, as they are often one of the greatest difficulties in families and barriers
to education (Lecavalier, Leone, & Wiltz, 2006). At least one challenging behaviour is usually present for children with ASDs
(Matson, Nebel-Schwalm, & Matson, 2007), and, in some circumstances, such behaviour can place others at risk of injury
(Murphy, Healy, & Leader, 2009). These problems can exclude individuals from accessing community services, preventing
effective integration into society (Matson et al., 2007; Murphy et al., 2009).
Selecting an appropriate measurement system for these variables is a necessary decision facing clinicians in early
intervention. Measurement must be accurate to effectively assess progress (Reed & Osborne, 2013) and inform future
treatment. This requires the measurement tools to be appropriate for the target population and treatment goals (Lord et al.,
2005). The decision should also consider how much unique information the system provides about a child’s condition;
minimising unnecessary assessments is important as financial resources and time are both limited for many services
providing intervention for ASDs. Two commonly used assessment tools for children with ASDs are the Vineland Adaptive

* Corresponding author at: School of Social Sciences, University of the Sunshine Coast, Locked Bag 4, Maroochydore DC, 4558 Queensland, Australia.
Tel.: +61 407 145 449.
E-mail addresses: blane@usc.edu.au (B.R. Lane), jessica.paynter@aeiou.org.au (J. Paynter), rsharman@usc.edu.au (R. Sharman).

1750-9467/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.rasd.2013.07.011
 B.R. Lane et al. / Research in Autism Spectrum Disorders 7 (2013) 1196–1203 1197

Behaviour Scales, second edition (VABS-II; Sparrow, Cicchetti, & Balla, 2005), and the Behaviour Assessment System for
Children, second edition (BASC-2; Reynolds & Kamphaus, 2004). The VABS-II focuses on adaptive behaviours, but includes
maladaptive behaviour indices on the parent rating form. In contrast, the BASC-2 focuses on challenging behaviours, but
includes adaptive scales on both the parent and teacher rating forms.
The use of multiple informants (such as parents and teachers) assumes that these individuals will provide unique
information beyond what is possible from one informant. It might be expected that they would establish similar
assessments, but this is not always the case (Kalyva, 2010; Kanne, Abbacchi, & Constantino, 2009). Rating differences likely
represent much more than measurement error (Achenbach, 2011; Hartley, Zakriski, & Wright, 2011), and potentially arise
due to changes in behaviour across different situations or environments (Kanne et al., 2009). For instance, in the home
environment, there may be few demands placed on the child beyond normal activities. In contrast, the teaching or
intervention environment likely contains more goal-directed activities. These environments likely affect children on a
person-by-situation basis (Hartley et al., 2011), with effects possibly being moderated according to preferences, such as for
structure versus autonomy. However, studies suggest environmental factors cannot account for all rating disparities. For
instance, researchers have found modest parent–teacher correlations even when rating aggression following specific events
(Hartley et al., 2011) or when teachers rated the children’s behaviour based on observations in a home-based intervention
context (Reed & Osborne, 2013). Another potential source of ratings differences draws from characteristics of the informant.
Recent research suggested that parent education level and age affect perceptions of autism symptoms (Hattier, Matson,
Belva, & Adams, 2013). Past studies have found that stress or depression levels correlate with more severe parent ratings
(Szatmari, Archer, Fisman, & Streiner, 1994; Youngstrom, Loeber, & Stouthamer-Loeber, 2000), possibly through reducing
tolerance for disruptive behaviour or a lack of receptiveness to social advances. These factors may affect ratings through
influencing interactions with the child (changing the child’s behaviour) or through a cognitive bias (changing the informant’s
perception).
When assessing informant differences, past research has identified the value of using a combination of correlation
statistics and mean difference testing (Szatmari et al., 1994; Voelker, Shore, Lee, & Szuszkiewicz, 2000). Correlations provide
evaluation of whether the children are rank-ordered consistently between informants, whereas mean differences can reveal
potential tendencies to rate higher or lower. When taken together, these statistics can indicate whether there are differences
and whether these differences are systematic or random. This approach requires that parent and teacher ratings be matched
for each assessed child.
Parent and teacher congruence has been examined for many behaviours and age groups of children with ASDs (Geiger,
Smith, & Creaghead, 2002; Kanne et al., 2009; Lecavalier et al., 2006; Murray, Ruble, Willis, & Molloy, 2009; Reed & Osborne,
2013). Few studies were found reporting on parent–teacher rating congruence for children with ASDs on the BASC and VABS
(limited to Barnhill et al., 2000; Foley Nicpon, Doobay, & Assouline, 2010; Szatmari et al., 1994). One of these studies included
correlation statistics, using the original VABS (N = 83; Szatmari et al., 1994). Szatmari et al. found that although parent–
teacher correlations were moderate to high (r = .42–.83), parents systematically rated the children’s impairments as more
severe than teachers did. Voelker et al. (2000) found similar results with a sample of children with low intellectual quotients
(IQs) without ASDs. However, they highlighted that the norm referencing method for the original VABS resulted in parent
scores having a lower absolute limit, artificially biasing scores to make them appear more severe than teacher scores. Also in
a study with children with non-ASD disabilities, Hundert, Morrison, Mahoney, and Vernon (1997) found that parent ratings
were only more severe than teacher ratings for children with severe developmental delays and not for those with moderate
or no delays. These findings support Voelker et al.’s (2000) argument that differences are due to a lower absolute limit, as
scores for children with moderate and no impairment would tend to be insufficiently low to elicit the effect and thus show no
parent–teacher differences. No study (to the authors’ knowledge) has reported on parent–teacher ratings using the updated
VABS-II, so it is unclear whether these results would replicate for this instrument.
Barnhill et al. (2000) examined differences between parent and teacher ratings of children and adolescents with Asperger
disorder on the original BASC (N = 20). They found that, on average, teachers rated all composites (externalising problems,
internalising problems, behavioural symptoms, and adaptive skills) more positively than did parents. Foley Nicpon et al.
(2010) found similar differences between parent and teacher ratings on the updated BASC-2 for children (n = 39) and
adolescents (n = 15) with ASDs and intellectual giftedness. However, both studies did not include correlation statistics,
making it unclear whether these differences were systematic. Furthermore, as Foley Nicpon et al. did not match the parent
and teacher ratings, extra parent ratings were included due to a higher response rate from parents. Studies with other
measures of challenging behaviour have not replicated these results (Lecavalier et al., 2006; Reed & Osborne, 2013). For
instance, Reed and Osborne, using the Connors’ Rating Scale (CRS-R; Conners, 1997) as a measure of challenging behaviour,
found mean parent–teacher differences only for oppositional behaviours (e.g., rule breaking, authority problems) in a sample
of young children with ASDs (N = 52; M age = 44.4 months). Teachers rated these problems more severely than parents did.
This contrasts with results on the BASC, but may be attributable to differences in the age of the children and the context of the
teacher relationship. Reed and Osborne’s teacher respondents were involved in a home-based treatment program.
Correlation statistics showed that parent–teacher correspondence was poor for all behaviours, except attention deficit
hyperactivity disorder (ADHD) symptoms when assessed by a parent with low stress.
The present study focused on children diagnosed with ASDs, aged 2½ to 6 years who participated in an early intervention
program at an AEIOU centre.The AEIOU foundation is a not for profit organisation in Queensland, Australia, providing full-
time early intervention to children with ASDs. The AEIOU program aims to support development in four areas: social
1198 B.R. Lane et al. / Research in Autism Spectrum Disorders 7 (2013) 1196–1203

emotional, language and communication, physical, and cognitive (Paynter, Scott, Beamish, Duhig, & Heussler, 2012).
However, the intervention is tailored to individualised plans for each child based on a comprehensive assessment. Parents are
actively encouraged to participate in program activities, goal setting, and decision-making. Staff include speech therapists,
occupational therapists, early childhood teachers, and childcare professionals at each centre. However, teacher respondents for
this study included only early childhood teachers or childcare professionals, both of whom have high personal contact with the
children on ratios of 1:2 to 1:4 staff members per child. Children enrolled full-time spend a minimum of 25 h per week in the
program. The existence of differences in informant ratings would have implications for the identification of strengths and needs
and provision of early intervention programs such as the AEIOU program. Specifically, it highlights whether having two
informant ratings (parents and teachers) provides different or additional information to a single rating.
Little research has considered the extent to which the VABS-II and BASC-2 provide unique information about a child with
an ASD. In particular, no studies have reported correlation statistics for parent and teacher ratings of children with ASDs on
the BASC-2 or updated VABS-II and no studies have examined parent and teacher ratings on these measures concurrently or
in the early intervention context. The BASC-2 and VABS-II are comprehensive measurement tools and thus, it is difficult to
obtain complete data for research purposes. Having data for both these measures allows comparisons between their
properties and stronger inferences about the results. Furthermore, this study used a unique, relatively homogenous teacher
sample working in the early intervention context. This controls for a number of participant characteristics, such as exposure
to ASD cases, which may influence perceptions. Thus, this study makes a novel contribution to this research area, while
controlling for several key variables. The study’s purpose was to compare parent and teacher ratings of young children with
ASDs on the BASC-2 and VABS-II, and examine the level of congruence.

2. Method

2.1. Participants

Participants included a set of 41 parents and teachers of children with ASDs enrolled in an AEIOU program. At the time of
this study, entry to this program was reliant on having a medical practitioner’s (paediatrician, child psychiatrist, or
neurologist) DSM-IV-TR (American Psychiatric Association, 2000) diagnosis of an ASD (Autistic Disorder, Asperger Disorder,
or Pervasive-Developmental Disorder—Not Otherwise Specified) and an age at intake between 30 and 71 months. Diagnosis
was verified by parent-completion of the Social Communication Questionnaire (SCQ) and inclusion required a score of 11 or
higher as recommended by previous research for children in the preschool period (Lee, David, Rusyniak, Landa, &
Newschaffer, 2007). After removing two children who scored below this cut-off, scores on the SCQ ranged between 13 and 28
(M = 19.03, SD = 4.18). This left a set of 39 cases (32 male, 7 female), with ages at the assessment times ranging from 34 to 71
months (M = 51.63, SD = 9.60).
For the BASC-2, the time between parent and teacher completion ranged between 0 and 42 days (M = 18.86, SD = 9.37).
One case with a difference greater than 50 days was excluded. For the VABS-II, the time ranged between 0 and 34 days
(M = 11.38, SD = 9.13). Four cases with a difference greater than 50 days were excluded. These exclusions left a subset of 22
matched parent–teacher pairs for the BASC-2 and 28 for the VABS-II. There were 18 cases where parent–teacher data were
available for both the BASC-2 and the VABS-II, with time between these assessments ranging from 58 to 100 days (M = 74.33,
SD = 12.42) for teacher assessments and from 64 to 224 days (M = 106.80, SD = 35.34) for parent assessments. Assessment
times differed as the VABS-II was added in a later term to accommodate changes to required outcome measures. Most
respondents indicated that the primary carer was female (97%, but with five missing data), however no data were collected
on which parent filled out the survey materials. Table 1 presents additional parent characteristics data. The same teacher
respondent completed the BASC-2 and VABS-II materials for each child.

2.2. Materials

2.2.1. Social Communication Questionnaire (SCQ)

The SCQ is a 40-item parent-completed measure, derived from the Autism Diagnostic Interview-Revised to assess the
presence of autistic behaviours (Rutter, Bailey, & Lord, 2003). Items require yes or no responses to produce a total score
between 0 and 39. Seven items are excluded for nonverbal children, producing a score between 0 and 32. A cut-off of 11 was
used as recommended for identifying cases of probable ASDs for children receiving preschool special education services
(Lee et al., 2007).

Table 1
Demographic data for parent respondents.

Yes No Missing

Female gender 33 (97%) 1 (3%) 5 (13%)

Tertiary or higher education 23 (68%) 11 (32%) 5 (13%)
Culturally or linguistically diverse background 10 (29%) 25 (71%) 4 (10%)
Earns below $37,960 per year 13 (36%) 23 (64%) 3 (8%)
Child lives with both parents 33 (92%) 3 (8%) 3 (8%)
 B.R. Lane et al. / Research in Autism Spectrum Disorders 7 (2013) 1196–1203 1199

2.2.2. Behaviour Assessment System for Children, second edition (BASC-2)

The BASC-2 uses a multidimensional assessment approach with parent, teacher, and self-forms (Reynolds & Kamphaus,
2004). Versions are available for parents and teachers for preschoolers (ages 2–5), children (ages 6–11), and adolescents (12–
21). For this study, only the parent and teacher forms for preschoolers are relevant. The preschool forms assess behaviour in 12
primary scales arranged into four composites. These include the behavioural symptoms (atypicality and withdrawal),
internalising (anxiety, depression, and somatisation), externalising (hyperactivity and aggression), and adaptive skills
(adaptability, social skills, functional communication, and activities of daily living) composites. An attention problems scale is
also included separately. On the parent and teacher forms, respondents rate behavioural frequency on a 4-point scale (Never to
Almost Always). Total raw scores for each primary scale then convert to composite scores, T-scores (m = 50, s = 10), and
percentile ranks.
The BASC-2 manual documents coefficient alpha scores above .8 for all scales and composites, demonstrating
appropriate internal consistency (Reynolds & Kamphaus, 2004). Test–retest reliability scores ranged above .70. For
each scale, median interrater reliability scores ranged between .53 and .74 for scores between teacher respondents, and
in the .70s for parent respondents (Reynolds & Kamphaus, 2004). Construct validity was demonstrated though
correlations with similar behavioural scales (Reynolds & Kamphaus, 2004). Both teacher and parent forms showed high
correlations with the Achenbach System of Empirically Based Assessment Caregiver-Teacher report forms and the
original BASC. The teacher form also showed high correlations with Conners’ Teaching rating scale-revised, while the
parent form showed similar correlations with the Child Behaviour Checklist and the Behaviour Ratings Inventory of
Executive Functioning.

2.2.3. Vineland Adaptive Behaviour Scales, second edition (VABS-II)

The VABS-II has parent and teacher forms, assessing adaptive behaviour in four domains with subscales in each (in
brackets) including: communication (receptive, expressive, and written); daily living (personal, domestic/academic, and
community/school community); socialisation (interpersonal relationships, play and leisure time, and coping skills); and
motor skills (fine and gross motor; Sparrow et al., 2005). Raw scores convert to standard scores (m = 100, s = 15) for the
composites, v-scale scores (m = 15, s = 3) for the subdomains, and age-equivalent scores. Lower scores indicate greater
impairment. The parent form also includes a maladaptive behaviour index with scales for externalising and internalising
problems.
The VABS-II manuals report extensive reliability statistics for the preschool age range (Sparrow et al., 2005). For the
composites, test–retest and split-half scores ranged from .83 to .94, indicating high stability. Concerning interrater
reliability, Sparrow et al. (2005) reported mean correlation coefficients for domains of .54 and .78 and subdomains of .49 and
.81 for teachers and parents, respectively.

2.3. Procedure

AEIOU collected the data as part of their usual intake assessment processes. The Griffith University ethics department
granted approval for use of the data in research (GU Ref No: EBL/88/10/HREC) and parents accessed an information sheet on
the project and gave informed consent. Parents were sent a demographics form and the SCQ at enrolment to the AEIOU
program and the BASC-2 and VABS-II at different points during enrolment. Teachers and childcare professionals received the
BASC-2 and VABS-II at the AEIOU centre.

3. Results

The data met assumptions for parametric testing. Some subscale data were not normal; however the correlation statistics
used are robust to such violations (Havlicek & Peterson, 1977).

3.1. Descriptive statistics

For the BASC-2, scores between 60 and 69 on the challenging behaviour scales and 30 and 39 on the adaptive skills
scales indicate the individual is in the at-risk range (Reynolds & Kamphaus, 2004). Mean teacher ratings for the
behavioural symptoms index and adaptive skills composites, and the atypicality, withdrawal, social skills, and
functional communication scales all scored in the at-risk range (see Table 2). Mean parent ratings for the behavioural
symptoms index and the hyperactivity, withdrawal, and attention problems scales scored in the at-risk range.
Scores between 70 and 79 on the challenging behaviour scales and 20 and 29 on the adaptive skills scales indicate
clinically elevated symptoms (Reynolds & Kamphaus, 2004). No mean teacher ratings were in this range, but the
adaptive skills composite and the atypicality, social skills, functional communication, and activities of daily living scales
scored in this range for mean parent ratings. For both parents and teachers, many scales’ averages were in the normal
range (40–59), including the internalising and externalising composites, and the aggression, anxiety, depression,
somatisation, and adaptability scales. Teachers’ mean scores also placed hyperactivity and attention problems in the
normal range.
1200 B.R. Lane et al. / Research in Autism Spectrum Disorders 7 (2013) 1196–1203

Table 2
Descriptive statistics for parent and teacher ratings on the BASC-2.

Scalea Parent Teacher

M SD M SD r p

Externalising 56.14 9.48 54.20 8.67 .39 .035

Hyperactivity 60.14 15.49 56.64 9.95 .27 .110
Aggression 49.05 8.18 51.50 7.66 .53 .005
Internalising 50.86 11.17 52.64 8.69 .30 .089
Anxiety 47.00 14.35 54.45 9.82 .26 .121
Depression 54.18 12.38 54.91 7.98 .45 .018
Somatisation 52.55 7.79 48.14 6.89 -.08 .365
Behavioural symptoms 66.36 12.58 61.82 12.17 .19 .200
Atypicality 77.36 17.16 66.14 16.93 .47 .013
Withdrawal 67.64 8.87 66.45 14.49 .33 .064
Attention problems 64.23 7.50 58.59 9.04 .26 .121
Adaptive skills 25.59 8.46 36.32 5.89 .41 .030
Adaptability 41.73 9.89 42.91 9.25 .18 .218
Social skills 29.32 9.22 36.18 5.26 .74 .000
Functional communication 27.64 10.08 35.23 5.90 .42 .027
Activities of daily living 24.05 11.91
a
n = 22.

3.2. Parent–teacher correlations

Parent–teacher correlations for the BASC-2 composites were significant only for the externalising composite, r = .39,
p = .035, and adaptive skills composite, r = .41, p = .030. Five of the 11 subscales also had significant parent–teacher
correlations (see Table 2). Most were moderate correlations, with the exception of the social skills scale, r = .74, p = .000, and
the aggression scale, r = .53, p = .005, which both had strong correlations. In contrast to the BASC-2, all of the parent–teacher
correlations for the domains were significant on the VABS-II (see Table 3). Most of the subscale correlations were also
significant, with effects ranging from moderate to strong. In exception were the three subscales of the socialisation domain,
which were all non-significant.
A power analysis (conducted using G*Power; Faul, Erdfelder, Lang, & Buchner, 2007) revealed that the sample size for the
BASC-2 (n = 22) had sufficient power (.8) to detect effects larger than r = .48. For the VABS-II, the sample size (n = 28) was
sufficient to detect effects larger than r = .43.

3.3. Parent–teacher comparisons

Nine t-tests tested for significant differences between the select pairs (see Table 4). To control for family-wise error,
results were considered significant at the .01 level. For the BASC-2, parents and teachers only significantly differed on the

Table 3
Descriptive statistics for parent and teacher ratings on the VABS-II.

Scale n Parent Teacher

M SD M SD r p

Communication 28 71.54 18.05 73.32 15.16 .73 .000

Receptive 27 9.33 3.50 10.11 2.81 .49 .005
Expressive 28 8.46 3.29 8.89 2.56 .74 .000
Written 27 13.56 3.85 12.93 4.08 .68 .000
Daily living 28 72.82 12. 58 72.61 13.69 .68 .000
Personal 28 9.68 2.44 9.68 2.68 .48 .005
Domestic 27 10.89 2.33
Academic 28 11.29 3.16
School community 28 11.50 2.37 11.32 2.92 .38 .024
Socialisation 28 72.71 10.66 69.46 14.52 .43 .012
Interpersonal relationships 28 9.11 1.95 9.29 2.58 .31 .054
Play and leisure time 28 9.57 2.36 9.68 3.07 .30 .059
Coping skills 28 11.82 2.60 10.46 2.52 .25 .099
Motor skills 28 77.21 12.34 72.04 15.34 .70 .000
Gross 28 12.07 1.94 11.04 1.99 .48 .005
Fine 28 10.46 2.55 10.36 3.14 .53 .002
Adaptive behaviour composite 28 70.54 11.40 68.39 14.21 .72 .000
Maladaptive behaviour index 27 19.89 8.64
Externalising 27 6.44 2.82
Internalising 27 7.30 3.73
 B.R. Lane et al. / Research in Autism Spectrum Disorders 7 (2013) 1196–1203 1201

Table 4
Differences between parent and teacher scores on the BASC-2 and VABS-II.

Mean differencea SD t df p dzd

b
BASC-2
Externalising 1.92 10.01 .905 21 .376 .19
Internalising 1.77 11.94 .697 21 .494 .15
Behavioural symptoms 4.55 15.77 1.352 21 .191 .29
Adaptive skills 10.73 8.11 6.206 21 .000 1.33
VABS-IIc
Communication 1.79 12.56 .948 27 .351 .14
Daily living .21 10.57 .143 27 .887 .02
Socialisation 3.25 13.89 1.238 27 .226 .23
Motor skills 5.18 11.04 2.482 27 .020 .47
Composite 2.14 9.98 1.136 27 .266 .22
a
Mean difference is the parent mean minus the teacher mean.
b
n = 22.
c
n = 28.
d
Effect size calculated using G*Power (Faul et al., 2007), dz > .2 is small dz > .6 is moderate, dz > 1.2 is large (Hopkins, 2002).

adaptive skills composite. Teacher ratings (M = 36.32, SD = 5.89) were significantly less severe than parent ratings (M = 25.59,
SD = 8.46), t (21) = 6.21, p = .000, dz = 1.33. For the VABS-II, no comparisons reached significance at the specified .01 level.
A power analysis, using a .01 alpha level, revealed that the BASC-2 sample size had sufficient power (.8) to detect
differences larger than dz = .72. The VABS-II sample size had sufficient power (.8) to detect differences with effect sizes larger
than dz = .63.
Correlations between the difference scores showed that, for the BASC-2, the degree of parent–teacher differences on one
composite generally corresponded to the degree of difference on other composites, especially for the challenging behaviour
composites (r = .62–.74) compared with the adaptive skills composite (r = .26–.29). Similar results appeared for the
overlapping VABS-II composites (r = .31–.86). For the overlapping difference scores between the BASC-2 and VABS-II, only
the VABS-II communication and BASC-2 adaptive skills composite correlation was significant (r = .52, p = .013).

4. Discussion

This study aimed to compare parent and teacher ratings of young children with ASDs on the BASC-2 and VABS-II. Different
patterns emerged for parent–teacher ratings for these instruments. For the BASC-2, only the difference for adaptive skills
reached significance, contrasting with past research finding differences for all composites (Barnhill et al., 2000; Foley Nicpon
et al., 2010). On average, parents rated the children’s adaptive skills in the clinically elevated range, while teachers rated the
children in the at-risk range. The significant parent–teacher correlation further suggested this tendency was systematic. This
effect was present for the social skills and functional communication scales, which also attracted comparatively high parent–
teacher correlations compared to the challenging behaviour scales. However, the adaptability scale showed little difference
between parents and teachers, while also having a non-significant correlation. The previous studies for parent and teacher
ratings on the BASC did not report correlation statistics, preventing a comparison. Thus, this is a novel result.
It is interesting that, in contrast to the BASC-2 adaptive skills ratings, no VABS-II adaptive scales showed significant
differences between the parents and teachers, despite ratings being on most of the same children. If an effect of a comparable
size to the BASC-2 results did exist in this population, the sample size was sufficient to detect it. This may reflect differences
in the items used to operationalise adaptive skills in each measure for those who score low and warrants further
investigation. This result contrasts with past research using the original VABS (Hundert et al., 1997; Szatmari et al., 1994;
Voelker et al., 2000), finding that parent ratings tended to be systematically of greater severity. This supports Voelker et al.’s
argument that scoring artefacts on the original VABS explained past findings of consistently more severe parent ratings.
Revisions to the VABS-II included additional items on the development of language, correcting the lower absolute limit for
these ratings (Sparrow et al., 2005). Similar to the BASC-2, the VABS-II adaptive scales had significant parent–teacher
correlations, which corresponds with past research (Hundert et al., 1997; Szatmari et al., 1994; Voelker et al., 2000).
In contrast to adaptive behaviours, parent–teacher correlations for challenging behaviours on the BASC-2 were mostly
small to moderate. Whereas the aggression, depression, and atypicality scales and overall externalising composite had
moderate to large and significant correlations, the hyperactivity, anxiety, somatisation, withdrawal, and attention problems
scales and internalising composite and behavioural symptoms index had non-significant correlations. The fact that the
internalising scales require the observer to make inferences about the child may explain lower correlations within this
domain (Achenbach, 2011). In addition, some items on these scales rely on verbal abilities (e.g., ‘‘complains about health’’ or
‘‘complains of being cold’’), which can be reduced in this population (American Psychiatric Association, 2000; Mahan &
Matson, 2011). However, reasons for the low behavioural symptoms correlation are unclear. It is notable that this study’s
BASC-2 scale counterparts to the CRS-R (attention problems and hyperactivity) attracted similarly low correlation statistics
to those reported in past research (Reed & Osborne, 2013). This suggests that parent–teacher rating discordance for ADHD
symptoms are a robust finding for young ASD populations.
1202 B.R. Lane et al. / Research in Autism Spectrum Disorders 7 (2013) 1196–1203

Finding many low parent–teacher correlations for challenging behaviours suggests that there are differences between
parents and teachers, but no systematic factor guiding these differences. In contrast, parent–teacher ratings of adaptive skills
on both the BASC-2 and VABS-II had comparatively higher correspondence. This could suggest that different environments
consistently elicit similar adaptive skills, making the relative abilities of the children clearer. In contrast, challenging
behaviours are likely more situation-driven. A situation characteristic in this study is that children enrolled in the AEIOU
program interact with other children with ASDs. Thus, the teacher respondents observed these children in this special, social
context where the behaviour of one child potentially affected the behaviours of others around them (Farmer & Cadwallader,
2000). Research has also recognised communication difficulties to complicate behaviour problems, as children with
difficulties may not be able to convey needs effectively to each other (Carpenter & Drabick, 2011). In the home context,
complications would be less likely as whereas other children with ASDs may ignore or respond inappropriately to social
advances, parents and siblings may be more likely to respond or reward such overtures. Another notable implication of these
environments is that the teachers are likely exposed to a wider variety of ASD cases than are parents, which could influence
perceptions (Reed & Osborne, 2013). This would be especially relevant if the parent has no other children for comparison
(Barnhill et al., 2000). Indeed, the results showed that teachers rated atypicality (the presence of odd behaviour) in the at-risk
range, but parents rated it in the clinical range. The children’s behaviour likely does not seem as peculiar to teacher
respondents compared with parent respondents.
A limitation of this study requiring consideration is sample size. Power analysis revealed a high probability of Type II errors
when evaluating moderate correlations. For some of the BASC-2 parent–teacher correlations with moderate effect sizes, the
sample size was too small to evaluate significance accurately. Despite this limitation, the study did have sufficient power to
detect correlations with large effect sizes. Thus, we can be confident that the parent–teacher correlations were genuinely higher
for the adaptive compared to challenging behaviours scales. For evaluating mean parent–teacher differences, the study had
sufficient power to detect most moderate and large effect sizes. No differences had non-significant moderate or large effect
sizes, but one (differences in motor skills ratings) approached a moderate effect size and significance. Thus, although our
findings suggest no parent–teacher differences, there is an inflated chance of a Type II error and we cannot confidently conclude
no systematic differences exist. There is clearly a need for further research with a larger sample size.
Similar to past research, this study accessed a special subset of the ASD population—those enrolled in an early
intervention program. As measures were of direct relevance to treatment, parents and teachers were likely motivated to
complete the scales mindfully. However, it should be noted that this sample excludes many who are not diagnosed or
receiving other intervention services. There may be characteristics of such individuals that encourage such exclusions, which
may affect results (Reed & Osborne, 2013). For instance, those with low socioeconomic statuses who cannot afford services
are perhaps more stressed, which has a suggested link with parent–teacher differences (e.g., Szatmari et al., 1994). In
addition, a relatively high proportion of parents identified culturally and linguistically diverse backgrounds. This figure is
perhaps unsurprising considering that in Queensland, 39% of residents have at least one parent born overseas (Australian
Bureau of Statistics, 2011). However, taken together, these factors highlight that this study’s high parent–teacher
correspondence for adaptive behaviour ratings is potentially unique to the sample’s context. Despite these limitations, using
parent and teacher ratings on two measures provided greater depth of information and allowed control of participant
characteristics that would usually make comparisons between studies difficult. Furthermore, past research has sometimes
used general education teachers who may lack familiarity with the child or with children with ASDs in general, or sufficient
time to complete sizable assessment forms such as the VABS-II or BASC-2.
It is important for assessments to maximise unique information gained about the child, while minimising repetition and
use of unnecessary measurement tools to maximise resources and minimise distress and burden on families and teachers.
Many of the challenging behaviour scales on the BASC-2 attracted relatively low parent–teacher correlations, compared with
high correlations on the VABS-II. This suggests that parents and teachers provide more unique information on the BASC-2 in
contrast to greater concordance on the VABS-II. An exception is socialisation on the VABS-II, with reasonable discordance
appearing between the respondents. These results build on conclusions in past research that suggested caution is necessary
when interpreting challenging behaviour assessments (Reed & Osborne, 2013). Furthermore, they affirm calls for greater
consideration of behaviour in the wider psychosocial context (i.e., with multiple rating sources; Hartley et al., 2011). The
results also suggest that the updated VABS-II has overcome limitations present in the scoring of the original VABS (Voelker
et al., 2000). Although statistical power limited the results and there is thus a need for further research and replication with
larger samples, our results provide initial evidence that parent–teacher correlations for adaptive behaviours are higher than
for challenging behaviours. The implication of this finding is that multiple informants are likely useful and necessary when
assessing challenging behaviours (using a system such as the BASC-2). For adaptive behaviours, it may be important to gain
multiple informant ratings when determining initial eligibility to a program, in terms of establishing whether there are
divergent ratings from onset. If initial ratings are similar, it may be feasible to rely on a single informant to track further
progress and aid goal setting.

Acknowledgements

The authors would like to acknolwedge the work of the AEIOU foundation and their commitment to research by allowing
us access to their data. Thank you also to the Australian Government Department of Housing, Community Services, and
Indigenous Affairs (FaHCSIA) for their financial support provided to the AEIOU program.
 B.R. Lane et al. / Research in Autism Spectrum Disorders 7 (2013) 1196–1203 1203

References

Achenbach, T. M. (2011). Commentary: Definitely more than measurement error: But how should we understand and deal with informant discrepancies? Journal
of Clinical Child and Adolescent Psychology, 40, 80–86 http://dx.doi.org/10.1080/15374416.2011.533416
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text revision). Washington, DC: American Psychiatric
Association.
Australian Bureau of Statistics. (2011). Census: Data & analysis Retrieved May 30, 2013, from http://www.abs.gov.au/census.
Barnhill, G. P., Hagiwara, T., Myles, B. S., Simpson, R. L., Brick, M. L., & Griswold, D. E. (2000). Parent, teacher, and self-report of problem and adaptive behaviors in
children and adolescents with Asperger syndrome. Assessment for Effective Intervention, 25, 147–167 http://dx.doi.org/10.1177/073724770002500205
Carpenter, J. L., & Drabick, D. A. G. (2011). Co-occurrence of linguistic and behavioural difficulties in early childhood: A developmental psychopathology
perspective. Early Child Development and Care, 181, 1021–1045 http://dx.doi.org/10.1080/03004430.2010.509795
Conners, C. K. (1997). Conners’ parent rating scale. NJ: Pearson Assessment.
Farmer, T. W., & Cadwallader, T. W. (2000). Social interactions and peer support for problem behavior. Preventing School Failure, 44, 105–109 http://dx.doi.org/
10.1080/10459880009599791
Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences.
Behavior Research Methods, 39, 175–191 http://dx.doi.org/10.3758/BF03193146
Foley Nicpon, M., Doobay, A. F., & Assouline, S. G. (2010). Parent, teacher, and self perceptions of psychosocial functioning in intellectually gifted children and
adolescents with autism spectrum disorder. Journal of Autism and Developmental Disorders, 40, 1028–1038 http://dx.doi.org/10.1007/s10803-010-0952-8
Geiger, D. M., Smith, D. T., & Creaghead, N. A. (2002). Parent and professional agreement on cognitive level of children with autism. Journal of Autism and
Developmental Disorders, 32, 307–312 http://dx.doi.org/10.1023/a:1016382819186
Hartley, A. G., Zakriski, A. L., & Wright, J. C. (2011). Probing the depths of informant discrepancies: Contextual influences on divergence and convergence. Journal of
Clinical Child and Adolescent Psychology, 40, 54–66 http://dx.doi.org/10.1080/15374416.2011.533404
Hattier, M. A., Matson, J. L., Belva, B. C., & Adams, H. (2013). The effects of informant age and education level on childhood ASD symptom endorsement. Research in
Autism Spectrum Disorders, 7, 398–404 http://dx.doi.org/10.1016/j.rasd.2012.11.001
Havlicek, L. L., & Peterson, N. L. (1977). Effect of the violation of assumptions upon significance levels of the Pearson r. Psychological Bulletin, 84, 373–377 http://
dx.doi.org/10.1037/0033-2909.84.2.373
Hopkins, W. G. (2002). A scale of magnitudes for the effect statistics: In a new view of statistics Retrieved May 30, 2013, from http://www.sportsci.org/resource/stats/
effectmag.html.
Hundert, J., Morrison, L., Mahoney, W., & Vernon, M. L. (1997). Parent and teacher assessments of the developmental status of children with severe, mild/
moderate, or no developmental disabilities. Topics in Early Childhood Special Education, 17, 419–434 http://dx.doi.org/10.1177/027112149701700404
Kalyva, E. (2010). Multirater congruence on the social skills assessment of children with Asperger syndrome: Self, mother, father, and teacher ratings. Journal of
Autism and Developmental Disorders, 40, 1202–1208 http://dx.doi.org/10.1007/s10803-010-0978-y
Kanne, S. M., Abbacchi, A. M., & Constantino, J. N. (2009). Multi-informant ratings of psychiatric symptom severity in children with autism spectrum disorders: The
importance of environmental context. Journal of Autism and Developmental Disorders, 39, 856–864 http://dx.doi.org/10.1007/s10803-009-0694-7
Lecavalier, L., Leone, S., & Wiltz, J. (2006). The impact of behaviour problems on caregiver stress in young people with autism spectrum disorders. Journal of
Intellectual Disability Research, 50, 172–183 http://dx.doi.org/10.1111/j.1365-2788.2005.00732.x
Lee, L. C., David, A. B., Rusyniak, J., Landa, R., & Newschaffer, C. J. (2007). Performance of the Social Communication Questionnaire in children receiving preschool
special education services. Research in Autism Spectrum Disorders, 1, 126–138 http://dx.doi.org/10.1016/j.rasd.2006.08.004
Lord, C., Wagner, A., Rogers, S., Szatmari, P., Aman, M., Charman, T., et al. (2005). Challenges in evaluating psychosocial interventions for autistic spectrum
disorders. Journal of Autism and Developmental Disorders, 35, 695–708 http://dx.doi.org/10.1007/s10803-005-0017-6
Magiati, I., Tay, X. W., & Howlin, P. (2012). Early comprehensive behaviorally based interventions for children with autism spectrum disorders: A summary of
findings from recent reviews and meta-analyses. Neuropsychiatry, 2, 543–570 http://dx.doi.org/10.2217/npy.12.59
Mahan, S., & Matson, J. L. (2011). Children and adolescents with autism spectrum disorders compared to typically developing controls on the Behavioral
Assessment System for Children, Second Edition (BASC-2). Research in Autism Spectrum Disorders, 5, 119–125 http://dx.doi.org/10.1016/j.rasd.2010.02.007
Matson, J. L., & Jang, J. (2013). Autism spectrum disorders: Methodological considerations for early intensive behavioral interventions. Research in Autism Spectrum
Disorders, 7, 809–814 http://dx.doi.org/10.1016/j.rasd.2013.01.006
Matson, J. L., Nebel-Schwalm, M., & Matson, M. L. (2007). A review of methodological issues in the differential diagnosis of autism spectrum disorders in children.
Research in Autism Spectrum Disorders, 1, 38–54 http://dx.doi.org/10.1016/j.rasd.2006.07.004
Murphy, O., Healy, O., & Leader, G. (2009). Risk factors for challenging behaviors among 157 children with autism spectrum disorder in Ireland. Research in Autism
Spectrum Disorders, 3, 474–482 http://dx.doi.org/10.1016/j.rasd.2008.09.008
Murray, D. S., Ruble, L. A., Willis, H., & Molloy, C. A. (2009). Parent and teacher report of social skills in children with autism spectrum disorders. Language, Speech,
and Hearing Services in Schools, 40, 109–115 http://dx.doi.org/10.1044/0161-1461(2008/07-0089)
Ozonoff, S., Goodlin-Jones, B. L., & Solomon, M. (2005). Evidence-based assessment of autism spectrum disorders in children and adolescents. Journal of Clinical
Child and Adolescent Psychology, 34, 523–540 http://dx.doi.org/10.1207/s15374424jccp3403_8
Paul, R., Miles, S., Cicchetti, D., Sparrow, S., Klin, A., Volkmar, F., et al. (2004). Adaptive behavior in autism and pervasive developmental disorder-not otherwise
specified: Microanalysis of scores on the Vineland Adaptive Behavior Scales. Journal of Autism and Developmental Disorders, 34, 223–228 http://dx.doi.org/
10.1023/B:JADD.0000022612.18116.46
Paynter, J., Scott, J., Beamish, W., Duhig, M., & Heussler, H. (2012). A pilot study of the effects of an Australian centre-based early intervention program for children
with autism. Open Pediatric Medicine Journal, 6, 7–14, Retrieved May 30, 2013, from http://www.benthamscience.com/open/topedj/articles/V006/
7TOPEDJ.pdf.
Prior, M., Roberts, J. M. A., Rodger, S., Williams, K., & Sutherland, R. (2011). A review of the research to identify the most effective models of practice in early intervention
for children with autism spectrum disorders. Australian Government Department of Families, Housing, Community Services and Indigenous Affairs Retrieved
May 30, 2013, from http://www.fahcsia.gov.au/sites/default/files/documents/09_2012/review_of_the_research_report_2011_0.pdf.
Reed, P., & Osborne, L. A. (2013). The role of parenting stress in discrepancies between parent and teacher ratings of behavior problems in young children with
autism spectrum disorder. Journal of Autism and Developmental Disorders, 43, 471–477 http://dx.doi.org/10.1007/s10803-012-1594-9
Reynolds, C. R., & Kamphaus, R. W. (2004). Behavior assessment system for children. American Guidance Services: Circle Pines, MN.
Rutter, M., Bailey, A., & Lord, C. (2003). The social communication questionnaire: Manual. Los Angeles, CA: Western Psychological Services.
Sparrow, S. S., Cicchetti, D. V., & Balla, D. A. (2005). Vineland adaptive behavior scales (2nd ed.). Minneapolis, MN: NCS Pearson Inc.
Szatmari, P., Archer, L., Fisman, S., & Streiner, D. L. (1994). Parent and teacher agreement in the assessment of pervasive developmental disorders. Journal of Autism
and Developmental Disorders, 24, 703–717 http://dx.doi.org/10.1007/BF02172281
Voelker, S. L., Shore, D. L., Lee, C. H., & Szuszkiewicz, T. A. (2000). Congruence in parent and teacher ratings of adaptive behavior of low-functioning children. Journal
of Developmental and Physical Disabilities, 12, 367–376 http://dx.doi.org/10.1023/a:1009436230984
Youngstrom, E., Loeber, R., & Stouthamer-Loeber, M. (2000). Patterns and correlates of agreement between parent, teacher, and male adolescent ratings of
externalizing and internalizing problems. Journal of Consulting and Clinical Psychology, 68, 1038–1050 http://dx.doi.org/10.1037//0022-006X.68.6.1038