Printer Friendly

The real-world effectiveness of early teaching interventions for children with autism spectrum disorder.

Educationally based programs are often the only form of intervention given to children with autism spectrum disorder (ASD) and are often the only form of support that their families receive (e.g., see Howlin & Moore, 1997; Osborne & Reed, in press, for surveys). Often these interventions are offered early in the child's development, and there is a perception that interventions offered early are more effective than those offered later (Lovaas, 1987). However, despite the ubiquity and perceived importance of educational approaches to the management of ASD, there are relatively few quantitative investigations of the comparative efficacy of such programs.

A number of early-teaching interventions have been suggested as offering benefit to some children with ASD. Applied behavior analysis (ABA) approaches have been widely investigated (e.g., Eikeseth, Smith, Jahr, & Eldevik, 2002; Howard, Sparkman, Cohen, Green, & Stanislaw, 2005; Lovaas, 1987; Smith, Annette, & Wynn, 2000). In addition, special nursery placements (see Charman, Howlin, Berry, & Prince, 2004; Gabriels, Hill, Pierce, Rogers, & Wehner, 2001), floortime (Greenspan & Wieder, 1997), and portage (e.g., Smith, 2000) have been offered as approaches to managing the problems associated with ASD. Some of these latter approaches have the benefit of being reasonably cost effective, but they suffer from the problem of having a scant evidence base with respect to their effectiveness in promoting the child's intellectual, educational, and social functioning.

The effectiveness of placement in specialist nursery provision has been studied, often as control conditions in the assessment of ABA programs. Such programs are often referred to in this literature as "generic" (e.g., Lovaas, 1987) and/or "eclectic" (e.g., Howard et al., 2005) programs. This is typically because such approaches do not follow a particular "brand name" approach to education, and they offer a variety of teaching methods (e.g., picture exchange systems, activities drawn from the TEACCH [Treatment and Education of Autistic and Related Communication Handicapped Children] model, circle time, social stories) that do not necessarily adhere to a single underlying philosophy, as do ABA approaches. The results from these studies of special nursery placements have not been particularly encouraging. Three studies have shown small improvements in intellectual functioning following special educational placement. Sheinkopf and Siegel (1998) found a 2-point increase in IQ across a year in children in special school provision, and Eikeseth et al. (2002) found a 4-point improvement in IQ over a year (using one-to-one techniques such as TEACCH and sensory-integration therapy). However, there was no improvement in adaptive behavior (as measured by the Vineland Adaptive Behaviour Scale). Gabriels et al. (2001), in a retrospective, 3-year study of children with ASD undergoing an eclectic approach to intervention, found a 5-point increase in IQ over 3 years.

In contrast, four studies have found no improvement in functioning. Shallows and Graupner (1999) found a 7-point decrease in IQ over a year. The remaining three studies in this latter group found no overall change in the standardized score of children in special educational placements with an eclectic approach to intervention. Howard et al. (2005) compared two groups of children with ASD undergoing either an intensive program in an autism special school or a less intensive program in a local community special education classroom. In neither case was there any change in a variety of assessments of intellectual functioning. Lovaas (1987) found no improvement for children undergoing generic educational interventions over a 3-year period. Charman et al. (2004) found no improvement in adaptive behavior scores as measured by the Vineland Adaptive Behaviour Scale (see also Eikeseth et al., 2002).

Unfortunately, the usefulness of many of these studies in assessing the impact of special educational provision on children with ASD is limited by the fact that many of these groups are derived from archive records (e.g., Gabriels et al., 2001; Sheinkopf & Siegel, 1998), rather than being studied contemporaneously with the other children in the study. Additionally, many of these studies employ different measurements both across participants and at baseline and follow up (e.g., Howard et al., 2005; Lovaas, 1987), thus, reducing the reliability of the data (see Magiati & Howlin, 2001). There are few studies of special educational placement that are prospective in nature and that use the same tests at baseline and follow up, which was the first aim of the current study.

Portage is an approach offered by numerous local education authorities in the United Kingdom for children with ASD. Portage is a home-based, parent-administered early intervention for developmental disabilities. Smith (2000) provides some data to suggest that portage programs can have an impact on social development of children with autism, and Reed, Osborne, and Gibson (2000) discuss some possible developments and limitations of this approach. However, this approach, although commonly used in practice, suffers from a great paucity of empirical evidence regarding its effectiveness.

Much current debate in educational literature has centered on ABA techniques as an intervention for ASD problems (see Connor, 1998; Mudford, Martin, Eikeseth, & Bibby, 2001). There are a number of different ABA approaches that have been outlined in a variety of sources (e.g., Greer, 1997; Koegel & Koegel, 2006; Lovaas, 1981; Lovaas & Smith, 1989; Sundberg & Michael, 2001; Twyman, 1998). In general, these approaches share a number of common features: (a) the one-to-one teaching of children with autism by adult tutors; (b) a discrete-trial reinforcement-based method; and (c) an intensive regime (up to 40 hr a week, for 3 years, in some instances). The initial results reported by Lovaas (1987) concerning the effectiveness of such an approach were remarkable. The children undergoing this approach made gains of up to 30 IQ points (a finding noted in some children with ASD undergoing special educational programs, see Gabriels et al., 2001), and just less than half of these children appeared to recover, that is, they were not noticeably different from normally developing children after 3 years of the intervention (a finding yet to be replicated).

There have been a number of critiques of this original piece of research (i.e., Lovaas, 1987) that have focused on problems both with the internal and external validity of the study (e.g., Connor, 1998; Gresham & MacMillan, 1997; Mudford et al., 2001). In terms of the internal validity of the Lovass (1987) study, it should be noted that different IQ tests were used at baseline and at follow up to assess the children's intellectual functioning, thus reducing the reliability of the measurement (Magiati & Howlin, 2001). This is not a unique problem with ABA studies (see earlier text). In terms of the threats to the external validity of the Lovaas (1987) study, the reliance on IQ as a sole measure may be questioned, given that IQ is not necessarily the main problem in autism functioning. Second, the sample chosen for the study reported by Lovaas (1987)were verbal, relatively high-functioning participants, who may have performed equally well with any intervention of a reasonable input (such as some of those noted earlier). Finally, the study reported by Lovaas (1987) was a clinic-based study and may not generalize to interventions as they are typically used in the parents' home, severely compromising the usefulness of the study. This issue has been highlighted in a number of recent reports of home-based ABA programs, which have questioned the extent to which the results of clinic-based assessments can be generalized to the type of ABA program that occurs in the community (Mudford et al.).

Although several of these critiques have been addressed in subsequent research, few, if any, studies have addressed them in a single investigation. For example, Eikeseth et al. (2002), and Howard et al. (2005), have used community-based rather than clinic-based samples, but neither used the same tests at baseline and at follow up. In light of this, a second aim of the current study was to assess the effectiveness of ABA approaches on a more typical sample of children with ASD, using the same instruments at baseline and follow up.

Thus, there are a number of issues that are clearly unresolved concerning the effectiveness of educational interventions for ASD. It appears critical to be able to assess the effectiveness of the ABA approach in community-based, rather than clinic-based, settings with participants more typically representative of those who present to local service providers (i.e., more severe symptoms, less linguistically able). It also appears important to utilize a wide range of instruments in the assessment procedure, not only to examine intellectual functioning, but also educational functioning, and adaptive behavior. Simultaneously providing evidence on the effectiveness of other interventions than ABA would not only allow these interventions to be assessed, but also would allow a well-matched alternate-treatment control group for the ABA studies, a comparison so far missing (Hohmann & Shear, 2002).

Given the previous considerations, the current study directly compared the impact of existing ABA, special nursery placements, and portage programs on a variety of aspects of the children's abilities. The latter two were selected because special nursery placement is a commonly occurring program offered to children with ASD, which has received little direct assessment in terms of its effectiveness. Portage was chosen as, again, it is increasingly offered to children with ASD (see Reed et al., 2000; Smith, 2000). The portage intervention also allows comparison of a very intensive intervention (ABA) with a less intensive intervention (portage) in a community-based setting. This comparison formed part of the original clinic-based study conducted by Lovaas (1987), and the current comparison allows assessment of the generalization to a community-based sample.



Participants were selected on the basis of four criteria. They were (a) 2:6 (2 years 6 months) to 4:0 years old, (b) at the start of their first teaching intervention, (c) receiving no other major intervention during the period of the assessment, and (d) they had an independent diagnosis of ASD made by specialist pediatricians following initial referral from a general medical practitioner. In addition to this independent diagnosis, several of the scales that were administered allowed further independent assessment of the degree of the autism severity (Gilliam Autism Rating Scale [GARS]; Gilliam, 1995) and the degree of behavioral problems (Conners' Rating Scale [CRS-R]; Conners, 1997).

A total of 53 participants were identified in conjunction with local authorities in the southeast of England during 2002 to 2004. None of the families contacted refused to participate in the study. Of the participants initially recruited, five were excluded from the study: three for compromised treatment integrity (i.e., at the follow-up assessment it became apparent that the child had received both ABA and special nursery interventions in two cases, and portage and special nursery interventions in one case); one for missing data; and one for highly discrepant GARS scores at baseline and ceiling effects on follow up).

Assignment to a group was on the basis of the intervention being offered to the child in their particular area. For example, if a child was in an area that offered a special nursery placement, then that child was assigned to that group. The areas involved in the study offered a similar socioeconomic profile, all being in southeast England. Thus, although the allocation to group was not truly random, the child's characteristics did not influence group assignment, and the groups offered a well matched profile (see Results and Table 2). Ethical approval for the study (University College London Hospital Trust Ethics Committee) was granted based on this understanding.

In terms of the estimated sample size required for finding statistical effects, previous studies have employed between 11 and 19 participants per group. Given the average effect size in these studies of about 1.0, the minimum number of participants for a study with high power (0.8) would be approximately 8. Given this, the number of participants per group (12 to 20) was thought to be adequate in this study.


Three early educational interventions for autism were studied (i.e., they were not part of a specially organized trial)--applied behavior analysis (ABA), special nursery placements, and portage. These interventions were selected as they represent some of the most commonly occurring community-based interventions, and they provided a broad spectrum of the types of approach currently offered in the United Kingdom.

Applied Behavior Analysis. The ABA programs included in this study were provided by a range of organizations, and followed one of a number of well recognized ABA procedures: (a) UCLA Lovaas-type approaches (see Lovaas, 1981; Lovaas & Smith, 1989); (b) verbal behavior (see Sundberg & Michael, 2001); or (c) Comprehensive Application of Behavior Analysis to Schooling (CABAS; see Greet, 1997). All of these programs were overseen by appropriately trained supervisors (who were either Board Certified Behavior Analyst [BCBA] accredited or who had master teacher-level CABAS qualifications). The interventions were conducted by tutors trained in accordance with the appropriate intervention manuals associated with the ABA approach offered.

All of these programs shared key ABA features. All were home-based and offered almost exclusively one-to-one teaching for the child with autism, and the intensity (hours per week) of the interventions were typically quite high (see Table 1). Typically, a session would last 2 to 3 hr, and would comprise approximately 8 to 14 tasks or drills per session (depending on the particular needs of the child). Typically these tasks would last about 5 to 10 min each and would be repeated until some criterion performance was reached. Each task would be separated by a 5- to 10-min break, or down time. The programs used an antecedent (question/task), behavior (response) sometimes prompted if necessary, and consequence procedure as outlined in the various manuals. Reinforcement was usually a tangible such as food, but could also be praise and activities depending on what was effective with the individual child. No aversive stimuli were used in any of the programs.

Special Nursery Placements. The special nursery placements (Nursery) occurred across several school provisions (see Table 1). The classes were small, with about six to eight children in each. All classes were under the supervision of a teacher with postgraduate qualifications in teaching and specialist training in special educational needs. All curriculum and practices had been approved by Ofsted Reports (U.K. government inspection reports given regularly to all schools). In addition to the teacher, each class had two or three learning support assistants, who would help work with the children in small groups. Thus, most teaching was conducted in small groups rather than individually (about four times as much group work as individual work). The intensity of the intervention (hours per week) was moderate compared to the ABA group.

The children attended the nursery for a number of 2- to 3-hr sessions per week, depending on the severity of the child's autism (see Table 1 for the range of time inputs). Typically, a session would start and end with six to eight children in a group with the teacher at the front. The teacher usually guided a song, or other introduction, and the children were encouraged to take turns in answering their names or responding, often involving doing an individual activity (e.g., picking up name card, shaking an instrument, etc.), while the others were encouraged to respond and comment. A key feature was the use of materials and methods appealing to children with ASD, such as bright colored visual materials, glitter, water, paint, sand, or musical instruments. Much of the school's environment and many of the tasks given to the children were presented in a highly structured method as outlined by the TEACCH methodology (Mesibov, Schopler, & Hearsey, 1994).

Thus, the focus was on learning and motivation, developing shared attention, and participation (with support from up to two or three other adults). Other sessions included a free choice of activities for the children, with adults focusing their support on children working toward their learning objectives, all the time encouraging and prompting social interaction, turn-taking, shared attention, and responding. Another session involved outdoor group activities, such as running across the playground, together in pairs, to practice physical coordination, group participation, and so forth. Most of the reinforcers were social (e.g., saying "good boy," or clapping, or providing the opportunity to play with a desired toy, etc., but always in the context of the group).

Portage. Portage is a home-based teaching program for preschool children with special educational needs. The portage program has been extensively used with children with developmental delay (see Cameron, 1997), but has recently been developed to accommodate children with ASD (see Reed et al., 2000; Smith, 2000). This program involves slightly increased time input, and a focus on developing attentional responses, such as joint attention and attention sharing behaviors. The intervention is a relatively low-intensity program, with the majority of the work being conducted one-to-one with the child, although about 25% of the time involves group work (see Table 1).

The portage program is supervised by a trained portage supervisor, who has graduate-level qualifications and who attended training courses in the delivery of portage for children with ASD. The program typically follows a manual written by the portage service provider. The supervisor visits the parents once every 1 to 2 weeks, and parents and caregivers are shown how to apply this system during these visits.

The training sessions conducted with the child are brief, usually about 40 to 60 min per day, and are scheduled when the parent believes the child will be at his or her most receptive. Thus, the program shares some common elements with more motivationally oriented approaches such as floortime (Greenspan & Wieder, 1997). Typically, the parent will teach the child in a one-to-one situation and will target several skills per week. The children are taught new skills through the use of questions and tasks, prompts, and rewards. The skills to be taught are outlined in the portage manuals, and each skill acquired is used to build larger and more complex behavioral responses. Monitoring and evaluation of progress occurs at the supervisor's visits.

Comparison of Approaches. The key characteristics of the different interventions, along with a description of their main features (overall hours per week, hours of one-to-one and group, number of tutors involved, etc.) reported by tutors in pre- and poststudy questionnaires are shown in Table 1. These questionnaires were given at the start of the intervention and at the end and were compared to check the degree to which the planned intervention was adhered to over the course of the program. In all cases, the descriptions of the program given in the postintervention questionnaires were used for the data in Table 1, but it should be noted that in most cases the postintervention descriptions adhered to those outlined at the start of the intervention. Similarly, the descriptions of the program given by the supervisors and the parents were very similar to one another. To this extent, the data obtained on the characteristics of the intervention were verifiable from multiple sources, and fidelity to the initial aims of treatment, insofar as it can be achieved in home-based settings, was reasonable.

Thus, inspection of Table 1 shows that the ABA program had the highest time intensity, followed by nursery, and portage. The ABA and portage approaches delivered mainly one-to-one teaching, whereas the nursery placements offered mainly group instruction. All of the programs offered exposure to a range of adult tutors in addition to the teacher/supervisor. It should be noted that all of the supervisors/teachers leading the various programs were professionally accredited. That is, the ABA supervisors were all BCBA or CABAS accredited, the nursery teachers were all certified teachers, and the portage home visitors had all been trained and held Portage certificates. Each intervention was conducted in accordance to either the appropriate manual or in accordance with best teaching practice in special schools.


Gilliam Autism Rating Scale. The GARS (Gilliam, 1995) comprises four subtests, each describing behaviors symptomatic of autism (Stereotyped Behaviors, Communication, Social Interaction, and Developmental Disturbances). The raw scores from these subscales can be converted into standard scores (M = 100, SD = 15). These subscales combine to give an Autism Quotient, high scores meaning greater autistic severity (M = 100 [average autistic severity], SD = 15). The scale is appropriate for persons ages 3 to 22 years and is completed by parents or professionals in about 10 min. Its internal reliability is 0.96, and it has high criterion validity with the Autism Behavior Checklist (0.94).

Psychoeducational Profile (Revised). The Psychoeducational Profile Revised (PEP-R; Schopler, Reichler, Bashford, Lansing, & Marcus, 1990) is a developmental test designed for assessing both the typical strengths and characteristic weaknesses of children with autism. The test measures functioning in seven developmental domains: Imitation, Perception, Fine and Gross Motor Skills, Eye-Hand Coordination, and Nonverbal and Verbal Conceptual Ability. The mental age required to perform the tests ranges from 1 to 72 months. The test also gives an overall developmental functioning score that can be converted into an overall score (mental age/chronological age x 100). The internal reliability of the PEP-R for children with autism ranges from 0.85 (Perception) to 0.98 (Cognitive Verbal Performance), and it has high criterion validity with other tests for intelligence, such as the Merrill Palmer Scale of Mental Tests (0.85). The PEP-R was used because its low floor and high ceiling made it appropriate to administer to the current sample both at baseline and at follow up. This avoids one major criticism of previous work that different tests are administered at these times reducing the reliability of the measure (Magiati & Howlin, 2001).

British Abilities Scales II. The British Abilities Scales: Second Edition (BAS II; Elliott, Smith, & McCulloch, 1996) is a battery of tests of cognitive abilities that index educational achievement. It is suitable for use with children and adolescents ages 2 years, 6 months (2:6) to 17 years, 11 months (17:11). For the current purposes, the Early Years Battery was employed, which is designed for children under the age of 6 years. The present use of the test concerned educational achievement, so the Verbal Comprehension, Early Number Concepts, Picture Matching, and Naming Vocabulary subscales were used. These scales allow the calculation of a General Cognitive Ability scale (M = 100, SD = 15), which represents early educational achievement.

Vineland Adaptive Behavior Scales. The Vineland Adaptive Behavior Scales (VAB; Sparrow, Balla, & Cicchetti, 1984) is a semistructured interview, administered to a parent or other caregiver of the child. It can be used for children from birth to 5:11 years, making it suitable for the present cohort. The VAB scales assess children's day-to-day adaptive functioning. Scores from four domains of adaptive behavior were used in the present study (Communication, Daily Living Skills, Socialization, and Motor Skills). The raw scores can be converted to standard scores, and a Composite Overall score can be derived based on sum of scale standard scores (M = 100; SD = 15). The internal reliability of the Overall Composite score is 0.93.

Conners' Rating Scales--Revised. The Conners' Rating Scales--Revised (CRS-R; Conners, 1997) assess children (ages 3-17) for behavioral problems and attention deficit/hyperactivity disorder. The short version of the form was used, which takes 5 to 10 minutes to complete. The short version of the CRS-R comprises 28 items for the teacher, using a 0 (Not True at All) to 4 (Very Much True) Likert-type scale. There are four subscales of the test: (a) Oppositional Behavior--indicative of rule breaking, authority problems, and ease of anger; (b) Cognitive Problems--indexing inattention, difficulty in organizing work, or concentrating for sustained periods; (c) Hyperactivity--suggesting difficulties in sitting still, restlessness, and impulsivity; and the (d) ADHD index--identifying children likely to suffer from ADHD. The range of internal reliability of the subscales is from 0.77 to 0.96. Although not used primarily for autism per se, this tool has been identified as important in the assessment of the effects of interventions on behavioral difficulties for children with autism (see, Handen, Johnson, & Lubetsky, 2000).


The children were visited by an experienced and qualified senior educational psychologist (10 years postqualification), who completed the PEP-R and BAS II assessments. These first set of measures took about 120 to 180 min to complete, and all the measures were completed in a single session. In addition, the educational psychologist helped the parent to complete the GARS and VAB measures during this visit. After 9 months, the follow-up measures were taken by the same educational psychologist, who again completed all of the measures as previously described. At this time, the educational psychologist also helped the parent to complete the follow-up GARS and VAB measures. The educational psychologist had no previous experience of the children being tested in this study and was not informed of the child's group status (but may have been able to obtain that information by the location of the visit or through the parents).

In addition to these measures, the supervisors or teachers delivering the interventions were asked to complete a CRS-R concerning the child's behavior and a questionnaire concerning the nature of the intervention. The supervisors/teachers were asked to complete these forms at the start of the intervention (at baseline) and again after 9 months (follow up). These questionnaires were completed at the same time as the baseline and follow-up child assessment. Finally, the parents were also asked to complete a questionnaire regarding the characteristics of the program that their child had experienced.



The data analysis was divided into a number of sections for ease of understanding. Initially, the baseline scores of the children were analyzed with respect to the main outcome measures: autistic severity (GARS), intellectual functioning (PEP-R), educational functioning (BAS II), and adaptive behavioral functioning (VAB). These data were analyzed to ascertain whether the groups were well matched. The analysis was conducted by means of a multivariate analysis of covariance (MANCOVA) with the three intervention groups as independent variables, the overall baseline scores for the four scales as dependent variables, and the participants' age as a covariate. To document further the precise nature of the problems experienced by the children at intake, their scores on the four subscales of the GARS, and the four subscales of the CRS-R, at baseline, were analyzed separately using univariate analysis of covariance, with group as an independent variable and age at intake as a covariate.

The next series of analyses dealt with the changes that occurred in the overall functioning scores over the period of the intervention. First, the changes in autistic severity were analyzed to determine if any of the three interventions had an impact on the severity of the autistic characteristics (separately from changes in intellectual or educational functioning). Therefore, this analysis was conducted by analysis of covariance (ANCOVA), with the three intervention types as the independent variable, change in the overall GARS as a dependent variable, and participants' ages at intake as a covariate.

Following this analysis, the changes in the overall scores for intellectual (PEP-R), educational (BAS II), and adaptive behavioral functioning (VAB) were analyzed by MANCOVA, with the intervention group as an independent variable, the three overall outcome measures as the dependent variable, and the participants' ages and autistic severities at intake as covariates. The changes in the subscales of these measures across the three intervention groups were also analyzed separately by ANCOVA, using age and autistic severity as covariates, where the initial overall differences had proved to be statistically significant.

Finally, changes in the behavioral problems experienced by the children, as noted by the CRS-R, were analyzed by means of a MANCOVA, with the intervention group as an independent variable, the four Conner's subscales as the dependent variable, and the age and autistic severity at intake as covariates.


The baseline data from the 48 participants who completed the study are described in Tables 2, 3, and 4. Inspection of these baseline data shows that the participants in each of the three groups were well matched in terms of their baseline scores on all overall measures. In terms of the group composition: the ABA group had 12 participants (11 male and 1 female), with a mean age of 40 months (range 32-47 months) at intake; the special nursery group had 20 participants (18 male and 2 female), with a mean age of 43 months (range 41-48 months); and the portage group had 16 participants, with a mean age of 38 months (range 30-45 months). A MANCOVA was conducted on the baseline scores for the GARS, PEP-R, BAS II, and VAB as dependent measures, with intervention group as an independent variable, and age as a covariate. The Pillai's Trace criterion was adopted as the most conservative of the MANOVA test statistics (Olson, 1979). This analysis revealed no statistically significant differences between the groups at baseline, Pillai's Trace = .244, F(8,84) = 1.46, p > .10.

In addition to these overall cognitive and behavioral functioning measures, it is important to note that the children were also assessed on their specific behavioral problems through the subscales of the GARS and the CRS-R. This allows further documentation of the subject equivalence across the groups.

The subscales of the GARS deal with the degree of stereotyped behaviors, communication difficulties, social interaction, and developmental disturbances shown by the children. All of these scales offer standardized scores, with a mean of 10 and standard deviation of 3 (with high scores meaning more problems). The group means for stereotyped behaviors at baseline are shown in Table 3. Inspection of these data shows that none of the groups were discrepant from one another on any of the subscales for the GARS. Separate ANCOVAs conducted on the baseline scores across group, with age at intake as a covariate, revealed no statistically significant difference for any scale, all ps > .30.

The scores from the baseline CRS-R also offer a standardized T score (M = 50 and SD = 10, with high scores meaning more problems). The group means for the four subscales (oppositional behavior, cognitive problems, hyperactivity, and ADHD) are shown in Table 4. Inspection of these data shows that none of the groups were discrepant from one another on any of the subscales for the CRS-R. Separate ANCOVAs conducted on the baseline scores across group, with age at intake as a covariate, revealed no statistically significant difference for any scale, all ps >. 10.


The overall autism severity of the participants was assessed at baseline and follow up using the GARS. A change in a negative direction indicated an improvement (i.e., a lessening) in the severity of autistic symptoms, whereas a positive change indicated a worsening of the symptoms over time. Analysis of the difference score between the overall GARS score obtained at baseline and follow-up revealed very little change over this period for any of the intervention groups (see Table 2). These difference scores were analyzed by an ANCOVA, with intervention type (ABA, Nursery, and Portage) as a between-subject factor and age at intake as a covariate. This analysis revealed no statistically significant difference between the groups, F(2,44) = .56, p > .60.


Table 2 displays the mean change scores (follow-up standardized score minus baseline standardized score) for all three intervention types, along with the mean baseline and follow-up scores. Intellectual functioning was measured by the overall PEP-R score ([developmental age/chronological age] x 100), educational functioning was measured by the general cognitive ability score of the BAS II, and adaptive behavior change was measured by the composite VAB.

Inspection of Table 2 shows that for intellectual functioning (PEP-R), the ABA and nursery interventions produced gains of around 10 to 13 points over the 9-month period, which were somewhat higher than the gains produced by the portage intervention. Educational functioning (BAS II) showed the biggest gain in the ABA group (approximately 18 points), which exceeded the change in the other three interventions (approximately 5-8 points). For adaptive behavioral functioning, the nursery placements produced the largest gain, followed by ABA, with no gains being shown for portage.

These data were analyzed by a MANCOVA with the intellectual (PEP-R), educational (BAS II), and adaptive behavior (VAB) measures as dependent variables; intervention type (ABA, nursery, and portage) as the independent variable; and the participants' ages and their autistic severity (GARS) at the start of the intervention as covariates. This analysis revealed a statistically significant effect of intervention, Pillai's Trace = .30, F(6,84) = 2.47, p < .05, partial [eta.sup.2] = .150. Separate univariate analyses of covariance (ANCOVAs) were performed on each of the three dependent scores (PEP-R, BAS II, and VAB), with intervention as the independent variable, and age and GARS as covariates. These analyses revealed a statistically significant effect of intervention type on intellectual functioning (PEP-R), F(2,43) = 3.88, p < .05, partial [eta.sup.2] = .153; educational functioning (BAS II), F(2,43) = 4.52, p < .05, partial [eta.sup.2] = .174; and adaptive behavior (VAB), F(2,43) = 3.80, p < .05, partial [eta.sup.2] = .150. Tukey's honestly significant difference (HSD) tests were conducted, which revealed statistically significant pairwise differences between the ABA and portage for intellectual functioning (PEP-R), p < .05, between ABA and each of the other two interventions for educational functioning (BAS II), ps < .05. The Tukey's HSD tests for the adaptive behavior score (VAB) revealed a statistically significant difference between the nursery and portage interventions, p < .05. None of the other pairwise differences proved to be statistically significant.


The mean change scores (follow-up standard score minus baseline standard score) for the individual subscales of each test of functioning (i.e., PEP-R, BAS II, and VAB) can be seen in Table 3, along with the mean baseline and follow-up scores. The top panel of Table 3 shows the change in the standard scores for the subscales of the intellectual functioning (PEP-R) measure. In general, the ABA intervention showed the greatest change scores of the interventions. Table 3 also shows that there were statistically significant improvements in six of the subscales of the PEP-R for the ABA intervention (imitation, perception, gross motor skills, hand--eye coordination, cognitive performance, and cognitive verbal performance). There were statistically significant changes in three of the subscales for the nursery intervention (gross motor skills, cognitive performance, and cognitive verbal performance). However, in only the imitation score was there a statistically significant difference between the interventions, F(2,39) = 3.34, p < .05, partial [eta.sup.2] = .146. Tukey's HSD tests revealed that ABA produce a greater change than nursery on this measure, p < .05, but none of the other pairwise comparisons proved statistically significant.

For the subscales of the educational functioning (BAS II) measure, the ABA intervention produced statistically significant gains in all the subscales. The nursery intervention produced statistically significant gains in all the subscales, except verbal comprehension. Across the interventions, ABA produced statistically significantly greater improvements than the other interventions, these being statistically significant for both verbal comprehension, F(2,39) = 5.00, p < .01, partial [eta.sup.2] = .204 and naming vocabulary, F(2,39) = 10.08, p < .001, partial [eta.sup.2] = .341. For both measures, follow-up Tukey's HSD tests showed that the ABA intervention outperformed both the nursery and portage interventions. For the subscales of the VAB scale there was a mixed pattern of results. The nursery intervention produced statistically significant improvements for the communication and socialization scales, and the nursery intervention tended to outperform the other two interventions marginally on all the scales, except communication. However, only this latter scale showed a statistically significant difference between the interventions, F(2,39) = 4.95, p < .01, partial [eta.sup.2] = .191, with the ABA intervention outperforming the portage intervention, p < .05.


In addition to the overall intellectual and cognitive changes, the changes in behavioral problems as indexed by the CRS-R are shown in Table 4, along with the group means at baseline and follow up. In terms of oppositional behavior, nursery and portage programs showed more improvement than the ABA group; whereas for cognitive problems, ABA and nursery showed more change than portage. All groups showed similar improvements for hyperactivity and ADHD problems.

These data were analyzed by a MANCOVA with intervention group as a dependent variable, and the age and autistic severity of the participants as covariates. This analysis showed a statistically significant difference between the groups, Pillai's Trace = .521, F(8.82) = 3.61, p < .001, partial [eta.sup.2] = .260. Separate ANCOVAs conducted on each subscale revealed significant differences on the oppositional problems scale, F(2,43) = 5.22, p < .01, partial [eta.sup.2] = .195. Here both the nursery and portage groups showed significantly greater improvements than the ABA group (Tukey's HSD tests, p < .05). There were also statistically significant differences between the groups on the cognitive problem scale, F(2,42) = 3.82, p < .05, partial [eta.sup.2] = .151. Tukey's HSD tests showed that the nursery group differed statistically significantly from the portage group, p < .05. There were no group differences on either of the hyperactivity and ADHD scales, both ps > .20.


The current study examined the effectiveness of three early-teaching interventions for children with ASD. The primary aims of this study were to examine the impact of these interventions as they occur in the community rather than the clinic, and to examine their impact on a set of participants not heavily selected on criteria designed to maximize possible outcome efficacy. In this way, it was hoped to produce a study with a high external validity. In addition, it was hoped to show some of the factors that contribute to the success of these interventions. In summary, the results shed new light on the real-world effectiveness of several of the educational interventions. Particularly, they partially replicated the effectiveness of previous studies of clinic-based ABA programs in a community-based sample. They also showed strong gains over a relatively short period of time, and demonstrated the effectiveness of nursery placements in promoting aspects of intellectual and adaptive behavioral functioning.

The participants in this study tended to have moderate autistic symptoms, being in the mid range of the standard scores generated by the GARS. The participants' intellectual functioning, educational functioning, and adaptive behavioral functioning, measured at the start of the intervention, all showed quite severe impairments; the mean scores were around 50 on each standardized instrument, indicating that these participants were at the severe end of those taken on previous studies of ABA (e.g., Lovaas, 1987, excluded children with IQs lower than 50). To this extent, the current participants were more representative of children with ASD who present for intervention at local authorities (see Conner, 1998).

The present results indicated that, on the basis of the GARS rating of autism symptom severity, none of the interventions produced recovery, as noted in some previous studies (e.g., Lovaas, 1987). This finding corroborates a number of other studies of the effectiveness of ABA approaches and adds to the current weight of evidence that this approach will not cure autism. However, it should be noted that there are a number of possible reasons for this discrepancy between this study and that of Lovaas (1987); the length of time of the intervention (9 months vs. 36 months), the place of intervention (community vs. clinic), and the severity of the participants symptoms (severe vs. milder). Whatever the reason for this difference, it appears that in a typical community-based intervention, over 9 months, there is no sign of recovery from the symptoms of autism.

Irrespective of the impact on the autistic severity of the participants' ASD, the ABA program produced significant gains on both intellectual and educational functioning, and the nursery intervention produced a strong impact on all three measures of functioning, with the improvements resulting from portage being limited to only a few measures. The comparison between the ABA intervention and the portage intervention in the current study offers a chance to replicate the original comparison in the Lovaas (1987) study between a high intensity and low intensity treatment program, with the same results; the high-intensity program produced stronger improvements (see also Smith, Eikeseth, Klevstrand, & Lovaas, 1997).

With respect to the ABA intervention, these data show that ABA has a strong impact on intellectual and educational functioning. These gains in functioning replicate, in part, the effects of previous ABA studies, which were conducted in a clinic, on a home-based sample (Eikeseth et al., 2002). The gains are also of a similar size to previous longer period studies of ABA (Lovaas, 1987), but were achieved over a somewhat brief period (9 months). In addition, they show that ABA can be effective for children with relatively severe difficulties (see also Smith et al., 2000).

Although ABA also showed some impact on adaptive behavioral functioning, it was the special nursery program that had the greatest impact on this aspect of functioning (with an impact also on intellectual and educational functioning). This aspect of the data was borne out by the teachers' ratings of behavioral problems, which showed that nursery placement affected these problems to the greatest extent. Thus, these data show, for the first time, the impact of special nursery placement on these measures, and offer support for other studies that have shown some improvement that is due to special nursery placement (Gabriels et al., 2001). They also stand in contrast to the often reported failures to produce gains in the nursery control groups of ABA intervention studies (see Howard et al., 2005; Lovaas, 1987). It is difficult to be precise about the reasons why the current study produced gains whereas other studies did not, although there are some differences that might bear subsequent investigation. In the current study, the special nursery placement was heavily reliant on group-teaching approaches, whereas in previous studies (e.g., Howard et al.) this has not always been the case. There also remains the possibility that there are individual differences between the effectiveness of special nursery placements and the receptiveness of the children to the approach. The precise nature of the special nursery placements is, of course, difficult to determine on a day-to-day basis, and they did contain elements of ABA and TEACCH approaches. A precise description of special nursery placement, often characterized as an eclectic approach would be a useful step in this field, as it remains a very widely used early intervention for children with ASD. Thus, disentangling these influences remains a key objective, but it is an objective that may be very difficult to implement in a natural setting.

That ABA has a strong impact on intellectual/educational gains, and special nursery on adaptive behaviors, may reflect the difference in teaching approaches between the two interventions (one-to-one vs. group). This pattern of data might lead to the suggestion that any program should include both relatively intense one-to-one work, followed by group work such as occurs in a special nursery intervention.

The current study was primarily concerned with determining whether a number of interventions would produce gains in children with ASD when they were conducted in the community. It was not a study designed to answer the question of why those interventions worked; given the state of knowledge of outcome effectiveness in the community this approach seems a reasonable first step.

There are a number of caveats to be made when interpreting these data. The main caveat concerns the fact that the groups were not randomly selected for each of the interventions. The participants received whatever intervention was being provided in their locality. This aspect of the design was a deliberate part of the study, adopted to ensure a high external validity. Low external validity has been a criticism of many previous studies (see Conner, 1998); children with ASD are not randomly allocated to treatment in the real world. However, despite this concern, inspection of the groups of participants reveals that they were very evenly matched, suggesting that there were no preexisting systematic biases in the participants across the groups. This should be the case, as the geographical location of the participant, rather than any characteristic, determined their allocation to intervention. Additionally, this study was not designed to address exactly what aspect of the intervention was responsible for any observed effectiveness. The nature of the interventions was characterized by the tutors and parents, and some variables, such as time input, were identified as important. However, further work will be needed to establish which aspects of these complex interventions are critical in generating intellectual and behavioral change in the participants. Additionally, further measures perhaps tailored to particular curriculums would be helpful for assessing the impact of specific early teaching interventions, although these would limit the generality of findings to particular locations. (For example, curriculum key stages may well differ between the United States and the United Kingdom, and certainly with other regions.)

One issue that should be commented on in this study concerns the fidelity of the treatment regimes in the current study. This study was an investigation into how the programs sometimes studied in the clinic fared when placed in the community. This is a key issue about the effectiveness of such programs, especially ABA (see Mudford et al., 2001). As the programs were conducted in the home, their day-to-day fidelity was not directly measured. It should be noted that all programs were conducted by trained supervisors in accordance with the program manuals, which should ensure some treatment fidelity, as in other studies. However, the lack of direct observation may present a limitation on the current work. It should be noted that it is not clear the degree to which any program has fidelity to the manual in the face of specific demands of individual children, and that many programs are individually tailored to the child, and therefore vary from individual-to-individual, and from day-to-day within individuals. Of course, stronger measures of the treatment fidelity would improve the internal validity of the study and would be welcomed in future work, but if the characteristics of the program conducted in a study with high internal validity do not correspond to the way such a program would be conducted in the majority of cases, the findings would be of limited external value.

In summary, the current report failed to note any evidence of recovery from autism produced by any early intervention. In terms of intellectual functioning, applied behavior analysis and special nursery interventions produced gains (of the same magnitude as many gains produced by previous longer term clinic-based ABA programs). The results from clinic-based ABA trials were partially replicated on a community-based sample; specifically with respect to intellectual and educational skills. Special nursery placement was also found to be effective for improving adaptive behavior and educational skills. Of course, the relative importance of educational skills versus adaptive behavioral skills at preschool for subsequent school functioning and school inclusion needs to be investigated.

Manuscript received June 2006; accepted October 2006.


Cameron, R. J. (1997). Early intervention for young children with developmental delay: The portage approach. Child: Care, Health and Development, 23, 11-27.

Charman, T., Howlin, P., Berry, B., & Prince, E. (2004). Measuring developmental progress of children with autism spectrum disorder on school entry using parent report. Autism, 8, 89-100.

Conners, C. K. (1997). Conners' Rating Scales. Eagan, MN: Pearson Assessment.

Connor, M. (1998). A review of behavioural early intervention programmes for children with autism. Educational Psychology in Practice, 14, 109-117.

Eikeseth, S., Smith, T., Jahr, E., & Eldevik, E. (2002). Intensive behavioral treatment at school for 4- to 7-year old children with autism. Behavior Modification, 26, 49-68.

Elliott, C. D., Smith, P., & McCulloch, K. (1996). British Ability Scales: Second Edition. London: NFER-Nelson.

Gabriels, R. L., Hill, D. E., Pierce, R. A., Rogers, S. J., & Wehner, B. (2001). Predictors of treatment outcome in young children with autism. Autism, 5, 407-429.

Gilliam, J. E. (1995). Gilliam Autism Rating Scale. Austin, TX: Pro-Ed.

Greenspan, S. I., & Wieder, S. (1997). Developmental patterns and outcomes in infants and children with disorders in relating and communicating: A chart review of 200 cases of children with autistic spectrum diagnoses. Journal of Developmental and Learning Disorders, 1, 87-141.

Greer, R. D. (1997). The Comprehensive Application of Behavior Analysis to Schooling (CABAS[R]). Behavior and Social Issues, 7, 59-63.

Gresham, F. M., & MacMillan, D. L. (1997). Autistic recovery? An analysis and critique of the empirical evidence on the Early Intervention Project. Behavioral Disorders, 22, 185-201.

Handen, B. L., Johnson, C. R., & Lubetsky, M. (2000). Efficacy of methylphenidate among children with autism and symptoms of attention-deficit hyperactivity disorder. Journal of Autism and Developmental Disorders, 30, 245-255.

Hohmann, A. A., & Shear, M. K. (2002). Community-based intervention research: Coping with the "noise" of real life in study design. American Journal of Psychiatry, 159, 201-207.

Howard, J. S., Sparkman, C. R., Cohen, H. G., Green, G., & Stanislaw, H. (2005). A comparison of intensive behavior analytic and eclectic treatments for young children with autism. Research in Developmental Disabilities, 26, 359-383.

Howlin, P., & Moore, A. (1997). Diagnosis in autism: A survey of over 1200 parents in the U.K. Autism, 1, 135-162.

Koegel, R. L., & Koegel, L. K. (2006). Pivotal response treatments for autism. Baltimore, MD: Paul H. Brookes.

Lovaas, O. I. (1981). The ME book: Teaching developmentally disabled children. Austin, TX: Pro-Ed.

Lovaas, O. I. (1987). Behavioral treatment and normal educational and intellectual functioning in young autistic children. Journal of Consulting and Clinical Psychology, 55, 3-9.

Lovaas, O. I., & Smith, T. (1989). A comprehensive behavioral theory of autistic children: Paradigm for research and treatment. Journal of Behavior Therapy and Experimental Psychiatry, 20, 17-29.

Magiati, I., & Howlin, P. A. (2001). Monitoring the progress of preschool children with autism enrolled in early intervention programmes: Problems in cognitive assessment. Autism, 5, 399-406.

Mesibov, G. B., Schopler, E., & Hearsey, K. A. (1994). Structured teaching. In E. Schopler & G. B. Mesibov (Eds.), Behavioral issues in autism (pp. 195-207). New York: Plenum Press.

Mudford, O. C., Martin, N. T., Eikeseth, S., & Bibby, P. (2001). Parent-managed behavioral treatment for preschool children with autism: Some characteristics of U.K. programs. Research in Developmental Disabilities, 22, 173-182.

Olson, C. L. (1979). Practical considerations in choosing a MANOVA test statistic: A rejoinder to Stevens. Psychological Bulletin, 86, 1350-1352.

Osborne, L. A., & Reed, P. (in press). Caregivers' perceptions of communication with professionals during the diagnosis of autism. Autism.

Reed, P., Osborne, L. A., & Gibson, E. (2000). Portage as an intervention for autism. International Portage Newsletter, 13, 1-3.

Schopler, E., Reichler, R., Bashford, A., Lansing, M., & Marcus, L. (1990). Psychoeducational profile (Rev. ed.). Austin, TX: Pro-Ed.

Shallows, G. O., & Graupner, T. D. (1999, May). Replicating Lovaas' treatment and findings: Preliminary results. Paper presented at PEACH: Putting Research Into Practice Conference, London.

Sheinkopf, S. J., & Siegel, B. (1998). Home-based behavioral treatment of young children with autism. Journal of Autism and Developmental Disorders, 28, 15-23.

Smith, C. (2000). The development and evaluation of additional training for portage workers in addressing the needs of children with complex social and communication difficulties/autistic spectrum disorders. Unpublished doctoral dissertation, University of Southampton, U.K.

Smith, T., Eikeseth, S., Klevstrand, M., & Lovaas, O. I. (1997). Intensive behavioral treatment for preschoolers with severe mental retardations and pervasive developmental disorder. American Journal on Mental Retardation, 102, 238-249.

Smith, T. G., Annette, D., & Wynn, J. W. (2000). Randomized trial of intensive early intervention for children with pervasive developmental disorder. American Journal on Mental Retardation, 105, 269-285.

Sparrow, S., Balla, D., & Cicchetti, D. (1984). Vineland Adaptive Behavior Scales. Circle Pines, MN: American Guidance Service.

Sundberg, M. L., & Michael, J. (2001). The benefits of Skinner's analysis of verbal behavior for children with autism. Behavior Modification, 25, 698-724.

Twyman, J. S. (1998). The Fred S. Keller School. Journal of Applied Behavior Analysis, 31, 695-701.

Address correspondence to Phil Reed, Department of Psychology, Swansea University, Singleton Park, Swansea, SA2 8PP, U.K. (e-mail:

This research was funded by a grant from the South East Regional Special Educational Needs Partnership (SERSEN, U.K.), and we thank them for their support during the course of this research. Also, we are very grateful to the children, parents, teachers, and schools who kindly participated in this study. Some of these data were initially discussed at the European Association for Behaviour Analysis Conference held in Parma, Italy, 2003, and at the Behavioural Association of Ireland Conference held in Galway, Ireland, 2004.



Swansea University


Children, Young People, and Families Directorate, Oxford


PHIL REED Professor; and LISA A. OSBORNE, Senior Research Assistant, Department of Psychology, Swansea University, Swansea, U.K. MARK CORNESS, Principal Educational Psychologist, Children, Young People, and Families Directorate, Macclesfield House, Oxford, U.K.
Characteristics of the Three Interventions Studied

Variable ABA Nursery Portage

Mean intervention (hr) 30.4 12.7 8.5
Range (hr) 20-40 3-23 2-15
Interquartile range (hr) 28-34 12-15 3-9
One-to-one teaching (hr) 28.3 3.1 6.5
Group teaching (hr) 2.1 9.6 2.0
Tutors (family tutors) 4.4 (1.0) 4.0 (1.0) 4.0 (2.0)
Service providers 5 7 7
Treatment characteristics
 Based Home School Home
 Teaching One-to-one Group One-to-one
 Led by Tutor Tutor Parent
 Methods Reward Various Reward
 Intensity High Moderate Low

Mean (Standard Deviation) Baseline, Follow Up, and Change Scores for
the Three Intervention Groups

Group Scale Baseline Follow Up Change

ABA GARS 90.5 (13.9) 88.5 (15.4) -2.0 (8.4)
 PEP-R 55.6 (13.8) 69.2 (19.8) 13.6 (12.7)
 BAS II 56.8 (16.6) 74.5 (22.4) 17.8 (15.0)
 VAB 58.2 (8.0) 61.1 (10.6) 2.9 (6.5)

Nursery GARS 96.8 (10.4) 97.2 (11.0) 0.5 (9.3)
 PEP-R 51.9 (20.1) 62.0 (21.9) 10.2 (13.6)
 BAS II 57.8 (12.8) 65.7 (15.5) 7.9 (8.6)
 VAB 53.0 (4.6) 56.2 (8.1) 3.3 (6.3)

Portage GARS 91.3 (14.3) 92.8 (14.9) 1.5 (5.9)
 PEP-R 53.3 (16.1) 54.8 (19.2) 1.6 (11.2)
 BAS II 53.4 (10.9) 60.0 (13.9) 6.6 (9.1)
 VAB 58.6 (6.0) 57.2 (7.7) -1.4 (4.0)

Note. Gilliam Autism Rating Scale = GARS; Psychoeducational Profile
Revised = PEP-R; British Ability Scales: Second Edition = BAS II;
Vineland Adaptive Behavior = VAB. All standard scores (mean = 100,
standard deviation = 15).

Mean (Standard Deviation) Baseline, Follow Up, and Change Scores for
Each Subscale for the Three Intervention Groups

Group and Scale Baseline Follow Up

 GARS (M = 10, SD = 3)
 Stereotyped behaviors 8.1 (2.8) 7.1 (3.0)
 Communication 1.6 (3.3) 5.4 (4.8)
 Social interaction 7.4 (2.7) 7.0 (2.9)
 Developmental disturbance 9.2 (2.3) 9.2 (2.4)
 PEP-R (M = 100, SD = 15)
 Imitation 45.5 (30.9) 76.5 (24.9)
 Perception 78.2 (24.6) 99.8 (35.6)
 Fine motor 71.5 (18.7) 75.3 (19.2)
 Gross motor 60.3 (17.5) 91.1 (31.7)
 Hand-eye 64.8 (21.7) 77.3 (22.5)
 Cognitive 40.7 (23.2) 59.3 (22.2)
 Verbal 41.9 (22.6) 61.7 (24.9)
 BAS II (M = 50, SD = 10)
 Verbal comprehension. 23.3 (6.8) 32.8 (17.6)
 Picture matching 33.8 (13.3) 42.3 (15.3)
 Naming 22.4 (8.3) 35.6 (13.5)
 Early number skills 26.3 (5.3) 34.8 (10.9)
 VAB (M = 100, SD = 10)
 Communication 57.1 (7.6) 64.1 (14.9)
 Daily living skills 61.6 (6.7) 64.3 (10.8)
 Socialization 62.1 (9.3) 66.1 (9.2)
 Motor skills 73.8 (16.6) 71.8 (15.7)

Special Nursery
 GARS (M = 10, SD = 3)
 Stereotyped behaviors 9.3 (2.3) 9.6 (3.2)
 Communication 5.9 (5.6) 4.6 (5.3)
 Social interaction 9.1 (2.8) 8.6 (2.9)
 Developmental disturbance 9.6 (1.9) 9.7 (1.9)
 PEP-R (M = 100, SD = 15)
 Imitation 52.9 (30.3) 61.0 (30.5)
 Perception 80.4 (45.6) 81.2 (30.0)
 Fine motor 59.9 (21.0) 63.6 (18.2)
 Gross motor 67.2 (28.2) 80.5 (28.9)
 Hand-eye 66.0 (31.3) 71.0 (25.4)
 Cognitive 42.5 (20.2) 56.3 (24.3)
 Verbal 43.8 (23.8) 55.3 (26.8)
 BAS II (M = 50, SD = 10)
 Verbal comprehension 23.3 (6.8) 26.0 (7.0)
 Picture matching 33.8 (13.3) 37.1 (11.9)
 Naming 22.4 (8.4) 29.7 (10.1)
 Early number skills 26.3 (5.3) 32.1 (9.9)

 VAB (M = 100, SD = 10)
 Communication 57.1 (7.6) 58.3 (10.1)
 Daily living skills 61.6 (6.9) 60.9 (11.0)
 Socialization 62.1 (9.3) 61.6 (9.0)
 Motor skills 73.8 (16.6) 64.5 (14.0)

 GARS (M = 10, SD = 3)
 Stereotyped behaviors 8.9 (3.1) 8.8 (3.7)
 Communication 0.4 (1.8) 1.6 (3.5)
 Social interaction 8.3 (2.9) 8.1 (2.6)
 Developmental disturbance 9.2 (1.8) 9.6 (2.0)
 PEP-R (M = 100, SD = 15)
 Imitation 39.4 (22.1) 49.4 (26.8)
 Perception 74.8 (31.3) 76.8 (26.7)
 Fine motor 65.0 (19.9) 63.9 (21.6)
 Gross motor 68.3 (20.2) 79.3 (36.4)
 Hand-eye 64.7 (28.0) 63.8 (31.1)
 Cognitive 38.3 (17.6) 44.1 (23.4)
 Verbal 42.6 (15.9) 45.6 (18.7)
 BAS II (M = 50, SD = 10)
 Verbal comprehension 21.7 (4.3) 22.3 (3.9)
 Picture matching 28.8 (8.0) 35.8 (11.1)
 Naming 22.3 (6.3) 24.0 (8.5)
 Early number skills 26.1 (7.8) 27.8 (8.8)
 VAB (M = 100, SD = 10)
 Communication 57.1 (4.9) 54.8 (4.9)
 Daily living skills 64.6 (6.2) 63.1 (10.4)
 Socialization 61.0 (5.8) 60.7 (7.3)
 Motor skills 70.9 (11.1) 70.0 (13.7)

Group and Scale Change

 GARS (M = 10, SD = 3)
 Stereotyped behaviors -1.0 (1.5)
 Communication 3.8 (5.5)
 Social interaction -0.4 (1.6)
 Developmental disturbance 0 (1.1)
 PEP-R (M = 100, SD = 15)
 Imitation 31.0 (22.8) **
 Perception 21.6 (20.1) **
 Fine motor 3.8 (19.5)
 Gross motor 30.8 (25.0) **
 Hand-eye 12.4 (16.4) *
 Cognitive 18.6 (17.4) **
 Verbal 19.8 (24.5) *
 BAS II (M = 50, SD = 10)
 Verbal comprehension. 9.5 (14.3) *
 Picture matching 8.6 (8.8) **
 Naming 13.2 (11.9) **
 Early number skills 8.6 (9.6) **
 VAB (M = 100, SD = 10)
 Communication 7.0 (11.6)
 Daily living skills 2.7 (6.2)
 Socialization 4.0 (7.3)
 Motor skills -2.0 (11.6)

Special Nursery
 GARS (M = 10, SD = 3)
 Stereotyped behaviors 0.3 (2.1)
 Communication -1.3 (5.8)
 Social interaction -0.6 (2.5)
 Developmental disturbance -0.2 (1.2)
 PEP-R (M = 100, SD = 15)
 Imitation 8.2 (21.2)
 Perception 0.8 (34.6)
 Fine motor 3.7 (13.1)
 Gross motor 13.3 (19.7) **
 Hand-eye 5.0 (21.9)
 Cognitive 13.9 (16.0) **
 Verbal 11.5 (15.3) **
 BAS II (M = 50, SD = 10)
 Verbal comprehension 1.8 (4.1)
 Picture matching 7.9 (10.9) **
 Naming 2.7 (4.0) **
 Early number skills 4.3 (7.6) *

 VAB (M = 100, SD = 10)
 Communication 3.1 (6.4) *
 Daily living skills 3.1 (8.6)
 Socialization 5.0 (8.8) *
 Motor skills 3.2 (10.7)

 GARS (M = 10, SD = 3)
 Stereotyped behaviors 0.1 (2.5)
 Communication 1.2 (3.1)
 Social interaction 0.1 (1.1)
 Developmental disturbance 0.5 (1.5)
 PEP-R (M = 100, SD = 15)
 Imitation 10.1 (23.6)
 Perception 1.9 (20.8)
 Fine motor -1.1 (15.1)
 Gross motor 11.0 (26.0)
 Hand-eye -0.9 (18.9)
 Cognitive 5.9 (14.3)
 Verbal 3.1 (17.7)
 BAS II (M = 50, SD = 10)
 Verbal comprehension 0.7 (3.4)
 Picture matching 6.9 (7.6) **
 Naming 1.7 (4.6)
 Early number skills 1.8 (8.6)
 VAB (M = 100, SD = 10)
 Communication -2.3 (4.1) *
 Daily living skills -1.5 (6.6)
 Socialization -0.3 (3.9)
 Motor skills -0.9 (10.3)

Note. Gilliam Autism Rating Scale = GARS; Psychoeducational Profile
Revised = PEP-R; British Ability Scales: Second Edition = BAS II;
Vineland Adaptive Behavior = VAB.

* p < 0.05. ** p < 0.01.

Mean (Standard Deviation) Baseline, Follow Up, and Change Scores
for the Three Intervention Groups for Connors' Scales

Group Scale Baseline Follow Up Change

ABA Oppositional 62.1 (9.9) 61.3 (12.6) -0.8 (10.1)
 Cognitive 72.4 (19.4) 61.6 (16.1) -10.4 (26.6)
 Hyperactivity 63.2 (8.2) 55.8 (8.8) -7.0 (7.1)
 ADHD 64.6 (7.2) 57.8 (10.2) -6.9 (8.0)

Nursery Oppositional 70.4 (13.0) 59.8 (14.1) -10.7 (11.4)
 Cognitive 78.9 (10.9) 62.7 (12.7) -16.1 (14.1)
 Hyperactivity 68.6 (9.9) 62.0 (13.6) -6.5 (8.5)
 ADHD 69.0 (9.3) 62.3 (11.7) -6.7 (6.2)

Portage Oppositional 68.3 (6.1) 58.4 (1.8) -9.9 (6.3)
 Cognitive 74.9 (13.6) 74.8 (2.1) 0.4 (13.2)
 Hyperactivity 68.4 (6.3) 63.1 (0.6) -4.8 (6.4)
 ADHD 66.9 (7.1) 63.3 (0.9) -3.8 (7.1)
COPYRIGHT 2007 Council for Exceptional Children
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Reed, Phil; Osborne, Lisa A.; Corness, Mark
Publication:Exceptional Children
Geographic Code:1USA
Date:Jun 22, 2007
Previous Article:Co-teaching in inclusive classrooms: a metasynthesis of qualitative research.
Next Article:A national study of youth attitudes toward the inclusion of students with intellectual disabilities.

Related Articles
Autism rising: delivering services without draining parents and school systems. (research reflections).
One-on-one Tx for autism trumps mainstreaming.
Autism-specific screen outdoes general tool.
Optimum treatment of problem behaviors in children with autism spectrum disorders.
Video modeling for individuals with autism: a review of model types and effects.
Choosing effective treatments for children with autism.
Promoting the social and cognitive competence of children with autism: interventions at school.
Lawmakers address autism.
Conundrum of autism: a review of its causes and significant impact on the education of a school age child.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters