The Effects of Community Preschool Characteristics on Developmental Outcomes for Students With Autism Spectrum Disorder

Abstract

This study examined factors associated with 9-month outcomes of 78 preschoolers with autism spectrum disorder ([ASD]; 66 males, mean age = 44.8 [7.0] months) receiving preschool special education services. Use of recommended intervention practices was observed in each student’s primary educational setting: autism-only (n = 28), mixed-disability (n = 25), and general early education (n = 25). Recommended practice use across setting types and developmental changes were compared using generalized estimating equations. Autism-only settings had the highest implementation of classroom structure, classroom environment, and curriculum and instruction. General early education had greater implementation of practices to support social and peer relationships. Supporting social and peer relationships was the largest predictor of gains, controlling for setting type; the association was greater for children with lower initial receptive language or developmental skills. These findings highlight the importance of supporting and promoting peer engagement in students with ASD across preschool placement types.

Keywords

autism spectrum disorders preschool early intervention inclusion

This study examined the associations between characteristics of community-based preschool early intervention services and outcomes for children with autism spectrum disorder (ASD). Children with ASD who receive high-quality early intervention (EI) can make large gains in cognitive, communication, and adaptive behaviors skills, with positive long-term effects (Reichow et al., 2012; Weitlauf et al., 2014). Most studies demonstrating these gains were conducted in university-based research settings using expert clinicians and highly manualized interventions.

As Kasari and Smith (2013) note, few children served in public schools are represented in studies examining the “evidence base” because most studies are of small size and include homogeneous samples. In addition, these studies often exclude children who are more impaired, minimally verbal, non-English speaking, or who have multiple disabilities. While emerging research demonstrates that if expert support is offered, evidence-based interventions (EBIs) can be implemented in community settings (Shire & Kasari, 2014; Suhrheinrich et al., 2020; Vivanti et al., 2014), these studies provide little insight into outcomes associated with current community practices and usual care in authentic school settings.

Studying intervention as it is delivered in community settings can provide important insights into which practices may be most effective, given the resources available in these settings (Stahmer & Aarons, 2009). To date, few studies have measured both the type and the quality of intervention in community-based school settings and the associated outcomes. In addition, few studies have included diverse samples that represent the full range of backgrounds and functioning of children with ASD. The little research in this area suggests that EBIs for ASD rarely are found delivered to fidelity in the community (Stahmer et al., 2005; Wood et al., 2015) and that usual care is not associated with as large of gains as in expert-implemented interventions (Nahmias et al., 2019). Existing studies of outcomes in community settings are limited in that many were retrospective (Flanagan et al., 2012), lacked a comparison group, or relied on outcomes only from one program, type of intervention, or intervention setting (e.g., Eapen et al., 2013; Fernell et al., 2011; Magiati et al., 2007).

In addition to varying in the extent to which they use evidence-based practices, preschool EI settings vary considerably in the extent to which children with ASD interact with typically developing children. In the United States, the Individuals with Disabilities Education Improvement Act (2004) requires that students with disabilities, including autism, be educated in the least restrictive environment, stating that “to the maximum extent appropriate, children with disabilities . . . are educated with children who are nondisabled.” Most research to date has evaluated interventions implemented outside regular education classrooms that do not routinely offer such opportunities (National Research Council, 2001; Odom et al., 2010), and it has not compared inclusive with noninclusive settings. The few comparative studies are inconclusive (Boyd et al., 2014; Harris et al., 1990), although recent research is promising that EBIs (specifically Group-Early Start Denver Model) can be implemented in settings with peers without disabilities with similar outcomes to autism-specific settings (Vivanti et al., 2019).

To our knowledge, only one study has examined how community preschool setting characteristics are associated with child outcomes. Nahmias et al. (2014) found that children with ASD who received preschool special education services in inclusive placements had higher cognitive scores when they finished preschool than children who received preschool early intervention in more restrictive placements. This was particularly the case for children with baseline higher communication skills, lower social-emotional skills, and poorer adaptive behavior. The authors hypothesized that inclusive placements provide more opportunities to interact with typically developing peers, which, in turn, may be associated with better outcomes, but they lacked the data to test this hypothesis. This study also was hampered by its retrospective design, different measurement tools at baseline and follow-up, and limited characterization of both children and intervention.

Across intervention programs, children with ASD vary in their response to treatment, and predicting which children will respond to which intervention remains a challenge. In addition to the moderators of outcome reported by Nahmias and colleagues (2014), language abilities (Kasari et al., 2008), social communication skills (Kasari et al., 2008; Yoder & Stone, 2006a, 2006b), adaptive behavior (Eldevik et al., 2010; Flanagan et al., 2012; Remington et al., 2007), IQ (Eldevik et al., 2010; Magiati et al., 2007; Perry et al., 2011; Remington et al., 2007), object exploration (Carter et al., 2011; Yoder & Stone, 2006a, 2006b), age (Flanagan et al., 2012; Perry et al., 2011), and autism severity (Remington et al., 2007) have emerged as potential moderators of outcomes for young children with ASD. Their interaction with characteristics of preschool settings in predicting outcomes has not been examined prospectively.

We built on the study by Nahmias et al. (2014) and other research by following preschoolers with ASD as they received publicly funded preschool special education services. We carefully characterized children at baseline, measured characteristics of the intervention they received, and assessed outcomes at 9 months. This study takes advantage of the variation in authentic school-based intervention practices and provides insight into interventions characteristics most associated with positive outcomes and whether those outcomes differ by child characteristics. We attempted to replicate the findings from the Nahmias et al. (2014) paper and, given the more careful measurement, also explored other moderators, taking an agnostic approach. Findings from this study can provide a benchmark against which to measure future progress in community-based preschool interventions. If certain intervention characteristics are associated with better outcomes, it can lead to experimental studies of which strategies work best in authentic preschool settings.

Method

Participants

The sample consisted of children with ASD receiving preschool early intervention in a U.S. urban school district. Children were eligible for this study if they (a) were between 36 and 59 months of age, (b) had a documented diagnosis of ASD, and (c) received services through the county’s public preschool special education or behavioral health system. Exclusion criteria were (a) caregivers did not speak English or (b) either the caregiver or classroom staff for a given child did not provide informed consent to participate. The sample was recruited on a rolling basis through the preschool special education system from July 2014 to August 2016.

Procedure

The study was approved by the University of Pennsylvania Institutional Review Board. Students were assessed at two time points: “T1” at entry to the study and “T2” after 9 months of publicly funded preschool EI services. This time frame of 9 months was selected because it is the standard length of the academic year. Students were assigned to EI placements based on usual practices by service coordination teams, based on parent preference and provider availability, in accordance with Part B of Individuals with Disabilities Education Act (2004) and the state educational code (which both require that students with disabilities are educated in the least restrictive environment).

Measures

Mullen Scales of Early Learning (MSEL)

Our primary outcome measure was the MSEL (Mullen, 1995), a standardized, reliable, and valid measure of early cognitive development for children from birth to 68 months old. The Early Learning Composite (ELC) is based on four MSEL scales (visual reception [VR], fine motor [FM], expressive language [EL], and receptive language [RL]). The MSEL covers the full age range of our sample. Because more than half of the sample had T-scores below 20 on the RL and EL scales at baseline, developmental quotients (DQ) were calculated by dividing the age equivalent by the child’s chronological age in months and multiplying by 100 as done in similar situations (see Eapen et al., 2013; Kaale et al., 2012). For the ELC, the DQ was calculated by first averaging the age equivalences of the VR, FM, EL, and RL scales. The change on the MSEL ELC DQ between baseline and follow-up was used as the primary dependent variable. MSEL EL and RL scores were explored as predictors of change.

Educational Program Review

Our primary independent variables were derived from the Educational Program Review (EPR), a measure of classroom characteristics that was developed to characterize the quality of “business as usual” preschool settings in which children with autism receive early intervention services. It was adapted from the Professional Development in Autism Program Assessment (PDA, 2008) that was developed as part of a U.S. Department of Education Office of Special Education and Rehabilitative Services–funded center at the University of Washington for training personnel to provide evidence-based educational services to students. Adaptations of the PDA have been validated (Hume et al., 2011) and used in other studies of preschool intervention (Boyd et al., 2014; Suhrheinrich et al., 2020). The EPR consists of 50 items on seven subscales: Teaming, Classroom Structure, Classroom Environment, Curriculum and Instruction, Social/Peer Relationships, Management of Challenging Behaviors, and Building a Positive Instructional Climate. Items are rated on a 1 to 5 scale (1 = Minimal/no implementation, 3 = Partial Implementation, to 5= Full implementation) by a rater based on a 60-min direct observation, classroom staff interview, and review of the child’s Individualized Education Program. Sample items include “a daily schedule of activities is used by students and staff, individual schedules are utilized if necessary” (Classroom Structure); “a low student-teacher ratio is maintained to insure instructional fidelity and appropriate intensity of intervention” (Classroom Environment); and “staff plan daily opportunities for children with ASD to interact successfully with typically developing peers and use a range of evidence-based instructional strategies during these interactions” (Social/Peer Relationships). Postdoctoral fellows, graduate students, and bachelor’s level research assistants with experience in ASD interventions were trained to reliability by an expert coder. A subset EPRs was independently rated to assess ongoing interrater reliability. The Management of Challenging Behaviors scale was not included in analyses due to insufficient interrater reliability, and the Teaming scale was not included due to missing teacher interviews from four providers that impacted the ability for raters to score this scale. The remaining five scales demonstrated excellent interrater reliability, all above .75 (Fleiss, 1986). The intraclass correlations (ICC) for each scale were as follows: Classroom Structure = .81, Classroom Environment = .91, Curriculum and Instruction = .88, Social/Peer Relationships = .94, and Building a Positive Instructional Climate = .78. The EPR was conducted in the early education setting in which the student had spent the most time at approximately 4 and half months after study enrollment. Observations and provider interviews were scheduled based on the preferences and availability of providers and families.

Early education setting type

This also was used as an independent variable and was coded based on teacher report during the EPR as autism-only (i.e., class primarily composed of students with an educational classification of autism), mixed-disability (i.e., class composed of students with a variety of educational classifications), or regular early education environment (e.g., typical preschool, daycare, reverse mainstream classroom, inclusion classroom).

Community autism diagnosis

An autism diagnosis was confirmed using the Autism Diagnostic Observation Schedule–Second Edition (ADOS; Lord et al., 2012) or the Social Communication Questionnaire (SCQ; Rutter et al., 2003). The ADOS is a semi-structured play-based assessment considered to be the gold-standard observational measure for diagnosing ASD. It was administered by a graduate student or postdoctoral fellow in psychology trained to research reliability and supervised by a licensed clinical psychologist. Calibrated Severity Scores were used in analyses. The SCQ is a brief parent questionnaire that evaluates the presence of ASD based on questions from the Autism Diagnostic Interview–Revised (Lord et al., 1994). Raw scores were used as a measure of parent-reported ASD symptoms. In children under 5 years old, the SCQ has been found to have a sensitivity of .80 and a specificity of .60 to detect the presence of ASD versus nonspectrum when using a cut-off score of 11, and a sensitivity of .68 and a specificity of .74 when using a cut-off score of 15 (Corsello et al., 2007), so the cut-off score of 15 was also used to confirm a community ASD diagnosis.

Student adaptive functioning

This was measured using the Adaptive Behavior Assessment System–Second Edition (Harrison & Oakland, 2003), a parent-report questionnaire used to assess adaptive behavior in the home. Subscales include Communication, Social, and Daily Living Skills, which can be combined into a Global Adaptive Composite. Because our previous work suggests the potential role of adaptive behavior and social-emotional skills in moderating treatment effects (Nahmias et al., 2014), standard scores from the Social Composite and Global Adaptive Composite were explored as moderators.

Developmental Play Assessment (DPA)

The DPA was adapted from Lifter (2000) to measure children’s interest in playing with objects and toys. This measure was administered as described in Carter et al. (2011). Briefly, an assessor presented two standard sets of toys within the child’s reach for approximately 3.5 min each during a free-play session. Object interest (i.e., the number of toys with which children used differentiated play) at Time 1 was coded by raters trained to reliability. Interrater reliability was excellent (ICC = .93). Object interest was explored as a moderator due to its association with treatment gains in other studies (Carter et al., 2011; Yoder & Stone, 2006a, 2006b).

Classroom observation of on-task behavior

A validated measure was used to assess participants’ on-task behavior during instructional time via direct observations (Pisacreta et al., 2011). Observers were trained to 90% interrater reliability using scoring videos by a master coder before field-based observations. Ongoing interrater reliability was assessed in the field and was excellent (ICC = 0.99) (Fleiss, 1986). Participants were observed for 10 min during a typical classroom activity (e.g., academic instruction, circle time). Observers coded whether the student was on-task or off-task utilizing an interval-based coding system. The percentage of intervals the student was observed “on-task” was used in analyses.

Demographic data

These were collected from parents and classroom staff. Collections were done at Time 1 and Time 2 to use as covariates in analyses.

Analytic Strategy

All analyses were conducted using SPSS version 25. First, we conducted one-way analysis of variance (ANOVA) for continuous variables by site type. Bonferroni-corrected post hoc comparisons were utilized to explore differences by setting type on continuous variables. Chi-square tests were used to compare dichotomous variables by site type. Exact tests were used in analyses with dichotomous variables in which cell counts were below 5. Cohen’s d was used as an effect size metric. Cohen’s d can be interpreted as follows: 0.2 ≈ small, 0.5 ≈ medium, and 0.8 ≈ large (Cohen, 1988). Next, we used generalized estimating equations (GEE) with an independent correlation matrix and robust standard errors to assess change over time on the MSEL ELC across all participants, with teacher clusters at Level 2 to account for children nested within teachers. GEE was then used to examine the effects of child characteristics (e.g., baseline scores and demographic features) and setting characteristics (e.g., EPR scales and setting type) on changes in the MSEL ELC. We first examined unadjusted models to determine which putative moderators and covariates would be included in the final adjusted model. Variables with an unadjusted association with the outcome significant at p < .20 were included in the adjusted model, based on the recommendation of Hosmer et al. (2013), who found that the use of a more traditional p-value (such as .05) often fails to identify confounding variables, while the use of a higher p-value has the disadvantage of including variables of questionable importance.

Testing moderating effects

We tested for interactions between setting and EPR subscales and child characteristics that were included in the adjusted model (e.g., Time 1 MSEL RL, MSEL ELC, on-task behavior). Statistically significant interactions were then probed for regions of significance using the method described by Preacher et al. (2006). The region of significance defines the specific values of a moderator at which the regression of an outcome (i.e., MSEL ELC change) on a focal predictor (e.g., setting type and EPR scale) moves from nonsignificance to significance. If a region of significance contains no data, that region is considered uninterpretable.

Results

Sample Description

Students

As seen in the Participant Flow Diagram in Appendix A, 78 students completed the MSEL at T1 and T2, completed the EPR, and met ASD criteria on either the ADOS or SCQ, and therefore could be included in the analytic sample. Descriptive statistics for these students are presented in Table 1 and Appendix B. At the time of the EPR, 28 participants received services in an autism-only setting from 20 different providers, 25 in a mixed-disability setting from 17 different providers, and 25 in a regular early education environment setting from 25 different providers. Students were predominately male (84.6%) and most had a household income below US$40,000 (58.7%). The plurality of child participants were Black (43.6%), which is consistent with the ethnic background of students receiving preschool special education services in the school district. Ninety-seven percent met autism spectrum cut-off scores on the ADOS, with 3% meeting only on the SCQ. As seen in Table 1, some baseline participant characteristics varied by setting, including sex. Students in regular early education environments had statistically significantly higher receptive language, expressive language, developmental skills, adaptive behavior and social skills, and they had lower clinician-rated ASD symptoms than did students in the other two settings (all p-values < .05). Students in autism-only settings also were diagnosed at a younger age, on average, than children in the other two settings (both p-values < .05).

Table 1.

Characteristics of Student Participants in Community Preschool Study.

Measures and characteristics	Autism-only		Mixed-disability		Regular early education		Total		n
Measures and characteristics	M	SD	M	SD	M	SD	M	SD	n
Time 1 Age (months)	46.29	6.93	43.16	6.08	44.68	7.70	44.77	6.96	78
Time 1 MSEL Receptive Language DQ**	36.18^c	23.08	41.13^c	23.08	59.18^a,b	27.73	45.14	26.28	78
Time 1 MSEL Expressive Language DQ***	40.51^c	24.67	42.98^c	21.95	64.15^a,b	24.17	48.88	25.66	78
Time 1 MSEL Early Learning composite DQ***	46.92^c	18.76	50.39^c	15.69	65.67^a,b	19.75	54.05	19.71	78
ABAS General Adaptive composite**	55.46^c	8.37	57.83^c	9.87	66.30^a,b	14.62	59.80	12.02	70
ABAS Social composite**	58.63^c	9.45	60.39^c	9.12	69.83^a,b	15.56	62.89	12.57	70
ADOS Severity score**	7.86^b,c	1.51	6.04^a	1.51	6.60^a	2.25	6.87	1.92	78
Social Communication Questionnaire	20.19	5.84	19.33	5.44	17.88	5.60	19.17	5.65	75
DPA object interest	5.13	1.96	5.56	1.62	5.41	1.59	5.34	1.73	64
On-task behavior (percentage)	73.93	22.84	73.84	18.81	73.25	29.57	73.33	23.58	69
Age of ASD diagnosis (years)***	2.23^b,c	0.79	2.97^a	0.60	3.05^a	0.75	2.72	0.81	74
	n	%	n	%	n	%	n	%
Male**	24^b	85.71	25^{c, a}	100.00	17^b	68.00	66	84.62	78
Race/ethnicity									78
White	8	28.57	6	24.00	4	16.00	18	23.08
Black	11	39.29	12	48.00	11	44.00	34	43.59
Hispanic	2	7.14	5	20.00	4	16.00	11	14.10
More than one	6	21.43	1	4.00	5	20.00	12	15.38
Other/not reported	1	3.57	1	4.00	1	4.00	3	3.84
Household income									75
Under US$20,000	7	25.93	10	41.67	6	25.00	23	30.67
US$20,000– US$40,000	7	25.93	9	37.50	5	20.83	21	28.00
US$40,000–US$60,000	2	7.41	1	4.17	3	12.50	6	8.00
Over US$60,000	11	40.74	4	16.67	10	41.67	25	33.33

Note. MSEL = Mullen Scales of Early Learning; DQ = developmental quotients; ABAS = Adaptive Behavior Assessment System–Second Edition, ADOS = Autism Diagnostic Observation Schedule–Second Edition, EI = early intervention, DPA = Developmental Play Assessment; ASD = autism spectrum disorder.

Significantly different than autism-only setting. ^bSignificantly different than mixed-disability setting. ^cSignificantly different than regular early education environment.

Group difference p < .05. **Group difference p < .01. ***Group difference p < .001.

Classroom staff

As seen in Table 2 and Appendix C, of the 62 participating classroom staff that completed the demographic questionnaire, providers were largely female (97%) and White (66%). Less than half reported receiving regular autism training; many reported receiving training in programs and curricula that vary in their evidence base. Most staff (66%) were special education preschool teachers.

Table 2.

Characteristics of Staff in Participating Preschool Classrooms.

Characteristic	Autism-only		Mixed-disability		Regular early education		Total
Characteristic	n	%	n	%	n	%	n	%
Gender
Female	12	60.0	11	64.7	10	40.0	33	53.2
Male	0	0.0	0	0.0	1	4.0	1	1.6
Not reported	8	40.0	6	35.3	14	56.0	28	45.2
Race/ethnicity*
White	12^c	60.0	10^c	58.8	3^{a, b}	12.0	25	40.3
Black	2	10.0	2	11.8	5	20.0	9	14.5
Other	0	0.0	2	11.8	2	8.0	4	6.4
Not reported	6^c	30.0	3^c	17.6	15^a,b	60.0	24	38.7
Current role***
SPED preschool teacher	11^c	55.0	13^c	76.5	3^{a, b}	12.0	27	43.5
Gen Ed preschool teacher	0^c	0.0	0^c	0.0	6^{a, b}	24.0	6	9.7
SPED preschool teacher assistant/aide	2	10.0	2	11.8	0	0.0	4	6.5
One-to-one therapist	1	5.0	0	0.0	1	4.0	2	3.2
Gen Ed teacher assistant/aide	0	0.0	0	0.0	1	4.0	1	1.6
Administrator	0	0.0	0	0.0	1	4.0	1	1.6
Not reported	6	30.0	2^c	11.8	13^b	52.0	21	33.9
Education
Some college or voc./associates degree	1	5.0	1	5.9	4	16.0	6	9.7
College	5	25.0	8^c	47.1	3^b	12.0	16	25.8
Graduate/professional	8	40.0	4	23.5	5	20.0	17	27.4
Other	0	0.0	1	5.9	1	4.0	2	3.2
Not reported	6	30.0	3^c	17.6	12^b	48.0	21	33.9

Note: GenEd = General Education; SPED = Special Education; voc. = vocational.

Significantly different than autism-only setting. ^bSignificantly different than mixed-disability setting. ^cSignificantly different than regular early education setting.

Group difference p < .05. **Group difference p < .01. ***Group difference p < .001.

Use of Recommended Practices by Setting

As seen in Figure 1, the use of recommended practices based on the EPR varied by setting. Autism-only settings had significantly higher scores on classroom structure, classroom environment, and curriculum and instruction recommended practices than regular early education environments (all ps < .01, d = 1.7, 0.9, and 1.5, respectively). Mixed-disability settings had significantly higher scores on classroom structure (p < .001, d = 1.0) and curriculum and instruction (p < .001, d = 0.9) practices than regular early education environments, but lower classroom structure scores than autism-only settings (p = .006, d = −0.8). Regular early education environments settings had significantly higher implementation of recommended strategies to support social and peer relationships than the other two settings (all ps < .001, d = 1.6 for autism-only and 1.2 for mixed-disability).

Figure 1.

Use of recommended practices by setting type.

Developmental Outcomes

Full sample

The group means across all three groups improved in the Mullen ELC score between Time 1 (M = 54.05, SE = 2.44) and Time 2 (M = 56.41, SE = 2.72, B = 2.36, p = .03).

Unadjusted models

Table 3 presents the results of the GEE analyses predicting change in overall developmental ability, as measured by the MSEL ELC. In unadjusted analyses, only on-task behavior observed at baseline significantly predicted changes in Mullen ELC scores.

Table 3.

Unadjusted and Adjusted Generalized Estimating Equations on MSEL Early Learning Composite Change Score.

Variable	Unadjusted models		Adjusted model
Variable	Estimate	p	Estimate	p
Setting characteristics (reference group: regular early education)
Mixed-disability	−1.54	.53	−2.02	.51
Autism-only	−1.74	.54	−1.93	.62
EPR Social/Peer Relationships scale	1.90	.09	4.11	.01
EPR Classroom Environment scale	2.93	.11	0.73	.77
EPR Classroom Structure scale	−0.79	.57
EPR Curriculum and Instruction scale	−0.12	.93
EPR Positive Instructional Climate scale	0.40	.82
Not change setting	1.72	.47
Duration	−0.77	.26
Student and family characteristics
On-task behavior (percentage) Time 1	0.10	.02	0.13	<.001
MSEL ELC Time 1	−0.09	.10	−0.21	.02
MSEL Receptive Language Time 1	−0.06	.13	0.00	.97
MSEL Expressive Language Time 1	−0.02	.58
ABAS General Adaptive Composite Time 1	0.01	.90
ABAS Social Composite Time 1	−0.03	.73
SCQ	0.18	.27
ADOS Severity Score	0.19	.74
DPA Object Interest	0.17	.79
Male	−0.80	.77
Time 1 age	0.03	.86
Age at ASD diagnosis	−0.44	.76
Participate in Birth–3 EI services	1.27	.70
Maternal education	0.66	.37
Paternal education	−0.61	.53
Household income	0.20	.81
Race/ethnicity (White as reference group)
Black	0.44	.88	−1.64	.48
Hispanic	−0.23	.95	1.47	.63
More than 1	−2.96	.36	−4.70	.19
Other/Not reported	7.97	.19	8.25	.05

Note. MSEL = Mullen Scales of Early Learning; EPR = Educational Program Review; ELC = Early Learning Composite; ABAS = Adaptive Behavior Assessment System–Second Edition; SCQ = Social Communication Questionnaire; ADOS = Autism Diagnostic Observation Schedule–Second Edition; DPA = Developmental Play Assessment; ASD = autism spectrum disorder; EI = early intervention.

Adjusted models

Table 3 presents the results of the adjusted GEE analysis predicting change in MSEL ELC. In the adjusted analysis, only variables with a bivariate association statistically significant at p < .2 were included. Here, teacher use of recommended practices supporting social and peer relationships, child on-task behavior and MSEL ELC at Time 1 each statistically significantly predicted children’s developmental outcome at p < .05. Each point increase on the EPR Social/Peer Relationships scale was associated with a 4.11-point average increase in MSEL ELC change score. Each minute increase in observed on-task behavior was associated with a 1.30-point average increase in MSEL ELC change score. There were no statistically significant main effects for setting type.

Interactions

Of the initial putative moderators found in Nahmias et al. (2014), only receptive language on the MSEL met the criteria for inclusion in the adjusted model. Receptive language moderated the association between the implementation of recommended practices to support social and peer relationships (based on the EPR) and developmental changes (B = −0.12, SE = 0.03, p < .001; see Appendix D). Only the lower region of significance was interpretable (to the left of the vertical dotted line in Appendix D), meaning that for children with baseline MSEL RL developmental quotients below 58.43, children with lower baseline receptive language scores made greater gains on the MSEL ELC in settings with higher EPR Social/Peer Relationship Scale scores.

Baseline MSEL ELC also met criteria for inclusion in the adjusted model and was explored as a moderator. Baseline developmental scores moderated the association between the implementation of recommended practices to support social and peer relationships and developmental changes (B = −0.14, SE = 0.04, p < .001). Only the lower region of significance was interpretable (to the left of the vertical dotted line in Appendix D), meaning that for children with baseline MSEL ELC developmental quotients below 37.92, children with lower baseline receptive language scores made greater gains on the MSEL ELC in settings with higher EPR Social/Peer Relationship Scale scores.

On-task behavior did not moderate the association between the EPR Social/Peer Relationship Scale and MSEL ELC changes (p > .16). No tested variable moderated the association between setting type and developmental gains (p-values > .16).

Discussion

This study provides some of the most rigorous observational evidence of both expected outcomes across a variety of preschool early intervention programs in authentic school settings and the association of characteristics of these programs with developmental outcomes.

We found differences among autism-only, mixed-disability, and regular early education settings on many characteristics. Classroom staff in autism-only settings demonstrated greater implementation of recommended practices related to classroom structure, classroom environment, and curriculum and instruction than staff in regular early education environments, and greater classroom structure than in mixed-disability settings. Classroom staff in mixed-disability settings demonstrated greater implementation of recommended practices related to classroom structure and curriculum and instruction than staff in regular early education environments. Implementation of recommended practices for supporting social and peer relationships was highest in regular early education environments.

The characteristics of students in the three setting types also differed, although there was substantial overlap. Students receiving preschool special education services in regular early education environments had significantly higher receptive language, expressive language, developmental scores, adaptive behavior, and parent-reported social skills on average than students in placements primarily comprised of other students receiving special education services (i.e., autism-only and mixed-disability settings). Students in autism-only settings had higher clinician-observed autism severity and were diagnosed at a younger age than children in mixed-disability and regular early education environments. There also were more males in mixed-disability settings than in regular early education environments. These differences in child characteristics among the setting types were not observed in the Nahmias et al. (2014) study, in which there were minimal differences among children in different settings.

In this more rigorously designed prospective study, the results of our earlier work regarding the role of setting on outcome (Nahmias et al. 2014) were not replicated. However, in adjusted analysis, implementing recommended practices to support social and peer relationships was associated with positive outcome, even after controlling for other setting and child characteristics. Settings that served only children with ASD had higher implementation of almost all recommended practices, but only implementation of strategies to support social/peer relationships (more common in regular early educational environments) was significantly associated with overall developmental gains. While this finding is provocative, it is important to keep in mind that use of recommended classroom structure, curriculum and instruction, and classroom environment strategies may be associated with gains in domains not assessed in this study, such as challenging behaviors or academic readiness. A previous study (Boyd et al., 2014) found that children in community programs with at least partial implementation of recommended practices across domains, as measured by the EPR (average scores above 3 in each domain), demonstrated significant improvements in autism symptoms and severity, communication, and fine motor skills. Another possible explanation for the results of this study may be that community providers in this study need more training/support to appropriately individualize and tailor their use of recommended practices (Kasari & Smith, 2016). That is, positive EPR scores, averaged across students in the classroom, may not have captured important individual differences in how intervention was implemented.

While previous studies of interventions targeting social skills have demonstrated gains in those skills (National Autism Center, 2015; Odom et al., 2010; Steinbrenner et al., 2020), less research has explored the effect that targeting social and peer relationships has on developmental outcomes. One exception is a randomized controlled trial (RCT) of training in a school-based comprehensive treatment model (learning experiences and alternative program for preschoolers and their parents [LEAP]), which includes peer-mediated interventions as a core component. LEAP resulted in developmental gains (in addition to improved language, reduced autism symptom severity, improved social behavior and reduced problem behavior) compared with the control condition (Strain & Bovey, 2011). Supporting social and peer relationships may lead to improved attention when interacting with others, resulting in improved performance during developmental testing. Previous research with elementary school children without ASD (McClelland et al., 2000) found associations between learning-related social skills and later academic performance. In elementary school-age children with ASD (Pellecchia et al., 2016), increased social phobia symptoms were associated with decreased developmental gains. It may be that by targeting the core social deficit of ASD (American Psychiatric Association, 2013) leads to increased learning opportunities and generalization of gains by enabling children with ASD to take more advantage of interactions with peers that model age-appropriate behavior, social, play, and language skills.

Student characteristics also were associated with outcome. Specifically, increased on-task behavior predicted positive developmental changes. This suggests that children’s abilities to sustain attention and participate in the learning opportunities (or how well their early education placement facilitates this for them) are important predictors of developmental gains across setting types. Preschool students with ASD with poorer executive functioning may have worse intervention outcomes, which is consistent with some findings from children without ASD (Riggs et al., 2006). Our results are based on on-task behavior during typical classroom activities; therefore, it is unclear whether they are primarily driven by the student (e.g., the child’s executive functioning, including attention) or by the setting (e.g., teachers with better classroom management skills) characteristics. Further research is warranted to explore the relative contributions of these child and setting factors to impact developmental outcomes for preschoolers with ASD receiving community EI.

We found that greater implementation of recommended practices to support social and peer relationships was associated with greater developmental gains on average for children with ASD in our sample, and especially those lower baseline receptive language or developmental skills. This may suggest that a common educational approach of placing more severely impaired children with ASD in separate education settings away from typically developing peers (Strain, 2017) may result in a missed opportunity to maximize developmental gains. As Pellecchia and colleagues (2016) suggested, programs may need to adapt their educational approach to address social impairment and support social interactions, as these skills and opportunities may be necessary precursors to improving response to educational interventions.

Several study limitations warrant mentioning. First, students were not randomly assigned to placement, so unmeasured confounders of the association between placement and outcome cannot be ruled out (e.g., co-morbid concerns or disabilities, challenging behaviors, and broad attentional deficits). Challenges inherent with working with low-resource families (e.g., phone disconnection) and in authentic school settings (e.g., staff turnover) resulted in incomplete data collection. Our measurement of the intervention received was limited to a snapshot of the student’s programming, so we may not have fully captured the all of the intervention practices used with the student, consistency of strategy use, or intervention variability across the 9 months. Our primary outcome measure, the Early Learning Composite Developmental Quotient from the Mullen Scales of Early Learning, may not have been sensitive to more subtle or proximal developmental changes.

Despite these limitations, there are promising implications of this research. Using recommended practices to support social and peer relationships emerged as an important predictor of developmental gains for children with ASD receiving preschool special education services. Although the implementation of these strategies was greater in regular early education environments than settings that were primarily compromised of students receiving special education services, receiving preschool special education services in an inclusive classroom per se was not associated with more positive outcomes. This finding suggests that opportunities to interact with peers may be important for promoting gains, but inclusion alone is not sufficient. Consistent with the conclusions of Gutierrez et al. (2007), opportunities to interact must be paired with staff support to promote gains. In addition, while implementing strategies to support social and peer relationships was lower in autism-only and mixed-disability settings, it was not zero, suggesting opportunities to foster strategies to support social/peer engagement in those settings. Strategies by educational staff such as providing positive feedback to promote and maintain social interactions, not interfering with ongoing positive peer interactions, facilitating interactions by supporting and suggesting successful ways to interact with peers, encouraging peer partners or buddies, and considering peer relationships when assigning seats or groups can be implemented in a variety of preschool educational placement types.

Footnotes

Appendix

Appendix C.

Additional Classroom Staff Characteristics.

Characteristic	Autism-only		Mixed-disability		Regular early education		Total
Characteristic	n	%	n	%	n	%	n	%
Certifications^a
General Education	7	35.0	7	41.2	4	16.0	18	29.0
Special Education**	10^d	50.0	9^d	52.9	1^{b, c}	4.0	20	32.3
Early Childhood Education	4	20.0	6	35.3	6	24.0	16	25.8
Early Childhood Special Education	5	25.0	3	17.6	1	4.0	9	14.5
Elementary Education	0	0.0	1	5.9	1	4.0	2	3.2
Speech/Language Therapy	1	5.0	0	0.0	0	0.0	1	1.6
School Psychology	1	5.0	0	0.0	0	0.0	1	1.6
Child Development Associate	0	0.0	1	5.9	0	0.0	1	1.6
Autism Endorsement	0	0.0	0	0.0	1	4.0	1	1.6
None reported*	9	45.0	3^d	17.6	14^c	56.0	26	41.9
Programs/curricula trained in^a
Positive Behavior Supports*	14^d	70.0	12	70.6	7^b	28.0	33	53.2
Visual Supports*	12^d	60.0	8	47.1	5^b	20.0	25	40.3
Creative Curriculum	6	30.0	8	47.1	6	24.0	20	32.3
Discrete Trial Training***	13^c,d	65.0	4^b	23.5	2^b	8.0	19	30.6
Functional Behavioral Assessment**	11^c,d	55.0	4^b	23.5	3^b	12.0	18	29.0
PECS	8	40.0	5	29.4	2	8.0	15	24.2
Verbal Behavior*	8^c	40.0	2^b	11.8	4	16.0	14	22.6
Incidental Teaching*	8^c	40.0	1^b	5.9	4	16.0	13	21.0
Pivotal Response Training	5	25.0	4	23.5	2	8.0	11	17.7
Functional Communication Training	6	30.0	2	11.8	2	8.0	10	16.1
TEACCH	3	15.0	1	5.9	2	8.0	6	9.7
Floortime/DIR/RDI	2	10.0	2	11.8	1	4.0	5	8.1
High/Scope**	5^c	25.0	0^b	0.0	0	0.0	5	8.1
Peer-mediated Instruction	1	5.0	0	0.0	1	4.0	2	3.2
None reported**	6^d	30.0	3^d	17.6	18^{b,c b,c}	72.0	27	43.5
Autism training^a
No autism specific training	0	0.0	1	5.9	4	16.0	5	8.1
A couple workshops or courses	1	5.0	3	17.6	1	4.0	5	8.1
Some workshops or courses	1	5.0	4	23.5	2	8.0	7	11.3
Regular autism training	7	35.0	7	41.2	5	20.0	19	30.6
Certification or degree related to autism training	4	20.0	2	11.8	1	4.0	7	11.3
Not reported*	7	35.0	2^d	11.8	12^c	48.0	21	33.9
	M	SD	M	SD	M	SD	M	SD
Years in current position (n = 39)	3.4	3.9	3.6	5.3	7.0	10.5	4.5	7.0
Age (n = 33)	29.9	4.5	35.8	13.4	40.7	14.7	34.3	11.3

Note. DIR= the developmental, individual difference, relationship-based model; PECS = picture exchange communication system; RDI = relationship development intervention; TEACCH = treatment and education of autistic and related communications handicapped children.

Percentages do not add up to 100% because some providers selected multiple responses. ^bSignificantly different than Autism-Only setting. ^cSignificantly different than Mixed-Disability setting. ^dSignificantly different than Regular Early Education setting.

Group difference p < .05. **Group difference p < .01. ***Group difference p < .001.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors thank our funders, Autism Speaks (Grant #8874; PI Mandell) and the McMorris Autism Foundation (PIs Nahmias & Yerys). The research reported here was also supported in part by the Institute of Education Sciences, U.S. Department of Education, through grant #R305B090015 to the University of Pennsylvania. The opinions expressed are those of the authors and do not represent the views of the Institute or the U.S. Department of Education.

ORCID iD

Allison S. Nahmias

References

American Psychiatric Association (Ed.). (2013). Diagnostic and Statistical Manual of Mental Disorders: DSM-5 (5th ed.).

Boyd

B. A.

Hume

McBee

M. T.

Alessandri

Gutierrez

Johnson

. . .Odom

S. L.

(2014). Comparative efficacy of LEAP, TEACCH and non-model-specific special education programs for preschoolers with autism spectrum disorders. Journal of Autism and Developmental Disorders, 44(2), 366–380. https://doi.org/10.1007/s10803-013-1877-9

Carter

A. S.

Messinger

D. S.

Stone

W. L.

Celimli

Nahmias

A. S.

Yoder

(2011). A randomized controlled trial of Hanen’s “More Than Words” in toddlers with early autism symptoms. Journal of Child Psychology and Psychiatry, 52(7), 741–752. https://doi.org/10.1007/s10803-013-1877-9

Cohen

(1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.). Lawrence Erlbaum.

Corsello

Hus

Pickles

Risi

Cook

E. H.

Leventhal

B. L.

Lord

(2007). Between a ROC and a hard place: Decision making and making decisions about using the SCQ. Journal of Child Psychology and Psychiatry, 48(9), 932–940. https://doi.org/10.1111/j.1469-7610.2007.01762

Eapen

Črnčec

Walter

(2013). Clinical outcomes of an early intervention program for preschool children with autism spectrum disorder in a community group setting. BMC Pediatrics, 13(1), Article 3. https://doi.org/10.1186/1471-2431-13-3

Eldevik

Hastings

R. P.

Hughes

J. C.

Jahr

Eikeseth

Cross

(2010). Using participant data to extend the evidence base for intensive behavioral intervention for children with autism. American Journal on Intellectual and Developmental Disabilities, 115(5), 381–405. https://doi.org/10.1352/1944-7558-115.5.381

Fernell

Hedvall

Å.

Westerlund

Höglund Carlsson

Eriksson

Barnevik Olsson

Gillberg

(2011). Early intervention in 208 Swedish preschoolers with autism spectrum disorder. A prospective naturalistic study. Research in Developmental Disabilities, 32(6), 2092–2101. https://doi.org/10.1016/j.ridd.2011.08.002

Flanagan

H. E.

Perry

Freeman

N. L.

(2012). Effectiveness of large-scale community-based intensive behavioral intervention: A waitlist comparison study exploring outcomes and predictors. Research in Autism Spectrum Disorders, 6(2), 673–682. https://doi.org/10.1016/J.RASD.2011.09.011

10.

Fleiss

J. L.

(1986). The Design and Analysis of Clinical Experiments. John Wiley.

11.

Gutierrez

Jr. Hale

M. N.

Gossens-Archuleta

Sobrino-Sánchez

(2007). Evaluating the social behavior of preschool children with autism in an inclusive playground setting. International Journal of Special Education, 22(3), 26–30.

12.

Harris

S. L.

Handleman

J. S.

Kristoff

Bass

Gordon

(1990). Changes in language development among autistic and peer children in segregated and integrated preschool settings. Journal of Autism and Developmental Disorders, 20(1), 23–31. https://doi.org/10.1007/BF02206854

13.

Harrison

Oakland

(2003). Adaptive Behavior Assessment system (2nd ed.). The Psychological Corporation.

14.

Hosmer

Lemeshow

Sturdivant

(2013). Applied Logistic Regression (3rd ed.). John Wiley.

15.

Hume

Boyd

McBee

Coman

Gutierrez

Shaw

Odom

(2011). Assessing implementation of comprehensive treatment models for young children with ASD: Reliability and validity of two measures. Research in Autism Spectrum Disorders, 5(4), 1430–1440. https://doi.org/10.1016/j.rasd.2011.02.002

16.

Kaale

Smith

Sponheim

(2012). A randomized controlled trial of preschool-based joint attention intervention for children with autism: Preschool-based JA-intervention for children with autism. Journal of Child Psychology and Psychiatry, 53(1), 97–105. https://doi.org/10.1111/j.1469-7610.2011.02450.x

17.

Kasari

Paparella

Freeman

Jahromi

L. B.

(2008). Language outcome in autism: Randomized comparison of joint attention and play interventions. Journal of Consulting and Clinical Psychology, 76(1), 125–137. https://doi.org/10.1037/0022-006X.76.1.125

18.

Kasari

Smith

(2013). Interventions in schools for children with autism spectrum disorder: Methods and recommendations. Autism, 17(3), 254–267. https://doi.org/10.1177/1362361312470496

19.

Kasari

Smith

(2016). Forest for the trees: Evidence-based practices in ASD. Clinical Psychology: Science and Practice, 23(3), 260–264. https://doi.org/10.1111/cpsp.12161

20.

Lifter

(2000). Linking assessment to intervention for children with developmental disabilities or at-risk for Developmental delay: The Developmental Play Assessment (DPA) instrument. In Gitlin-Weiner

Sandgrund

Schaefer

C.E.

(Eds.), Play Diagnosis and Assessment (2nd ed., pp. 228–261). John Wiley & Sons.

21.

Lord

Rutter

DiLavore

Risi

Gotham

Bishop

(2012). Autism Diagnostic Observation Schedule–2nd edition (ADOS-2). Western Psychological Corporation.

22.

Lord

Rutter

Le Couteur

(1994). Autism Diagnostic Interview-Revised: A revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. Journal of Autism and Developmental Disorders, 24(5), 659–685. https://doi.org/10.1007/BF02172145

23.

Magiati

Charman

Howlin

(2007). A two-year prospective follow-up study of community-based early intensive behavioural intervention and specialist nursery provision for children with autism spectrum disorders. Journal of Child Psychology and Psychiatry, 48(8), 803–812. https://doi.org/10.1111/j.1469-7610.2007.01756.x

24.

McClelland

M. M.

Morrison

F. J.

Holmes

D. L.

(2000). Children at risk for early academic problems: The role of learning-related social skills. Early Childhood Research Quarterly, 15(3), 307–329. https://doi.org/10.1016/S0885-2006(00)00069-7

25.

Mullen

E. M.

(1995). Mullen Scales of Early Learning manual. American Guidance Service.

26.

Nahmias

A. S.

Kase

Mandell

D. S.

(2014). Comparing cognitive outcomes among children with autism spectrum disorders receiving community-based early intervention in one of three placements. Autism, 18(3), 311–320. https://doi.org/10.1177/1362361312467865

27.

Nahmias

A. S.

Pellecchia

Stahmer

A. C.

Mandell

D. S.

(2019). Effectiveness of community-based early intervention for children with autism spectrum disorder: A meta-analysis. Journal of Child Psychology and Psychiatry, 60(11), 1200–1209. https://doi.org/10.1111/jcpp.13073

28.

National Autism Center. (2015). Findings and conclusions: National Standards Project, Phase 2. Randolph, MA: Author.

29.

National Research Council. (2001). Educating Children with Autism. National Academies Press.

30.

Odom

S. L.

Collet-Klingenberg

Rogers

S. J.

Hatton

D. D.

(2010). Evidence-based practices in interventions for children and youth with autism spectrum disorders. Preventing School Failure: Alternative Education for Children and Youth, 54(4), 275–282. https://doi.org/10.1080/10459881003785506

31.

Pellecchia

Connell

J. E.

Kerns

C. M.

Xie

Marcus

S. C.

Mandell

D. S.

(2016). Child characteristics associated with outcome for children with autism in a school-based behavioral intervention. Autism, 20(3), 321–329. https://doi.org/10.1177/1362361315577518

32.

Perry

Cummings

Geier

J. D.

Freeman

N. L.

Hughes

Managhan

Williams

(2011). Predictors of outcome for children receiving intensive behavioral intervention in a large, community-based program. Research in Autism Spectrum Disorders, 5(1), 592–603. https://doi.org/10.1016/j.rasd.2010.07.003

33.

Pisacreta

Tincani

Connell

J. E.

Axelrod

(2011). Increasing teachers’ use of a 1:1 praise-to-behavior correction ratio to decrease student disruption in general education classrooms. Behavioral Interventions, 26(4), 243–260. https://doi.org/10.1002/bin.341

34.

Preacher

K. J.

Curran

P. J.

Bauer

D. J.

(2006). Computational tools for probing interactions in multiple linear regression, multilevel modeling, and latent curve analysis. Journal of Educational and Behavioral Statistics, 31(4), 437–448. https://doi.org/10.3102/10769986031004437

35.

Professional Development in Autism Center. (2008). PDA program assessment. PDA Center at University of Washington.

36.

Reichow

Barton

E. E.

Boyd

B. A.

Hume

(2012). Early intensive behavioral intervention (EIBI) for young children with autism spectrum disorders (ASD). The Cochrane Database of Systematic Reviews, 10, CD009260. https://doi.org/10.1002/14651858.CD009260.pub2

37.

Remington

Hastings

R. P.

Kovshoff

degli Espinosa

Jahr

Brown

Ward

(2007). Early intensive behavioral intervention: Outcomes for children with autism and their parents after two years. American Journal on Mental Retardation, 112(6), 418–438. https://doi.org/10.1352/0895-8017(2007)112[418:EIBIOF]2.0.CO;2

38.

Riggs

N. R.

Jahromi

L. B.

Razza

R. P.

Dillworth-Bart

J. E.

Mueller

(2006). Executive function and the promotion of social-emotional competence. Journal of Applied Developmental Psychology, 27(4), 300–309. https://doi.org/10.1016/j.appdev.2006.04.002

39.

Rutter

Bailey

Lord

(2003). The social communication questionnaire: Manual. Western Psychological Services.

40.

Shire

S. Y.

Kasari

(2014). Train the trainer effectiveness trials of behavioral intervention for individuals with autism: A systematic review. American Journal on Intellectual and Developmental Disabilities, 119(5), 436–451. https://doi.org/10.1352/1944-7558-119.5.436

41.

Stahmer

A. C.

Aarons

G. A.

(2009). Attitudes toward adoption of evidence-based practices: A comparison of autism early intervention providers and children’s mental health providers. Psychological Services, 6(3), 223–234. https://doi.org/10.1037/a0010738

42.

Stahmer

A. C.

Collings

N. M.

Palinkas

L. A.

(2005). Early intervention practices for children with autism: Descriptions from community providers. Focus on Autism and Other Developmental Disabilities, 20(2), 66–79. https://doi.org/10.1177/10883576050200020301

43.

Steinbrenner

J. R.

Hume

Odom

S. L.

Morin

K. L.

Nowell

S. W.

Tomaszewski

Savage

M. N.

(2020). Evidence-based practices for children, youth, and young adults with autism children, youth, and young adults with autism. The University of North Carolina at Chapel Hill, Frank Porter Graham Child Development Institute, National Clearinghouse on Autism Evidence and Practice Review Team.

44.

Strain

P. S.

(2017). Four-year follow-up of children in the LEAP randomized trial: Some planned and accidental findings. Topics in Early Childhood Special Education, 37(2), 121–126. https://doi.org/10.1177/0271121417711531

45.

Strain

P. S.

Bovey

E. H.

(2011). Randomized, controlled trial of the LEAP model of early intervention for young children with autism spectrum disorders. Topics in Early Childhood Special Education, 31(3), 133–154. https://doi.org/10.1177/0271121411408740

46.

Suhrheinrich

Rieth

Dickson

Roesch

Stahmer

(2020). Classroom Pivotal Response Teaching: Teacher training outcomes of a community efficacy trial. Teacher Education and Special Education, 43, 215–234. https://doi.org/10.1177/0888406419850876

47.

Vivanti

Dissanayake

Duncan

Feary

Capes

Upson

Hudry

(2019). Outcomes of children receiving group-Early Start Denver Model in an inclusive versus autism-specific setting: A pilot randomized controlled trial. Autism, 23(5), 1165–1175. https://doi.org/10.1177/1362361318801341

48.

Vivanti

Paynter

Duncan

Fothergill

Dissanayake

Rogers

S. J.

, & the Victorian ASELCC Team. (2014). Effectiveness and feasibility of the Early Start Denver Model implemented in a group-based community childcare setting. Journal of Autism and Developmental Disorders, 44(12), 3140–3153. https://doi.org/10.1007/s10803-014-2168-9

49.

Weitlauf

A. S.

McPheeters

M. L.

Peters

Sathe

Travis

Aiello

Warren

(2014). Therapies for children with autism spectrum disorder. Agency for Healthcare Research and Quality (US). https://www.ncbi.nlm.nih.gov/books/NBK241444/

50.

Wood

J. J.

McLeod

B. D.

Klebanoff

Brookman-Frazee

(2015). Toward the implementation of evidence-based interventions for youth with autism spectrum disorders in schools and community agencies. Behavior Therapy, 46(1), 83–95. https://doi.org/10.1016/j.beth.2014.07.003

51.

Yoder

Stone

W. L.

(2006a). A randomized comparison of the effect of two prelinguistic communication interventions on the acquisition of spoken communication in preschoolers with ASD. Journal of Speech, Language, and Hearing Research, 49(4), 698–711. https://doi.org/10.1044/1092-4388(2006/051)

52.

Yoder

Stone

W. L.

(2006b). Randomized comparison of two communication interventions for preschoolers with autism spectrum disorders. Journal of Consulting and Clinical Psychology, 74(3), 426–435. https://doi.org/10.1037/0022-006X.74.3.426