View in EDS HTML Full Text PDF Full Text

Profiles and Longitudinal Growth Trajectories of Teacher-Rated Academic Skills and Enablers in Autistic Children and Adolescents

Saved in:
Bibliographic Details
Title: Profiles and Longitudinal Growth Trajectories of Teacher-Rated Academic Skills and Enablers in Autistic Children and Adolescents
Language: English
Authors: Dawn Adams (ORCID 0000-0001-8001-0126), Matt Stainer, Kate Simpson, Jessica Paynter, Marleen Westerveld
Source: Journal of Autism and Developmental Disorders. 2025 55(1):267-283.
Availability: Springer. Available from: Springer Nature. One New York Plaza, Suite 4600, New York, NY 10004. Tel: 800-777-4643; Tel: 212-460-1500; Fax: 212-460-1700; e-mail: customerservice@springernature.com; Web site: https://link.springer.com/
Peer Reviewed: Y
Page Count: 17
Publication Date: 2025
Document Type: Journal Articles
Reports - Research
Education Level: Elementary Education
Secondary Education
Descriptors: Autism Spectrum Disorders, Elementary School Students, Secondary School Students, Student Motivation, Learner Engagement, Interpersonal Competence, Study Skills, Individualized Education Programs, Elementary School Teachers, Secondary School Teachers, Student Evaluation, Individualized Instruction, Learning Trajectories
DOI: 10.1007/s10803-023-06186-1
ISSN: 0162-3257
1573-3432
Abstract: In non-autistic children, academic skills are associated with academic enablers (motivation, engagement, study/interpersonal skills), but few studies have explored these in autistic children. This study identified profiles of academic skills and enablers in autistic students and explored the trajectory of each profile over time. Teachers completed the Academic Competences Evaluation Scales for autistic children in primary and secondary educational settings annually for 5 years. Latent profile analysis identified six profiles in the primary/younger cohort and seven in the secondary/older cohort. Whilst some profiles showed relative stability across skills and enablers, others profiles were more variable. The profiles remained stable and significantly different from each other over time, with no profile x time interactions identified. Autistic children may show variability across their academic skills and enablers. This highlights the importance of understanding each individual student and their profile of strengths and challenges when planning supports.
Abstractor: As Provided
Entry Date: 2025
Accession Number: EJ1460676
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwEv0xaY7QY2qG0fWayxYfxcAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDLqHRV_ALrUgR7fNggIBEICBm24CkLr1Xn6m1BJzoUQpyxp6JezYsZhymHRyZ29tz25E319kOtH_3Zc9nFS0IXwxKA4OBTQe1S0fE0VEGAyEGmtFLNgSlU2AWTveK3kf988tTTC8ycCWOPoXoo-6OMaUVnZFYuj8nDFRb7QTtgYDE0R0ZEnZNXf0qyR-mqwo5W-qZKx4lSvjfT1VkymjA6yyyoTIC_DOSkig5fZD
Text:
  Availability: 1
  Value: <anid>AN0182844507;aut01jan.25;2025Feb10.03:00;v2.2.500</anid> <title id="AN0182844507-1">Profiles and Longitudinal Growth Trajectories of Teacher-Rated Academic Skills and Enablers in Autistic Children and Adolescents </title> <p>In non-autistic children, academic skills are associated with academic enablers (motivation, engagement, study/interpersonal skills), but few studies have explored these in autistic children. This study identified profiles of academic skills and enablers in autistic students and explored the trajectory of each profile over time. Teachers completed the Academic Competences Evaluation Scales for autistic children in primary and secondary educational settings annually for 5 years. Latent profile analysis identified six profiles in the primary/younger cohort and seven in the secondary/older cohort. Whilst some profiles showed relative stability across skills and enablers, others profiles were more variable. The profiles remained stable and significantly different from each other over time, with no profile × time interactions identified. Autistic children may show variability across their academic skills and enablers. This highlights the importance of understanding each individual student and their profile of strengths and challenges when planning supports.</p> <p>Keywords: Autism; School; Learning; Longitudinal; Behaviour; Education Specialist Studies In Education</p> <p>Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s10803-023-06186-1.</p> <hd id="AN0182844507-2">Introduction</hd> <p>Many autistic children do not achieve academic grades reflective of their full potential (Ashburner et al., [<reflink idref="bib2" id="ref1">2</reflink>]; Mayes & Calhoun, [<reflink idref="bib25" id="ref2">25</reflink>]). To date, reviews have failed to identify consistent causes or factors associated with academic underachievement in autistic students (Keen et al., [<reflink idref="bib18" id="ref3">18</reflink>]; Kwok et al., [<reflink idref="bib20" id="ref4">20</reflink>]). Such findings are essential to identify students at risk and to provide targeted supports. Research into predictors may be complicated by significant within- and between-individual heterogeneity. Such heterogeneity has been noted within a single academic skill (e.g., reading; McIntyre et al., [<reflink idref="bib28" id="ref5">28</reflink>]; Solari et al., [<reflink idref="bib36" id="ref6">36</reflink>]) or across multiple academic skills (e.g., Bullen et al., [<reflink idref="bib3" id="ref7">3</reflink>]). Thus, the fact that much of the research has focussed on group-level comparisons fails to capture the heterogeneity of this population and may not yield meaningful predictors for individualising or differentiation of supports. Providing a more nuanced approach, some studies (e.g., Mayes & Calhourn, [<reflink idref="bib26" id="ref8">26</reflink>]) have begun to move beyond over- or under-achievement to explore profiles of academic achievement for autistic children and the factors associated with each profile. An improved understanding of autistic students' profiles of strengths and needs across a range of academic skills, alongside the factors associated with these profiles, could help inform tailored supports and strategies which support academic achievement for autistic students (Bullen et al., [<reflink idref="bib3" id="ref9">3</reflink>]).</p> <hd id="AN0182844507-3">Profiles of Academic Skills in Autistic Children</hd> <p>One approach to exploring profiles of academic achievement has been to focus on a single academic skill. Such approaches have been used with differing results within the area of reading. For example, Johnels et al. ([<reflink idref="bib14" id="ref10">14</reflink>]) classified 47% of their 8 year-old sample (<emph>n</emph> = 53) as "poor" readers (i.e., poor single-word reading/decoding and reading comprehension), 19% as "hyperlexic/poor comprehenders" and 34% as "skilled readers". However, McIntyre et al. ([<reflink idref="bib28" id="ref11">28</reflink>]) identified four profiles: those who consistently performed in the average range (32.1%), those with a particular difficulty in comprehension (20.6%), those who showed global disturbance (i.e., lower performance across all areas; 33.2%), and those with severe global disturbance (14.1%). Using latent profile analysis, Solari et al. ([<reflink idref="bib36" id="ref12">36</reflink>]) identified four profiles which were defined by performance on 12 language and reading skills assessments: average (31%), comprehension disturbance (18%), below average/intact receptive vocabulary (30%) and global disturbance (21%).</p> <p>A second approach to exploring academic achievement is to identify profiles across multiple academic skills or areas. Using cluster analysis, Wei et al. ([<reflink idref="bib39" id="ref13">39</reflink>]) identified four profiles in their sample of 130 6–9 year-old autistic children: homogenous high achievement in math and reading (39%), homogenous low achievement (32%) and unequal achievement excelling in either reading (hyperlexia; 9%) or calculation (hypercalculia; 20%). In contrast, Chen et al. ([<reflink idref="bib5" id="ref14">5</reflink>]) identified only two subgroups in their sample of 114 autistic males aged 7–12 years, all with IQs above 67: low achieving (37%, lower maths skills compared to reading skills) and high achieving (63%, higher maths skills than reading skills). Bullen et al. ([<reflink idref="bib3" id="ref15">3</reflink>]) reported on a sample of 78 autistic participants without an intellectual disability, aged 8–16 years. Similar to Wei et al., their hierarchical cluster analysis suggested two profiles: a low average academic achievement group (70%; average calculation but below average problem solving, reading comprehension and reading fluency) and high average academic achievement (above average calculation and problem solving, average reading comprehension and fluency). These differing numbers and descriptions of profiles may be due to sample differences or methodological differences, or they may highlight the broad heterogeneity in skills across autistic students across different samples.</p> <p>Studies that have explored the factors associated with specific profiles of academic skills and academic achievement in autistic students have shown inconsistent results. For example, language and communication skills differed among the three profiles of reading skills in Johnels et al. ([<reflink idref="bib14" id="ref16">14</reflink>]). However, conversational communication did not differ among the four profiles of literacy and mathematical skills in Wei et al. ([<reflink idref="bib39" id="ref17">39</reflink>]). Similarly, autism characteristics differed between two of the three reading profiles in Johnels et al. ([<reflink idref="bib15" id="ref18">15</reflink>]) but not between Solari et al.'s ([<reflink idref="bib36" id="ref19">36</reflink>]) four and Chen et al.'s ([<reflink idref="bib5" id="ref20">5</reflink>]) two profiles. Few studies have explored whether family demographics differ between profiles of academic skills. Those that have (e.g., Bullen et al., [<reflink idref="bib3" id="ref21">3</reflink>]; Wei et al., [<reflink idref="bib39" id="ref22">39</reflink>]) highlight possible differences in family socioeconomic status across profiles. Further research to explore possible differences in child, family and other factors between academic profiles of autistic students is therefore warranted.</p> <p>As academic achievement is based on more than just reading/language and numeracy skills, some studies have identified profiles of academic skills with other cognitive factors. In doing so, they have shown differing relationships between such factors and specific academic skills. For example, McDougal et al.'s ([<reflink idref="bib27" id="ref23">27</reflink>]) cluster analysis highlighted that mathematical skills are more impacted by poor divided attention than are literacy skills. Zaidman-Zait et al. ([<reflink idref="bib42" id="ref24">42</reflink>]) considered academic skills (based on direct assessments), engagement (based on teacher ratings) and social functioning (based on parent ratings) within their latent profile analysis. Based on data from 178 autistic children aged 10–11, they identified four profiles: elevated academic and elevated social school functioning, low academic and low social school functioning, low academic but average social school functioning, and average academic but low social school functioning. However, these profiles could only be derived from children who were able to complete direct assessments with the researchers, limiting the generalisability of such profiles across the broad range of abilities reported in autistic individuals. This approach of identifying profiles of academic skills with other classroom skills, some of which could be considered academic enablers, provides additional information for avenues for both classroom and broader intervention and supports.</p> <hd id="AN0182844507-4">Academic Enablers in Autistic Students</hd> <p>Academic enablers describe the attitudes and behaviours that facilitate a student's participation in, and benefit from, academic instruction. Academic enablers include motivation, engagement, study skills, and interpersonal skills (DiPerna et al., [<reflink idref="bib10" id="ref25">10</reflink>]). Academic achievement has been shown to be impacted by academic enablers in neurotypical (DiPerna & Elliott, [<reflink idref="bib8" id="ref26">8</reflink>]) and autistic students (Keen et al., [<reflink idref="bib17" id="ref27">17</reflink>]). For the purposes of this study, each of the academic enablers is defined as follows, based on the descriptions within Keen et al. ([<reflink idref="bib17" id="ref28">17</reflink>]). Motivation refers to the student's approach, persistence, and level of interest in academic subjects. Engagement relates to a student's attention and active participation in classroom activities. Study skills are behaviours that facilitate the processing of new material and test-taking. Interpersonal skills involve cooperative learning behaviour necessary to interact with others.</p> <p>In one of the only studies to specifically evaluate the collective impact of academic enablers in autistic students, Keen et al. ([<reflink idref="bib17" id="ref29">17</reflink>]) reported on the academic competence evaluation scales (ACES) in 133 autistic students in two groups: Kindergarten to Grade 2 and Grades 6–8. Students in the younger group in both inclusive and special educational settings, and students in the older group in special educational settings, scored significantly below the published norms for neurotypical students on all academic skills and academic enablers (except study skills). Students in the older group who were in mainstream (i.e., integrated/general education) settings scored significantly lower than the neurotypical published norms on all academic skills and on the academic enablers of interpersonal skills and motivation. However, there was no significant difference in the academic enabler skills of engagement and study skills. Importantly, specific enablers contributed unique and significant amounts of variance to different academic skills. These results suggest that the relationship between academic enablers and academic skills warrants further exploration.</p> <hd id="AN0182844507-5">The Current Study</hd> <p>Although multiple studies report that autistic students at a group level tend to achieve below their potential, there have been few consistent predictors or factors to explain the reasons behind this (Keen et al., [<reflink idref="bib18" id="ref30">18</reflink>]). As such, researchers (e.g., Chen et al., [<reflink idref="bib5" id="ref31">5</reflink>]; Zaidman-Zait et al., [<reflink idref="bib42" id="ref32">42</reflink>]) have recently moved away from a classification of underachievement and towards developing more nuanced descriptions of within-group profiles of academic skills. Such an approach identifies groups of students with relative strengths and challenges across academic skills. Once profiles are established, it provides the potential to identify whether any child or family factors are related to specific profiles. This then enables the identification of which students may be more likely to underachieve and, therefore, may benefit from tailored (early) supports in specific areas. However, caution must be applied when drawing causal conclusions based on cross-sectional research due to potential cohort effects (Johnels et al., [<reflink idref="bib14" id="ref33">14</reflink>]) and the limited information of the stability of profiles over time. Longitudinal studies are required to firstly identify specific profiles within the cohort and then to document each profile's trajectory over time. Given that a scoping review of trajectory research in autistic children (Gentles et al., [<reflink idref="bib12" id="ref34">12</reflink>]) identified only one study which reported on academic achievement, and that study reported on a single subtest only, there is a critical need for trajectory research in this area.</p> <p>To date, the studies exploring academic profiles in autistic children have relied upon direct, standardised tests of academic skills. Although scores on direct tests of academic skills provide insight into children's competency in areas such as reading/language and mathematics, they do not yield information on children's capacity to demonstrate such competencies in the classroom (Milgramm et al., [<reflink idref="bib29" id="ref35">29</reflink>]). Reliance on direct testing may also result in the exclusion of the many children who may not be able to complete such testing, or who perform at the floor on the measures. Both these limitations can be addressed by asking teachers to rate the child's skills compared to their peers. Teachers are the most frequent and consistent observers of children in the context of the classroom and can, therefore, provide informative perspectives on their learning and engagement (McDougal et al., [<reflink idref="bib27" id="ref36">27</reflink>]). Large datasets of over 5000 ratings of academic achievement have shown teacher ratings of student performance to be as reliable, stable, and heritable as test scores at every stage of the educational experience (Rimfeld et al., [<reflink idref="bib34" id="ref37">34</reflink>]). Not only can teacher assessments of student skills be provided for students regardless of their ability level, but they can also be reflective of student achievement across a broader timespan (Oswald et al., [<reflink idref="bib31" id="ref38">31</reflink>]). Teacher assessments can also provide information on students' non-cognitive, classroom-related behaviours—elements that would be difficult to measure through direct standardised assessments. Therefore, this study will use teacher ratings of the academic skills and academic enablers of autistic children so as to further our understanding of classroom-based profiles of learning in autistic students with a range of abilities.</p> <p>To the authors' knowledge, no research to date has used teacher ratings to characterise profiles of broader academic skills and academic enablers in autistic children. Similarly, no study has used longitudinal data to identify whether these different profiles of academic skills and academic enablers impact upon the later trajectories of academic skills. If academic enablers are associated with different academic skills or outcomes, they can be considered priority targets for intervention either prior to or early into schooling. Therefore, the research questions posed in this study were:</p> <p></p> <ulist> <item> Are there specific profiles of teacher-rated academic skills and enablers in autistic children? If so, are these the same in children beginning primary and secondary school?</item> <p></p> <item> Are these profiles of academic skills and enablers associated with child, family, or school characteristics (e.g., autism characteristics, child behaviour, family income)?</item> <p></p> <item> What are the trajectories of academic skills for each profile over time?</item> </ulist> <p>Based upon previous work on academic profiles (e.g., Zaidman-Zait et al., [<reflink idref="bib42" id="ref39">42</reflink>]), we hypothesised that, because of the wide heterogeneity of strengths and skills in school-age autistic children, more than one distinct profile of academic skills and enablers would emerge. Given the limited and mixed findings to date, no specific hypotheses were made in relation to child and family characteristics associated with academic profiles.</p> <hd id="AN0182844507-6">Methods</hd> <p>This study used data collected as part of the Longitudinal Study of Australian students with Autism (LASA), a study which focussed on academic and participation outcomes in autistic children across 6 years. The full protocol for this study is published Roberts et al. ([<reflink idref="bib32" id="ref40">32</reflink>]) Ethical approval for this study was obtained in 2015/16 from Human Research Ethics Committees (HRECs) at all participating universities and health authorities (see Roberts et al. ([<reflink idref="bib32" id="ref41">32</reflink>]) for all approval numbers). All parent participants gave informed, written consent to participate in the larger longitudinal study and also provided consent for the study team to contact the child's teacher to provide data for this study. All school principals provided consent for the study team to contact the child's teacher. If they wished to participate, the teacher provided consent to participate in the study.</p> <p>For transparency and clarity, the data from Keen et al. ([<reflink idref="bib17" id="ref42">17</reflink>]) are included within this study. This study differs from Keen et al. ([<reflink idref="bib17" id="ref43">17</reflink>]) as (a) Keen et al. was only looking at a single time point and this study is longitudinal and (b) Keen et al. looked at whether academic enablers predicted academic skills, whereas this study looks at identifying profiles across both academic skills and enablers.</p> <hd id="AN0182844507-7">Recruitment Procedure</hd> <p>The full recruitment procedure for the larger LASA study is described in the published protocol Roberts et al. ([<reflink idref="bib32" id="ref44">32</reflink>]). In brief, parents of autistic children who were aged either 4–5 years or 9–10 years were invited to participate in a 6 year longitudinal study through advertisements on social media and/or clinics across Australia. On enrolment into the study, all participants were requested to provide copies of their diagnostic reports. Participants were also asked to complete a measure of autism characteristics (social communication questionnaire (SCQ); Rutter et al., [<reflink idref="bib35" id="ref45">35</reflink>]) to confirm their child's diagnosis. The sample size of the larger longitudinal study was not set in advance to enable all participants who met inclusion criteria to participate.</p> <p>Data collection from schools began in the second year of the longitudinal study and was annual from thereon, so as to collect data across primary school (for the younger cohort) and secondary school (for the older cohort). If parents consented to information being collected from their child's school, principals and teachers were invited to complete an online questionnaire survey. All measures completed by teachers are described in the published protocol. This study focusses on data provided by the child's teacher to document the profiles and trajectories of academic skills over time. For clarity, within this study, time points are labelled Time 1–5, but these correspond to Years 2–6 of the broader longitudinal study. The final year of data collection was during the global COVID-19 pandemic. Home and school data were collected during a period when lockdown or school closures were not in place and had not been in place for at least 4 weeks.</p> <hd id="AN0182844507-8">Participants</hd> <p>For the initial teacher data collection time (Time 1), a total of 211 parents gave consent for the study team to contact their child's school. Consent was gained from principals before teachers were contacted. Following telephone calls and follow-up emails, 122 (57.8%) principals consented to the teacher receiving an invitation to participate. From the 122 teacher invitations that were sent out, 100 teachers (81.9%) completed the online questionnaire. This process was repeated each year, with 88 teachers participating at Time 2, 126 at Time 3, 131 at Time 4, and 108 at Time 5. In total, data from 216 autistic children are reported in this study. Of these, 44 children had full or partial teacher questionnaire data from one time point, 56 from two time points, 55 from three time points, 32 from four time points and 19 from all five possible time points across the study. Demographics for the child participants at Time 1 are presented in the upper half of Tables 1 and 2. Teacher demographics for the sample at each time point are presented in the lower half of each table.</p> <p>Table 1 Student and teacher/school demographics (younger cohort)</p> <p> <ephtml> <table frame="hsides" rules="groups"><thead><tr><th align="left" /><th align="left"><p>Time 1</p></th><th align="left"><p>Time 2</p></th><th align="left"><p>Time 3</p></th><th align="left"><p>Time 4</p></th><th align="left"><p>Time 5</p></th></tr></thead><tbody><tr><td align="left"><p>Student</p></td><td align="left" /><td align="left" /><td align="left" /><td align="left" /><td align="left" /></tr><tr><td align="left"><p> Gender: male</p></td><td align="left"><p>82%</p></td><td align="left"><p>81%</p></td><td align="left"><p>90%</p></td><td align="left"><p>93%</p></td><td align="left"><p>90%</p></td></tr><tr><td align="left"><p> Age in months mean (SD, range)</p></td><td align="left"><p>75.11 (7.36, 60–90)</p></td><td align="left"><p>85.28 (7.11, 73–102)</p></td><td align="left"><p>97.74 (7.58, 84–113)</p></td><td align="left"><p>110.60 (9.68, 97–126)</p></td><td align="left"><p>122.94 (7.34, 110–135)</p></td></tr><tr><td align="left"><p> Mainstream school</p></td><td align="left"><p>66.03%</p></td><td align="left"><p>69.81%</p></td><td align="left"><p>65.09%</p></td><td align="left"><p>58.49%</p></td><td align="left"><p>53.77%</p></td></tr><tr><td align="left" colspan="3"><p> Family income (AUD)</p></td><td align="left" /><td align="left" /><td align="left" /></tr><tr><td align="left"><p> 0–$37 k</p></td><td align="left"><p>8.91%</p></td><td align="left"><p>7.21%</p></td><td align="left"><p>8.91%</p></td><td align="left"><p>7.53%</p></td><td align="left"><p>5.26%</p></td></tr><tr><td align="left"><p> $37,001–$80 k</p></td><td align="left"><p>17.12%</p></td><td align="left"><p>18.92%</p></td><td align="left"><p>13.86%</p></td><td align="left"><p>9.68%</p></td><td align="left"><p>10.53%</p></td></tr><tr><td align="left"><p> $80,001–$180 k</p></td><td align="left"><p>50.45%</p></td><td align="left"><p>45.05%</p></td><td align="left"><p>45.54%</p></td><td align="left"><p>40.86%</p></td><td align="left"><p>42.11%</p></td></tr><tr><td align="left"><p> $180,001 k + </p></td><td align="left"><p>13.51%</p></td><td align="left"><p>16.22%</p></td><td align="left"><p>18.81%</p></td><td align="left"><p>23.66%</p></td><td align="left"><p>26.32%</p></td></tr><tr><td align="left"><p>Teacher</p></td><td align="left" /><td align="left" /><td align="left" /><td align="left" /><td align="left" /></tr><tr><td align="left"><p> Gender: female</p></td><td align="left"><p>93.18%</p></td><td align="left"><p>91.49%</p></td><td align="left"><p>89.71%</p></td><td align="left"><p>84.85%</p></td><td align="left"><p>84.31%</p></td></tr><tr><td align="left" colspan="2"><p> Years teaching experience</p></td><td align="left" /><td align="left" /><td align="left" /><td align="left" /></tr><tr><td align="left"><p> ≤ 5</p></td><td align="left"><p>32.50%</p></td><td align="left"><p>21.74%</p></td><td align="left"><p>32.31%</p></td><td align="left"><p>30.16%</p></td><td align="left"><p>18.37%</p></td></tr><tr><td align="left"><p> 6–10</p></td><td align="left"><p>27.50%</p></td><td align="left"><p>34.78%</p></td><td align="left"><p>18.46%</p></td><td align="left"><p>25.40%</p></td><td align="left"><p>26.53%</p></td></tr><tr><td align="left"><p> 11–15</p></td><td align="left"><p>12.50%</p></td><td align="left"><p>15.22%</p></td><td align="left"><p>15.38%</p></td><td align="left"><p>17.46%</p></td><td align="left"><p>20.41%</p></td></tr><tr><td align="left"><p> ≥ 16 + </p></td><td align="left"><p>27.50%</p></td><td align="left"><p>28.26%</p></td><td align="left"><p>33.85%</p></td><td align="left"><p>26.98%</p></td><td align="left"><p>34.69%</p></td></tr><tr><td align="left" colspan="4"><p> Years teaching autistic children/children with disabilities</p></td><td align="left" /><td align="left" /></tr><tr><td align="left"><p> ≤ 2</p></td><td align="left"><p>22.22%</p></td><td align="left"><p>20.93%</p></td><td align="left"><p>19.35%</p></td><td align="left"><p>16.67%</p></td><td align="left"><p>19.05%</p></td></tr><tr><td align="left"><p> 3–5</p></td><td align="left"><p>27.78%</p></td><td align="left"><p>20.93%</p></td><td align="left"><p>29.03%</p></td><td align="left"><p>24.07%</p></td><td align="left"><p>14.29%</p></td></tr><tr><td align="left"><p> 6–10</p></td><td align="left"><p>25.00%</p></td><td align="left"><p>32.56%</p></td><td align="left"><p>19.35%</p></td><td align="left"><p>29.63%</p></td><td align="left"><p>33.33%</p></td></tr><tr><td align="left"><p> ≥ 11</p></td><td align="left"><p>25.00%</p></td><td align="left"><p>25.58%</p></td><td align="left"><p>32.26%</p></td><td align="left"><p>29.63%</p></td><td align="left"><p>33.33</p></td></tr><tr><td align="left" colspan="4"><p> Duration known student/student in teacher's classroom</p></td><td align="left" colspan="2" /></tr><tr><td align="left"><p> 3–6 months</p></td><td align="left"><p>18.60%</p></td><td align="left"><p>22.72%</p></td><td align="left"><p>32.07%</p></td><td align="left"><p>18.87%</p></td><td align="left"><p>0.00%</p></td></tr><tr><td align="left"><p> 7–12 months</p></td><td align="left"><p>74.42%</p></td><td align="left"><p>52.27%</p></td><td align="left"><p>34.91%</p></td><td align="left"><p>31.13%</p></td><td align="left"><p>67.31%</p></td></tr><tr><td align="left"><p> 13–24 months</p></td><td align="left"><p>6.98%</p></td><td align="left"><p>18.18%</p></td><td align="left"><p>11.32%</p></td><td align="left"><p>6.60%</p></td><td align="left"><p>17.31%</p></td></tr><tr><td align="left"><p> 25 months + </p></td><td align="left"><p>0.00%</p></td><td align="left"><p>6.82%</p></td><td align="left"><p>12.26%</p></td><td align="left"><p>6.60%</p></td><td align="left"><p>15.38%</p></td></tr></tbody></table> </ephtml> </p> <p>N with complete ACES data varies across each year. Imputation was used to bring the full sample up to 106</p> <p>Table 2 Child, teacher and class demographics (older cohort)</p> <p> <ephtml> <table frame="hsides" rules="groups"><thead><tr><th align="left" /><th align="left"><p>Time 1</p></th><th align="left"><p>Time 2</p></th><th align="left"><p>Time 3</p></th><th align="left"><p>Time 4</p></th><th align="left"><p>Time 5</p></th></tr></thead><tbody><tr><td align="left"><p>Child</p></td><td align="left" /><td align="left" /><td align="left" /><td align="left" /><td align="left" /></tr><tr><td align="left"><p> Gender: male</p></td><td align="left"><p>72%</p></td><td align="left"><p>83%</p></td><td align="left"><p>75%</p></td><td align="left"><p>73%</p></td><td align="left"><p>75%</p></td></tr><tr><td align="left"><p> Age in months mean (SD, range)</p></td><td align="left"><p>132.5 (7.73, 119–149)</p></td><td align="left"><p>144.12 (7.57, 130–160)</p></td><td align="left"><p>156.52 (13.08, 143–170)</p></td><td align="left"><p>168.85 (7.51, 153–183)</p></td><td align="left"><p>181.81 (7.78, 168–197)</p></td></tr><tr><td align="left"><p> Mainstream school</p></td><td align="left"><p>83.51%</p></td><td align="left"><p>78.35%</p></td><td align="left"><p>77.32%</p></td><td align="left"><p>74.23%</p></td><td align="left"><p>65.98%</p></td></tr><tr><td align="left" colspan="2"><p> Family income (AUD)</p></td><td align="left" /><td align="left" /><td align="left" /><td align="left" /></tr><tr><td align="left"><p> 0–$37 k</p></td><td align="left"><p>14.95%</p></td><td align="left"><p>12.26%</p></td><td align="left"><p>10.58%</p></td><td align="left"><p>11.34%</p></td><td align="left"><p>8.51%</p></td></tr><tr><td align="left"><p> $37–$80 k</p></td><td align="left"><p>16.82%</p></td><td align="left"><p>8.87%</p></td><td align="left"><p>17.31%</p></td><td align="left"><p>16.49%</p></td><td align="left"><p>17.02%</p></td></tr><tr><td align="left"><p> $80–$180 k</p></td><td align="left"><p>52.34%</p></td><td align="left"><p>45.28%</p></td><td align="left"><p>46.15%</p></td><td align="left"><p>42.27%</p></td><td align="left"><p>40.43%</p></td></tr><tr><td align="left"><p> $180 k + </p></td><td align="left"><p>8.41%</p></td><td align="left"><p>9.43%</p></td><td align="left"><p>12.50%</p></td><td align="left"><p>18.56%</p></td><td align="left"><p>22.34%</p></td></tr><tr><td align="left"><p>Teacher</p></td><td align="left" /><td align="left" /><td align="left" /><td align="left" /><td align="left" /></tr><tr><td align="left"><p> Gender: female</p></td><td align="left"><p>85.18%</p></td><td align="left"><p>82.5%</p></td><td align="left"><p>89.28%</p></td><td align="left"><p>76.19%</p></td><td align="left"><p>76.47%</p></td></tr><tr><td align="left" colspan="3"><p> Years teaching experience</p></td><td align="left" /><td align="left" /><td align="left" /></tr><tr><td align="left"><p> ≤ 5</p></td><td align="left"><p>23.08%</p></td><td align="left"><p>26.32%</p></td><td align="left"><p>16.36%</p></td><td align="left"><p>29.51%</p></td><td align="left"><p>20.41%</p></td></tr><tr><td align="left"><p> 6–10</p></td><td align="left"><p>30.77%</p></td><td align="left"><p>26.32%</p></td><td align="left"><p>30.91%</p></td><td align="left"><p>21.31%</p></td><td align="left"><p>32.65%</p></td></tr><tr><td align="left"><p> 11–15</p></td><td align="left"><p>15.38%</p></td><td align="left"><p>13.16%</p></td><td align="left"><p>20.00%</p></td><td align="left"><p>13.11%</p></td><td align="left"><p>6.12%</p></td></tr><tr><td align="left"><p> ≥ 16 + </p></td><td align="left"><p>30.77%</p></td><td align="left"><p>34.21%</p></td><td align="left"><p>34.55%</p></td><td align="left"><p>36.07%</p></td><td align="left"><p>40.82%</p></td></tr><tr><td align="left" colspan="4"><p> Years teaching autistic children/children with disabilities</p></td><td align="left" /><td align="left" /></tr><tr><td align="left"><p> ≤ 2</p></td><td align="left"><p>25.53%</p></td><td align="left"><p>18.18%</p></td><td align="left"><p>16.32%</p></td><td align="left"><p>25.86%</p></td><td align="left"><p>16.67%</p></td></tr><tr><td align="left"><p> 3–5</p></td><td align="left"><p>21.28%</p></td><td align="left"><p>18.18%</p></td><td align="left"><p>18.36%</p></td><td align="left"><p>17.24%</p></td><td align="left"><p>20.83%</p></td></tr><tr><td align="left"><p> 6–10</p></td><td align="left"><p>29.79%</p></td><td align="left"><p>36.36%</p></td><td align="left"><p>29.59%</p></td><td align="left"><p>22.41%</p></td><td align="left"><p>22.92%</p></td></tr><tr><td align="left"><p> ≥ 11</p></td><td align="left"><p>23.40%</p></td><td align="left"><p>27.27%</p></td><td align="left"><p>35.71%</p></td><td align="left"><p>34.48%</p></td><td align="left"><p>39.58%</p></td></tr><tr><td align="left" colspan="4"><p> Duration known student/student in teacher's classroom</p></td><td align="left" /><td align="left" /></tr><tr><td align="left"><p> 3–6 months</p></td><td align="left"><p>25.77%</p></td><td align="left"><p>31.96%</p></td><td align="left"><p>25.51%</p></td><td align="left"><p>25.00%</p></td><td align="left"><p>0.00%</p></td></tr><tr><td align="left"><p> 7–12 months</p></td><td align="left"><p>44.33%</p></td><td align="left"><p>42.27%</p></td><td align="left"><p>43.88%</p></td><td align="left"><p>25.00%</p></td><td align="left"><p>47.95%</p></td></tr><tr><td align="left"><p> 13–24 months</p></td><td align="left"><p>18.56%</p></td><td align="left"><p>17.53%</p></td><td align="left"><p>19.39%</p></td><td align="left"><p>12.50%</p></td><td align="left"><p>22.45%</p></td></tr><tr><td align="left"><p> 25 months + </p></td><td align="left"><p>11.34%</p></td><td align="left"><p>8.24%</p></td><td align="left"><p>11.22%</p></td><td align="left"><p>18.75%</p></td><td align="left"><p>29.59%</p></td></tr></tbody></table> </ephtml> </p> <p>N with complete ACES data varies across each year. Imputation was used to bring the full sample up to 92</p> <hd id="AN0182844507-9">Measures</hd> <p></p> <hd id="AN0182844507-10">Demographic Characteristics</hd> <p>Child demographics (age, gender, age at diagnosis) were collected through the annual parent questionnaire. The teacher questionnaire began with a demographics section for the teachers. They were asked about their teaching experience (e.g., primary degree, length of time teaching), how many months they had known the child, and their school setting (e.g., mainstream, special education).</p> <hd id="AN0182844507-11">Academic Skills and Enablers</hd> <p>The Academic Competence Evaluation Scales—Teacher Form (ACES-TF; DiPerna & Elliott, [<reflink idref="bib9" id="ref46">9</reflink>]) is a teacher-rated assessment tool which measures teacher perceptions of academic skills and academic enablers in students in Grades K–12. The scale is completed based on direct observation of skills and enablers in the classroom setting by respondents who have known the student in their classroom for a minimum of 6 weeks. The first 33 items measure three areas of academic skills: Reading/Language, Mathematics and Critical Thinking. Reading/Language includes items relating to reading comprehension, written communication, and oral language. Mathematics includes items relating to measurement, problem solving, and arithmetic. Critical Thinking includes items relating to solving problems, synthesising information, and generalising skills. The "all grades" measure of Critical Thinking is used across this study (i.e., it does not change after Grade 3 to include the Grade 3–12-only items).</p> <p>For academic skills, the teacher rates the student's performance on each skill (e.g., "uses numbers to solve daily problems") compared to grade level: "far below grade level", "below grade level", "at grade level", "above grade level" or "far above grade level". The remaining items measure academic enablers, consisting of three subscales for the younger cohort and four subscales for the older cohort. Interpersonal skills includes items related to a student's ability to work, learn, and interact with others (e.g., "works effectively in a small group activity"). Engagement includes items relating to class participation and activity initiation (e.g., "speaks in class when called upon"). The Motivation subscale contains items relating to seeking out academic challenge and improvement (e.g., "attempts to improve on previous performance"). The items in the Study Skills subscale differ dependent upon grade, so these questions were only asked of the older cohort. Study skills questions ask about completion of homework and preparation for tasks or tests. For the academic enabler items, teachers are asked to rate how frequently the student demonstrates that behaviour: "never", "seldom", "sometimes", "often", or "almost always".</p> <p>The ACES demonstrate strong psychometric properties in neurotypical samples. Test reliabilities were 0.95 and 0.96 for academic skills and academic enablers, respectively, and interrater agreement was 0.99 and 0.61, respectively. Mean coefficient alphas for ACES-TF are 0.99 (DiPerna & Elliott, [<reflink idref="bib9" id="ref47">9</reflink>]). Keen et al. ([<reflink idref="bib17" id="ref48">17</reflink>]) reported Cronbach's alpha of 1.0 for Reading/Language, 0.97 for Mathematics, 0.97 for Critical Thinking, 0.94 for Interpersonal Skills, 0.93 for Engagement, 0.95 for Motivation, and 0.74 for Study Skills for their sample of autistic children. For the younger and older cohorts in this study, the Cronbach's alpha at the initial assessment was 0.97 for Reading/Language, 0.99/0.98 for Mathematics, 0.99/0.96 for Critical Thinking, 0.96/0.94 for Interpersonal Skills, 0.92/0.94 for Engagement, 0.96/0.93 for Motivation, and 0.91 for Study Skills (older cohort only).</p> <hd id="AN0182844507-12">Child Internalising and Externalising Behaviour</hd> <p>The Strengths and Difficulties Questionnaire teacher version (SDQ; Goodman, [<reflink idref="bib13" id="ref49">13</reflink>]) is a brief, 25-item screening questionnaire. It is widely used in research, with 48 published studies exploring its reliability and validity across 131,223 children (Stone et al., [<reflink idref="bib37" id="ref50">37</reflink>]). Based upon their review of these studies, Stone et al. ([<reflink idref="bib37" id="ref51">37</reflink>]) concluded that the psychometric properties of the SDQ are strong, especially for the teacher version. This study reports on the Internalising Behaviour, Externalising Behaviour, and Total Behaviour Problems subscale (TBPS) scores at Time 1. Cronbach's alpha for this sample at this time point was acceptable for both cohorts on the Externalising subscale (α = 0.70), the Internalising subscale (α = 0.74 younger cohort; α = 0.69 older cohort), and the TBPS (α = 0.77 younger cohort; α = 0.73 older cohort).</p> <hd id="AN0182844507-13">Autism Characteristics</hd> <p>The Social Communication Questionnaire (SCQ) (Rutter et al., [<reflink idref="bib35" id="ref52">35</reflink>]) is a behavioural checklist that requires parents to indicate the presence of certain social, communicative or stereotyped behaviours by answering yes or no to 40 items. A higher score represents a higher number of behaviours which may be considered characteristic of autism.</p> <hd id="AN0182844507-14">Data Analysis</hd> <p>Data analysis proceeded in three phases: profile identification through Latent Profile Analysis, comparison of child and family factors between profiles, and then examination of each profile's trajectory and comparison of outcomes across time. These were undertaken separately for the younger and the older cohorts.</p> <p></p> <ulist> <item> Latent profile analysis</item> <p></p> <item> For the current study, it was important to use an assumption-free method to develop the profiles, thereby letting the groups emerge from the observed data. Therefore, latent profile analysis was used to examine how children clustered along the three variables of performance (Reading/Language, Mathematics, and Critical Thinking) and the four variables of academic enablers (Interpersonal Skills, Engagement, Motivation and, for the older cohort only, Study Skills) at Time 1.</item> <p></p> <item> The number of profiles for each cohort was selected by using an analytic hierarchy process (see Akogul & Erisoglu, [<reflink idref="bib1" id="ref53">1</reflink>]) using the compare solutions function of the tidyLPA package (Rosenburg et al., [<reflink idref="bib33" id="ref54">33</reflink>]) in R. This process compares the fits of different numbers of profiles across a range of fit criteria (e.g., AIC, BIC). To ensure that data were included from all children (when they had missing data for a year, for example), imputation was used to replace the missing data (as is standard practice). To ensure that this was not biased, this process was run 100 times for each cohort and the distribution of profiles plotted to examine the best profile fit number.</item> <p></p> <item> Comparison of factors between profiles</item> <p></p> <item> To examine whether child or family factors varied across profiles, a series of ANOVAs (when outcome variables were continuous) and Chi-squared tests (when outcome variables were categorical) were used. Factors compared were child sex, autism characteristics (SCQ total score), school type (mainstream/special), household income (Likert scale), and child behaviour (SDQ Internalising, Externalising, and TBPS subscales). For the ANOVA analyses, effect sizes are reported in terms of partial eta-squared (η<sups>2</sups>) with the conventions of effect size of η<sups>2</sups> = 0.01 = small, η<sups>2</sups> = 0.06 = medium, and η<sups>2</sups> = 0.14 = large (Cohen, [<reflink idref="bib7" id="ref55">7</reflink>]).</item> <p></p> <item> Profile-Based Longitudinal Trajectories of Academic Performance</item> <p></p> <item> Mixed-effect modelling was used to examine whether performance on reading/language, mathematics and critical thinking varied across profile and across year of study, and whether there was an interaction between them. Mixed-effect models have gained popularity in recent years due to their ability to simultaneously model variability across factors of interest (fixed effects) and factors that may contribute to the variability but are not of interest (random effects). Mixed-effect models allow the model fit to vary across the random effects. This means that by including participants as a random effect, the model can account for individual variability and participants can be included in the model even when they do not have data in all cells of the design. Mixed-effect models were fit using the lmerTest package in R (Kuznetsova et al., [<reflink idref="bib19" id="ref56">19</reflink>]). Estimated marginal means were extracted, and when effects were significant, contrasts were run using the emmeans package (Lenth et al., [<reflink idref="bib21" id="ref57">21</reflink>]), with visualisation of effects being produced with ggplot2 (Wickham, [<reflink idref="bib41" id="ref58">41</reflink>]) with the viridis package for accessible plot colours (Garnier, [<reflink idref="bib11" id="ref59">11</reflink>]).</item> </ulist> <hd id="AN0182844507-15">Results</hd> <p></p> <hd id="AN0182844507-16">Younger Cohort: Profiles of Academic Skills and Enablers</hd> <p>To address Research Question 1 and to identify profiles of academic skills (Reading/Language, Mathematics, and Critical Thinking) and academic enablers (Interpersonal Skills, Engagement, and Motivation) in 5–6 year-old autistic students, latent profile analysis was conducted on the subscales of the ACES at Time 1. Across 100 random imputations of the data, the latent profiling analysis centred around seven profiles for the younger cohort (see supplementary material). The seven profiles identified by the latent profile analysis can be seen in Fig. 1. The <emph>z</emph>-scores on the ACES subscales (three academic skills, three academic enabler subscales) for the seven profiles are shown in Fig. 1. Profile 1 is characterised by 8.5% of the sample; Profile 2, 17.9%; Profile 3, 20.8%; Profile 4, 15.1%; Profile 5, 9.4%; Profile 6, 14.2%; and Profile 7, 14.2%. Based on Fig. 1, the profiles were named to describe the profile across the academic skills and academic enablers within this cohort (i.e., of young autistic children): (<reflink idref="bib1" id="ref60">1</reflink>) Low consistent; (<reflink idref="bib2" id="ref61">2</reflink>) Medium Low consistent; (<reflink idref="bib3" id="ref62">3</reflink>) Average academics, poorer enablers; (<reflink idref="bib4" id="ref63">4</reflink>) Medium High consistent; (<reflink idref="bib5" id="ref64">5</reflink>) High consistent; (<reflink idref="bib6" id="ref65">6</reflink>) Low academics with strength in enablers and (<reflink idref="bib7" id="ref66">7</reflink>) Medium High academics with poorer enablers.</p> <p>Graph: Fig. 1 Variable estimates (scaled) of the ACES subscales across the seven profiles in the younger cohort, along with the number of children within each profile. As scores are scaled (z-scores), 0 represents "average" performance within the sample. Values above 0 represent "above average" and values below 0 represent "below average" performance within this cohort. L language, M mathematics, CT critical thinking, IS interpersonal skills, En engagement, Mo motivation</p> <hd id="AN0182844507-17">Younger Cohort: Comparison of Factors Between Profiles</hd> <p>In order to identify factors that differ between academic profiles, a series of Chi-square tests and ANOVAs were undertaken on child gender, SCQ score, school type (mainstream/special), household income (Likert scale) and SDQ Internalising, Externalising, and TBPS subscales. The results are presented in the stacked bar charts in Fig. 2, the results of the statistical comparisons are in Supplementary Table S5. There were no significant differences between profiles in gender, SCQ score, or household income. There was a significant difference in school type (mainstream/special), with the largest proportion of students in special schools in the Low consistent (Profile 1) and Medium Low consistent (Profile 2) profiles.</p> <p>Graph: Fig. 2 Time 1 younger cohort factors by profiles</p> <p>There were significant effects of profile on SDQ Internalising (<emph>p</emph> = 0.036, η<sups>2</sups> = 0.30, large effect size), SDQ Externalising (<emph>p</emph> < 0.001, η<sups>2</sups> = 0.46, large effect size), and SDQ TBPS (<emph>p</emph> < 0.001, η<sups>2</sups> = 0.50, large effect size). Tukey post-hoc tests revealed that there was a significantly lower score for internalising in the Medium High consistent (Profile 4) compared to the Low consistent profile (Profile 1; <emph>p</emph> = 0.023), but there were no other differences. For SDQ Externalising, scores in Medium High consistent (Profile 4) were significantly lower than Low consistent (Profile 1; <emph>p</emph> = 0.003), Medium Low consistent (Profile 2; <emph>p</emph> = 0.005), and Average consistent, poorer enablers (Profile 3; <emph>p</emph> = 0.037). The High consistent (Profile 5) was significantly lower than the Low consistent (Profile 1; <emph>p</emph> = 0.035).</p> <p>In terms of SDQ TBPS, scores were higher in Low consistent (Profile 1) than in the Medium High consistent (Profile 4; <emph>p</emph> < 0.001), High consistent (Profile 5; <emph>p</emph> = 0.014), and Medium High academics with poorer enablers. (Profile 7; <emph>p</emph> = 0.015). The Medium Low consistent (Profile 2) TBPS score was also higher than the Medium High consistent (Profile 4; <emph>p</emph> = 004). The profiles with the highest SDQ TBPS score were the Low consistent ( <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mover><mi>x</mi><mo>¯</mo></mover></math> </ephtml> = 20.4), Medium Low consistent ( <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mover><mi>x</mi><mo>¯</mo></mover></math> </ephtml> = 16.5), and Average ( <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mover><mi>x</mi><mo>¯</mo></mover></math> </ephtml> = 16.5). The profiles with the lowest SDQ TBPS were Profile 4 (Medium High consistent, <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mover><mi>x</mi><mo>¯</mo></mover></math> </ephtml> = 6.83) and Profile 5 (High consistent, <ephtml> <math xmlns="http://www.w3.org/1998/Math/MathML"><mover><mi>x</mi><mo>¯</mo></mover></math> </ephtml> = 6.83). As can be seen in Fig. 2, the SDQ TBPS for the two mixed profiles (Profile 6; Low academics with strength in enablers and Profile 7; Medium High academics with poorer enablers) appear relatively similar but may be made up of differing proportions of Internalising and Externalising scores.</p> <hd id="AN0182844507-18">Younger Cohort: Trajectory of Academic Skills Over Time</hd> <p>The trajectory of academic skills over the course of the study was then explored separately for each area: Reading/Language, Mathematics and Critical Thinking. Each of these are described in turn below.</p> <hd id="AN0182844507-19">Reading/Language Across Time</hd> <p>A mixed-effect model was used to examine the effect of Profile and Year on scores for Reading/Language, with subject included as a random effect. Estimated marginal means from the modelling can be seen in the lower panel of Fig. 3. The model revealed that there was a significant effect of Profile (<emph>F</emph> (<reflink idref="bib6" id="ref67">6</reflink>, 93.83) = 8.72, <emph>p</emph> < 0.001), but no effect of Year (<emph>p</emph> = 0.389) and no interaction (<emph>p</emph> = 0.807). This suggests that each profile developed on the Time 1 data has a significantly different trajectory of reading/language skill development (based on teacher ratings compared to peers) over time. Table S2 (Supplementary material) summarises the post-hoc comparisons with Tukey corrected <emph>p</emph>-values. This shows that across the study, Profiles 1 (Low consistent), 2 (Medium Low consistent), 3 (Average academics, poorer enablers) and 6 (Low academics with strength in enablers) were significantly lower for Reading/Language than Profile 7 (Medium High academics with poorer enablers). Profile 2 (Medium Low consistent) was lower than Profiles 4 (Medium High consistent) and 5 (High consistent), while Profiles 4 and 5 were higher than Profile 6 (Low academics with strength in enablers).</p> <p>Graph: Fig. 3 Estimated marginal means of the z-scores in reading/language, mathematics, and critical thinking across Time 1–5 for each of the seven profiles identified at Time 1 in the younger cohort</p> <hd id="AN0182844507-20">Mathematics Across Time</hd> <p>The estimated marginal means from the modelling of the ACES Mathematics scores for each profile are shown in the middle panel of Fig. 3. Similar to the Reading/Language subscale, a mixed-effect model showed that there was a significant main effect of Profile membership on Mathematics scores (<emph>F</emph> (<reflink idref="bib6" id="ref68">6</reflink>, 91.1) = 5.34, <emph>p</emph> < 0.001), but no effect of Year (<emph>p</emph> = 0.277) and no interaction (<emph>p</emph> = 0.721). Table S3 (Supplementary material) summarises the post-hoc comparisons with Tukey corrected <emph>p</emph>-values for the main effect of Profile. This shows that across time for the Mathematics subscale, Low consistent (Profile 1) was lower than High consistent (Profile 5) and Medium High academics with poorer enablers (Profile 7). Medium High consistent (Profile 4) and High consistent (Profile 5) were higher than Low academics with strength in enablers (Profile 6), who were lower than Medium High academics with poorer enablers (Profile 7).</p> <hd id="AN0182844507-21">Critical Thinking Across Time</hd> <p>The estimated marginal means from the modelling of the ACES Critical Thinking are shown in the upper panel of Fig. 3. Similar to the other two ACES subscales, a mixed-effect model on Critical Thinking scores revealed a main effect of Profile (<emph>F</emph> (<reflink idref="bib6" id="ref69">6</reflink>, 86.31) = 7.25, <emph>p</emph> < 0.001), but no effect of Year (<emph>p</emph> = 0.838) and no interaction (<emph>p</emph> = 0.202). As per the Reading/Language and Mathematics subscales, this means that each profile identified at Time 1 remains significantly different over time. Post-hoc comparisons with Tukey corrected <emph>p</emph>-values of the profiles (Table S4, Supplementary material) show that the Low Consistent (Profile 1) were lower than Profiles 4 (Medium High consistent), 5 (High consistent) and 7 (Medium High academics with poorer enablers). Profile 2 (Medium Low consistent) was lower than Profile 4 (Medium High consistent) and Profile 5 (High consistent). Profiles 2 (Medium Low consistent), 5 (High consistent) and 7 were all higher than Profile 6 (Low academics with strength in enablers).</p> <hd id="AN0182844507-22">Older Cohort: Profiles of Academic Skills and Enablers</hd> <p>Latent profile analysis was again used to address Research Question 1 and to identify profiles of teacher academic skills (Reading/Language, Mathematics, Critical Thinking) and academic enablers (Interpersonal Skills, Engagement, and Motivation) in older autistic students at Time 1 of the LASA study. Across 100 random imputations of the data, the number of profiles that was fit by the latent profiling analysis centred around six profiles for the older cohort (see Supplementary material). The six profiles identified by the latent profile analysis can be seen in Fig. 4.</p> <p>Graph: Fig. 4 Variable estimates (scaled) of the ACES subscales across the six profiles in the older cohort, along with the number of children within each profile. As scores are scaled (z-scores), 0 represents "average" performance within the sample. Values above 0 represent "above average" and values below 0 represent "below average" performance within this cohort. L language, M mathematics, CT critical thinking, IS interpersonal skills, En engagement, Mo motivation, SS study skills</p> <p>Profile 1 is characterised by 15.3% of the sample; Profile 2, 28.6%; Profile 3, 16.3%; Profile 4, 12.2%; Profile 5, 15.3%; and Profile 6, 12.2%. Based on Fig. 1, the profiles were named (<reflink idref="bib1" id="ref70">1</reflink>) Low consistent, (<reflink idref="bib2" id="ref71">2</reflink>) Average consistent, (<reflink idref="bib3" id="ref72">3</reflink>) Medium High academics with stronger enablers, (<reflink idref="bib4" id="ref73">4</reflink>) Low academics with relative strength in enablers, (<reflink idref="bib5" id="ref74">5</reflink>) Mixed academics with poorer enablers and (<reflink idref="bib6" id="ref75">6</reflink>) High academics with poorer enablers.</p> <hd id="AN0182844507-23">Older Cohort: Comparison of Factors Between Profiles</hd> <p>In order to identify factors that differ between the six profiles identified in the older cohort, a series of Chi-squares and ANOVAs were undertaken on child gender, SCQ score, school type (mainstream/special), household income, and SDQ scores. The results are presented in the stacked bar charts in Fig. 5 and the results of the statistical comparisons are in Supplementary Table S7. There were no significant differences in gender, SCQ score, school type, or household income.</p> <p>Graph: Fig. 5 Time 1 older cohort factors by profiles</p> <p>There was no significant effect of Profile on SDQ Internalising (<emph>p</emph> = 0.145, η<sups>2</sups> = 0.17, large effect size), but there was a significant effect on SDQ Externalising (<emph>p</emph> = 0.003, η<sups>2</sups> = 0.31, large effect) and on SDQ TBPS (<emph>p</emph> = 0.005, η<sups>2</sups> = 0.31, large effect size). Tukey post-hocs revealed that Externalising was significantly higher in the Low consistent profile (Profile 1) than in Medium High academics with stronger enablers (Profile 3; <emph>p</emph> = 0.001) and High academics with poorer enablers (Profile 6; <emph>p</emph> = 0.049). The SDQ Externalising score was also significantly higher in the Average consistent (Profile 2) than Medium High academics with stronger enablers (Profile 3; <emph>p</emph> = 0.045). For SDQ TBPS scores, Low consistent<emph> (</emph>Profile 1) was higher than Medium High academics with stronger enablers (Profile 3; <emph>p</emph> = 0.002) and High academics with poorer enablers (Profile 6; <emph>p</emph> = 0.002), but there were no other significant differences.</p> <hd id="AN0182844507-24">Older Cohort: Trajectory of Academic Skills Over Time</hd> <p>The trajectory of academic skills over the course of the study was then explored separately for each academic skill: Reading/Language, Mathematics and Critical Thinking. Each of these are described in turn below and presented in the upper, middle, and lower panels of Fig. 6.</p> <p>Graph: Fig. 6 Estimated marginal means of the z-scores in reading/language, mathematics, and critical thinking across Time 1–5 for each of the six profiles identified at Time 1 in the older cohort</p> <hd id="AN0182844507-25">Reading/Language Across Time</hd> <p>A mixed-effect model was used to examine the effect of Profile and Year on scores for Reading/Language, with participant included as a random effect. Estimated marginal means from the modelling can be seen in the lower panel of Fig. 6. The model revealed a significant effect of Profile (<emph>F</emph> (<reflink idref="bib5" id="ref76">5</reflink>, 70.29) = 4.11, <emph>p</emph> = 0.003), but no effect of Year (<emph>p</emph> = 0.923) and no interaction (<emph>p</emph> = 0.499). Table S6 (Supplementary material) documents post-hoc Profile comparisons with Tukey corrected <emph>p</emph>-values. Across the duration of the study for the Reading/Language subscale, Low consistent and Average consistent (Profiles 1 and 2) are significantly lower than Medium High academics with stronger enablers (Profile 3). Average consistent (Profile 2) is also significantly lower than High academics with poorer enablers (Profile 6).</p> <hd id="AN0182844507-26">Mathematics Across Time</hd> <p>As can be seen in the middle panel of Fig. 6, the mixed-effect model revealed an effect of Profile (<emph>F</emph> (<reflink idref="bib5" id="ref77">5</reflink>, 75.8) = 6.23, <emph>p</emph> < 0.001) but not of Year (<emph>p</emph> = 0.331), and no interaction (<emph>p</emph> = 0.255). Table S7 (Supplementary material) documents post-hoc Profile comparisons with Tukey corrected <emph>p</emph>-values. Low consistent, Average consistent, and Low academics with relative strength in enablers (Profiles 1, 2 and 4) are significantly lower than Medium High academics with stronger enablers (Profile 3). The Average Consistent profile (Profile 2) is also significantly below Mixed academics with poorer enablers (Profile 5) and High academics with poorer enablers (Profile 6).</p> <hd id="AN0182844507-27">Critical Thinking Across Time</hd> <p>The model of Critical Thinking across time in the older cohort is presented in the upper panel of Fig. 6. There was a significant effect of Profile (<emph>F</emph> (<reflink idref="bib5" id="ref78">5</reflink>, 64.03) = 4.49, <emph>p</emph> = 0.001) and also a significant effect of Year (<emph>F</emph> (<reflink idref="bib4" id="ref79">4</reflink>, 90.26) = 2.66, <emph>p</emph> = 0.034). Post-hoc analyses show that there was a significant increase in Critical Thinking skills from study Time 4 (age 12–13) to Time 5 (age 13–14) only (Tukey correct <emph>p</emph> = 0.040). There was no interaction. Post-hoc Profile comparisons with Tukey corrected <emph>p</emph>-values (Table S8, Supplementary material) show that for Critical Thinking across the study, the Low consistent, Average consistent and Low academics with relative strength in enablers (Profiles 1, 2 and 4) are significantly lower than Medium High academics with stronger enablers (Profile 3). Average consistent (Profile 2) is also significantly lower than High academics with poorer enablers (Profile 6).</p> <hd id="AN0182844507-28">Discussion</hd> <p>This is the first prospective longitudinal study of teacher-reported academic skills and academic enablers in autistic children. Using a teacher-report approach allowed for the documentation of skills observed in the classroom and for the inclusion of children with a range of levels of ability, including those who may not be able to complete valid standardised direct assessments. It is, therefore, an important addition to the literature given that a proportion of autistic students may experience intellectual (Maenner et al., [<reflink idref="bib23" id="ref80">23</reflink>]) or language challenges (Kwok et al., [<reflink idref="bib20" id="ref81">20</reflink>]) which can impact the validity of direct standardised assessments. The use of two cohorts allows for an accelerated longitudinal design, documenting trajectories across both primary and secondary school age groups. From these data, three major findings emerged. First, at a single time point, distinct profiles of academic skills and enablers were identified for the younger (seven profiles) and older (six profiles) cohorts. This highlights the large heterogeneity of academic profiles in autistic children and adolescents. It is important to look not only at the number of profiles but also at the profiles themselves. When doing this, it could be suggested that the older cohort had more "spiky" profiles (i.e., those with variability across skills) than the younger cohort. This may suggest that some variability across skills is present at entry into primary school, but that profiles may potentially become increasingly heterogeneous into secondary school. The second major finding is that child behaviour (particularly externalising behaviour) in both cohorts, and school placement in the younger cohort only, differed significantly between profiles. However, child autism characteristics, family income, and child gender did not differ between profiles in either cohort. Finally, the results demonstrate that profiles are associated with significantly different trajectories. This highlights the importance of early screening programs to identify those most likely to benefit from sustained support for their academic outcomes.</p> <hd id="AN0182844507-29">Profiles of Academic Skills and Enablers</hd> <p>Consistent with previous academic skills subtyping research (Bullen et al., [<reflink idref="bib3" id="ref82">3</reflink>]; Chen et al., [<reflink idref="bib5" id="ref83">5</reflink>]; Wei et al., [<reflink idref="bib39" id="ref84">39</reflink>]), multiple profiles were found within both age cohorts of this study. The latent profile analysis identified seven profiles in the younger cohort and six profiles within the older cohort, which is more than in other studies that have explored broader academic skillset profiles (e.g., Zaidman-Zait et al., [<reflink idref="bib42" id="ref85">42</reflink>]). Both groups had a Low consistent profile (i.e., low academics and enablers) but the proportions differed: 8.5% of the younger cohort and 15.3% of the older cohort. Both cohorts also had a profile which contained High academics with poorer enablers (14.2% of the younger cohort and 12.2% of the older cohort). However, the relative strengths and challenges within the other profiles were less similar between cohorts. Of interest is that for the younger cohort, there were four profiles which were described (based on visual inspection) as "consistent" across the academic and enabler skills, but only two for the older cohort.</p> <p>There are several possible explanations for the observed greater heterogeneity of skills and enablers within the profiles in the older cohort compared to the younger cohort. It may reflect a genuine increase in heterogeneity of skills within the cohort with age. Age has been documented to impact profiles of academic achievement (Whitby & Mancil, [<reflink idref="bib40" id="ref86">40</reflink>]). Differences in the development of complex cognitive and meta-cognitive skills, such as executive functioning, and differing levels of student motivation and engagement, become more apparent in adolescence than in childhood, thereby influencing the diversity of skills and challenges seen in adolescents compared to younger children (e.g., five distinct profiles of engagement in neurotypical students; van Rooij et al., [<reflink idref="bib38" id="ref87">38</reflink>]). These results may also reflect a change in the ecological environment in which skills are being assessed or in the knowledge of the person rating them. For many children, primary schools tend to be smaller, and children learn from a single teacher with a single cohort of peers. In contrast, secondary schools are larger, demanding greater academic and social independence from their students; also, students are taught by multiple teachers (Mandy et al., [<reflink idref="bib24" id="ref88">24</reflink>]). In addition, streaming by academic skills in secondary schools may impact academic, social, and psychological outcomes, with greater benefits observed for higher ability students/groups and potential challenges for those streamed to "lower" ability groups (e.g., see review by Johnston & Wildy, [<reflink idref="bib16" id="ref89">16</reflink>]). Such processes may impact academic skills and enablers in potentially differing ways, increasing heterogeneity among these. Finally, observed differences may represent cohort differences rather than age differences, reflecting existing differences between each group of participants at study entry. The younger cohort would (by definition) have received their diagnosis at or before 5 years of age, whereas the older cohort entered the study at 9–10 years, which may result in child participants who received their diagnosis across a broader age range. A systematic review (Loubersac et al., [<reflink idref="bib22" id="ref90">22</reflink>]) concluded that children with an intellectual disability or more significant social communication challenges are more likely to receive a diagnosis at an earlier age than those with higher levels of intellectual ability and/or social communication skills, which may then impact upon cohort profiles. Of course, the aforementioned possible explanations are not mutually exclusive, and it is likely that for many students, their profile reflects interactions among a number of these factors. Regardless of the cause, the diversity within the profiles in academic skills and academic enablers in both cohorts, and particularly in the older cohort, underlines the importance of not assuming that academic skills at a specific level will mean the student has academic enablers at the equivalent level, or vice versa. It also highlights the importance of student knowledge and accurate measurement of both academic skills and enablers to understand individual children's skills, presentation, strengths, and needs to inform and implement appropriate supports (Clark et al., [<reflink idref="bib6" id="ref91">6</reflink>]).</p> <hd id="AN0182844507-30">Factors Differing Between Profiles of Academic Skills and Enablers</hd> <p>Comparisons explored whether the academic skills/enablers profiles were associated with, or differed based upon, child (gender, school setting, autism traits, behavioural presentation) or family (income) characteristics. There were no significant differences across profiles in a number of variables that are more stable across time, including gender, autism traits and household income. The finding that level of autism characteristics (SCQ total score) did not differentiate groups was consistent with previous academic profile research by Solari et al. ([<reflink idref="bib36" id="ref92">36</reflink>]) and Chen et al. ([<reflink idref="bib5" id="ref93">5</reflink>]). This is in contrast to Johnels et al. ([<reflink idref="bib14" id="ref94">14</reflink>]) who, using the autism spectrum screening questionnaire (ASSQ), found autism traits did differ between two of their three groups. However, they also found that the same groups differed on scores of ability. This may be a confounding factor, as studies entering both autism characteristics and ability into regression find that autism characteristics do not explain a significant amount of unique variance once ability is entered into the regression (e.g., Miller et al., [<reflink idref="bib30" id="ref95">30</reflink>]).</p> <p>Potentially modifiable variables of school type and child behaviour significantly differed between profiles. More specifically, the proportion of children within mainstream schools was significantly different between profiles within the younger cohort only. In the younger cohort, the Low consistent and Medium Low consistent profiles had the lowest proportion of students in mainstream schools. While nonsignificant, a similar pattern was observed for the older cohort. The finding that school type may differ between profiles, at least within younger children, is consistent with findings by Bullen et al. ([<reflink idref="bib3" id="ref96">3</reflink>]) who found that children in the lower ability group for mathematics and literacy were more likely to be in specialist settings at initial data collection. Whether this reflects existing differences that informed school placement, or whether school placement impacted development or skills, or both, cannot be determined from our data. Nevertheless, our findings highlight the importance of understanding the potentially reciprocal effects of profiles, school placements, levels of support, and skills development.</p> <p>Child behaviour was a second variable that can change with supports/environment that showed significant differences between profiles in both cohorts. The Internalising subscale showed differences between profiles only for the younger cohort. There were significant differences between profiles for both the Externalising and TBPS subscales across both cohorts. As TBPS is a composite of Internalising and Externalising scores, differences may have been driven by the consistent and large effect of externalising behaviours in both cohorts. Consistent with previous studies exploring broad academic skillset profiles (e.g., Zaidman-Zait et al., [<reflink idref="bib42" id="ref97">42</reflink>]) and those exploring teacher ratings of academic competence in young autistic children (Milgramm et al., [<reflink idref="bib29" id="ref98">29</reflink>]), profiles characterised by lower abilities/more challenges tended to exhibit higher levels of externalising behaviour. Higher levels of externalising behaviour indicate more frequent, or higher intensity of, behaviours. These behaviours may be higher in these groups because increased levels of challenging or externalising behaviours are associated with lower ability, language skills and other cognitive domains (Carter Leno et al., [<reflink idref="bib4" id="ref99">4</reflink>]). Alternatively, the externalising behaviour may be the external presentation of an internal experience (e.g., pain, depression) which may also be impacting academic achievement. The causal pathway between academic underachievement and behaviour is as yet unclear and further research into the direction of the relationship is needed. Nevertheless, evaluation of externalising and total behaviours, alongside academic outcomes and enablers, may be of benefit in identifying students at risk or in understanding strengths/weaknesses. This is because it may be that suporting one aspect (e.g., behaviours) may support the other (e.g., academic outcomes), or vice versa.</p> <hd id="AN0182844507-31">Trajectories of Academic Skills Over Time</hd> <p>Overall, in both younger and older cohorts, there were main effects of Profile, but not of Year (with one exception being T4 to T5 of Critical Thinking in the older cohort) and no Profile × Year interactions. This suggests that there were different performance levels between the profiles, but with relative stability. These findings are in need of replication but highlight the potential utility of identifying profile membership early within a schooling journey to help to identify children early who may benefit from more individualised support.</p> <p>Exploring the trajectory of each profile on reading/language, mathematics, and critical thinking separately over the 5 years suggests that, for some profiles, there may be potentially differing development of reading/language, mathematics, and critical thinking skills. For example, in the younger cohort, Profile 3 (Average academics, poorer enablers) has visually different trajectories across the Reading/Language, Mathematics and Critical Thinking subscales over time, as does Profile 7 (Medium High academics, poorer enablers). In the older cohort, Profile 6's (High academics with poorer enablers) Critical Thinking trajectory does not (on visual inspection) appear to follow the same trajectory as the Reading/Language or Mathematics trajectories.</p> <p>Visually comparing profiles which, at T1, have similar academics but differing enablers, highlights the importance of considering broader academic skillset profiles in planning support, both concurrent and future. For example, in the younger cohort, Fig. 1 shows that Profile 4 (High consistent) and Profile 7 (Medium High consistent with poorer enablers) begin with academic skills within a similar range. However, the trajectories over time differ in Mathematics, with Profile 4 showing relatively consistent performance but Profile 7 (which had poorer enablers) showing a downward trend. A similar difference between profiles with relatively similar academic skills but differing enablers may also be present in the older cohort for Language (Profiles 3 and 5). However, larger longitudinal samples are needed to statistically explore these potential differences.</p> <hd id="AN0182844507-32">Limitations and Future Directions</hd> <p>This investigation of profiles of academic skills and enablers as rated by teachers within two cohorts of autistic children makes a novel contribution to the literature in identifying empirical profiles and factors differentiating profiles, and in investigating the trajectories of academic skills by profile over time. However, it is important to acknowledge the study's limitations. This research relied on teacher report to enable the inclusion of a broader range of students who may not be able to validly complete standardised measures (or, at least, not the same measures across participants). Whilst teacher-report data provide rich data, there is a risk for potential rater bias and a need for further information from additional informants to identify factors which may differ between the profiles. Complementing the teacher-reported data with direct assessment or student-led data would enhance our knowledge in this area. This could then allow for comparison of direct assessment (e.g., standardised academic achievement assessment) and teacher report. Such comparisons could be of value to explore perception versus actual ability which may be impacted, for example, by teacher attitudes or classroom/environmental barriers to displaying abilities. Standardised assessment would also enable investigation of profiles within each area, such as comparing word reading to reading comprehension, arithmetic, and word problems, which may relate to other factors such as language ability.</p> <p>The focus of this study was to identify profiles within the sample of autistic students. Future work could explore whether similar diversity across profiles exists in other groups and/or in non-autistic individuals, including gender diversity. This would be of value in informing understanding of all students and the unique strengths and challenges in the autism population. Finally, while our initial cohort size was relatively large, given the number of profiles identified (13 in total across cohorts) the size of samples within each profile was relatively small, which may have limited power to detect interactions and smaller effects. Replication with larger cohorts, such as through data pooling across sites and collaborative research, would enable more fine-grained evaluation in future research harnessing the potential of questionnaire measures for feasibility of larger scale studies. Larger samples would also reduce the need for data imputation, which has inherent limitations.</p> <hd id="AN0182844507-33">Implications/Conclusion</hd> <p>Our findings show that distinct profiles of academic skills and enablers are present across autistic students. These profiles show stability over time and are associated with different patterns of student behaviour. Further research in this area, replicating findings with larger samples, is needed. Nevertheless, a number of implications for both research and practice may be posited from these findings. First, there is a need to "assess not assume" academic skills and enablers. For example, our findings of mixed ability groups highlight the risk of assuming academic ability from academic enablers, such as the concept of "school readiness", which may lead to exclusion and marginalisation of students based on relatively distinct dimensions. The need for comprehensive assessment to understand strengths and needs is highlighted. Second, there is a need for further research into why autistic students are at particular risk of academic challenges, given that autism traits did not predict group membership. Third, there is a need to consider the impact of school placement and behaviour on student academic skills and enablers. This could include more fine-grained analysis of the mechanisms driving these significant differences and whether changes in one (e.g., school placement, child behaviour) would impact the other (e.g., academic skills and/or enablers). Finally, our findings highlight the need for further research in this area to inform a more fine-grained understanding of the factors related to academic skills, given divergence between these. Such understanding could then inform targeted assessment to identify students at risk, to inform interventions (e.g., targeting school placement fit, behaviour support) and to monitor outcomes over time.</p> <hd id="AN0182844507-34">Acknowledgements</hd> <p>We are grateful to the autistic children, their families and their teachers for giving their time to support this research study. The authors also acknowledge the contributions of the broader LASA team.</p> <hd id="AN0182844507-35">Author Contributions</hd> <p>Conceptualization: DA; Methodology: DA, KS, JP, MW; Formal analysis and investigation: MS, DA; Writing-original draft preparation, review and editing: DA, MS, KS, JP, MW; Funding acquisition: DA, JP, MW.</p> <hd id="AN0182844507-36">Funding</hd> <p>The financial support of the Cooperative Research Centre for Living with Autism (Autism CRC), established and supported under the Australian Government's Cooperative Research Centres Program, is also acknowledged.</p> <hd id="AN0182844507-37">Data Availability</hd> <p>The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.</p> <hd id="AN0182844507-38">Declarations</hd> <p></p> <hd id="AN0182844507-39">Competing interests</hd> <p>All authors declares that they have no competing interests with respect to this publication.</p> <hd id="AN0182844507-40">Supplementary Information</hd> <p>Below is the link to the electronic supplementary material.</p> <p>Graph: Supplementary file1 (DOCX 88 kb)</p> <hd id="AN0182844507-41">Publisher's Note</hd> <p>Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.</p> <ref id="AN0182844507-42"> <title> References </title> <blist> <bibl id="bib1" idref="ref53" type="bt">1</bibl> <bibtext> Akogul S, Erisoglu M. An approach for determining the number of clusters in a model-based cluster analysis. Entropy. 2017; 19; 9: 452-452. 10.3390/e19090452</bibtext> </blist> <blist> <bibl id="bib2" idref="ref1" type="bt">2</bibl> <bibtext> Ashburner J, Ziviani J, Rodger S. Surviving in the mainstream: Capacity of children with autism spectrum disorders to perform academically and regulate their emotions and behavior at school. Research in Autism Spectrum Disorders. 2010; 4; 1: 18-27. 10.1016/j.rasd.2009.07.002</bibtext> </blist> <blist> <bibl id="bib3" idref="ref7" type="bt">3</bibl> <bibtext> Bullen JC, Zajic MC, McIntyre N, Solari E, Mundy P. Patterns of math and reading achievement in children and adolescents with autism spectrum disorder. Research in Autism Spectrum Disorders. 2022. 10.1016/j.rasd.2022.101933</bibtext> </blist> <blist> <bibl id="bib4" idref="ref63" type="bt">4</bibl> <bibtext> Carter Leno V, Vitoratou S, Kent R, Charman T, Chandler S, Jones CRG, Happé F, Baird G, Pickles A, Simonoff E. Exploring the neurocognitive correlates of challenging behaviours in young people with autism spectrum disorder. Autism. 2019; 23; 5: 1152-1164. 10.1177/1362361318769176. 30288984</bibtext> </blist> <blist> <bibl id="bib5" idref="ref14" type="bt">5</bibl> <bibtext> Chen L, Abrams DA, Rosenberg-Lee M, Iuculano T, Wakeman HN, Prathap S, Chen T, Menon V. Quantitative analysis of heterogeneity in academic achievement of children with autism. Clinical Psychological Science. 2019; 7; 2: 362-380. 10.1177/2167702618809353. 31032147</bibtext> </blist> <blist> <bibl id="bib6" idref="ref65" type="bt">6</bibl> <bibtext> Clark M, Adams D, Roberts J, Westerveld M. How do teachers support their students on the autism spectrum in Australian primary schools?. Journal of Research in Special Educational Needs. 2020; 20; 1: 38-50. 10.1111/1471-3802.12464</bibtext> </blist> <blist> <bibl id="bib7" idref="ref55" type="bt">7</bibl> <bibtext> Cohen J. Statistical power analysis for the behavioral sciences. 19882; L. Erlbaum Associates</bibtext> </blist> <blist> <bibl id="bib8" idref="ref26" type="bt">8</bibl> <bibtext> DiPerna JC, Elliott SN. Development and validation of the academic competence evaluation scales. Journal of Psychoeducational Assessment. 1999; 17; 3: 207-225. 10.1177/073428299901700302</bibtext> </blist> <blist> <bibl id="bib9" idref="ref46" type="bt">9</bibl> <bibtext> DiPerna JC, Elliott SN. Academic competence evaluation scales. 2000; Psychological Corporation</bibtext> </blist> <blist> <bibtext> DiPerna JC, Volpe RJ, Elliott SN. A model of academic enablers and elementary reading/language arts achievement. School Psychology Review. 2002; 31; 3: 298-312. 10.1080/02796015.2002.12086157</bibtext> </blist> <blist> <bibtext> Garnier, S. (2018). viridis: Default color maps from "matplotlib". R package version 0.5. 1. https://CRAN.R-project.org/package=viridis</bibtext> </blist> <blist> <bibtext> Gentles SJ, Ng-Cordell EC, Hunsche MC, McVey AJ, Bednar ED, Chen YJ, Duku E, Kerns CM, Banfield L, Szatmari P, Georgiades S. Trajectory research in children with an autism diagnosis: A scoping review. Autism. 2023. 10.1177/13623613231170280. 37194194</bibtext> </blist> <blist> <bibtext> Goodman R. The strengths and difficulties questionnaire: A research note. Journal of Child Psychology and Psychiatry and Allied Disciplines. 1997; 38; 5: 581-586. 10.1111/j.1469-7610.1997.tb01545.x. 9255702</bibtext> </blist> <blist> <bibtext> Johnels JA, Carlsson E, Norbury C, Gillberg C, Miniscalco C. Current profiles and early predictors of reading skills in school-age children with autism spectrum disorders: A longitudinal, retrospective population study. Autism: the International Journal of Research and Practice. 2019; 23; 6: 1449-1459. 10.1177/1362361318811153</bibtext> </blist> <blist> <bibtext> Johnels JA, Fernell E, Kjellmer L, Gillberg C, Norrelgen F. Language/cognitive predictors of literacy skills in 12-year-old children on the autism spectrum. Logopedics, Phoniatrics, Vocology. 2022; 47; 3: 166-170. 10.1080/14015439.2021.1884897</bibtext> </blist> <blist> <bibtext> Johnston O, Wildy H. The effects of streaming in the secondary school on learning outcomes for Australian students—a review of the international literature. Australian Journal of Education. 2016; 60; 1: 42-59. 10.1177/0004944115626522</bibtext> </blist> <blist> <bibtext> Keen D, Adams D, Simpson K. Teacher ratings of academic skills and academic enablers of children on the autism spectrum. International Journal of Inclusive Education. 2023; 27; 10: 1085-1101. 10.1080/13603116.2021.1881626</bibtext> </blist> <blist> <bibtext> Keen D, Webster A, Ridley G. How well are children with autism spectrum disorder doing academically at school?. An Overview of the Literature. Autism. 2016; 20; 3: 276-294. 10.1177/1362361315580962. 25948598</bibtext> </blist> <blist> <bibtext> Kuznetsova A, Brockhoff PB, Christensen RHB. lmerTest package: Tests in linear mixed effects models. Journal of Statistical Software. 2017. 10.18637/jss.v082.i13</bibtext> </blist> <blist> <bibtext> Kwok EYL, Brown HM, Smyth RE, Oram Cardy J. Meta-analysis of receptive and expressive language skills in autism spectrum disorder. Research in Autism Spectrum Disorders. 2015; 9: 202-222. 10.1016/j.rasd.2014.10.008</bibtext> </blist> <blist> <bibtext> Lenth, R, Singmann, H, Love, J, Buerkner, P, & Herve, M. (2020). emmeans: Estimated marginal means. R package version 1.4. 8</bibtext> </blist> <blist> <bibtext> Loubersac J, Michelon C, Ferrando L, Picot MC, Baghdadli A. Predictors of an earlier diagnosis of autism spectrum disorder in children and adolescents: A systematic review (1987–2017). European Child & Adolescent Psychiatry. 2021; 32: 375-393. 10.1007/s00787-021-01792-9</bibtext> </blist> <blist> <bibtext> Maenner MJ, Shaw KA, Bakian AV, Bilder DA, Durkin MS, Esler A, Furnier SM, Hallas L, Hall-Lande J, Hudson A. Prevalence and characteristics of autism spectrum disorder among children aged 8 years—Autism and developmental disabilities monitoring network, 11 sites, United States, 2018. MMWR Surveillance Summaries. 2021; 70; 11: 1-16. 10.15585/mmwr.ss7011a1. 8639027</bibtext> </blist> <blist> <bibtext> Mandy W, Murin M, Baykaner O, Staunton S, Hellriegel J, Anderson S, Skuse D. The transition from primary to secondary school in mainstream education for children with autism spectrum disorder. Autism: the International Journal of Research and Practice. 2016; 20; 1: 5-13. 10.1177/1362361314562616. 25576142</bibtext> </blist> <blist> <bibtext> Mayes SD, Calhoun S. Learning, attention, writing and processing speed in typical children and children with ADHD, autism, anxiety, depression and oppositional-defiant disorder. Child Neuropsychology. 2007; 13; 6: 469-493. 10.1080/09297040601112773. 17852125</bibtext> </blist> <blist> <bibtext> Mayes SD, Calhoun SL. WISC-IV and WIAT-II profiles in children with high-functioning autism. Journal of Autism and Developmental Disorders. 2008; 38; 3: 428-439. 10.1007/s10803-007-0410-4. 17610151</bibtext> </blist> <blist> <bibtext> McDougal E, Riby DM, Hanley M. Profiles of academic achievement and attention in children with and without autism spectrum disorder. Research in Developmental Disabilities. 2020. 10.1016/j.ridd.2020.103749. 32858397</bibtext> </blist> <blist> <bibtext> McIntyre NS, Solari EJ, Grimm RP, Lerro L, Gonzales E, Mundy PC. A comprehensive examination of reading heterogeneity in students with high functioning autism: Distinct reading profiles and their relation to autism symptom severity. Journal of Autism and Developmental Disorders. 2017; 47; 4: 1086-1101. 10.1007/s10803-017-3029-0. 28160222</bibtext> </blist> <blist> <bibtext> Milgramm A, Christodulu KV, Rinaldi ML. Brief report: Predictors of teacher-rated academic competence in a clinic sample of children with and without autism spectrum disorder. Journal of Autism and Developmental Disorders. 2021. 10.1007/s10803-020-04680-4. 33893937</bibtext> </blist> <blist> <bibtext> Miller LE, Burke JD, Troyb E, Knoch K, Herlihy LE, Fein DA. Preschool predictors of school-age academic achievement in autism spectrum disorder. The Clinical Neuropsychologist. 2017; 31; 2: 382-403. 10.1080/13854046.2016.1225665. 27705180</bibtext> </blist> <blist> <bibtext> Oswald TM, Beck JS, Iosif AM, McCauley JB, Gilhooly LJ, Matter JC, Solomon M. Clinical and cognitive characteristics associated with mathematics problem solving in adolescents with autism spectrum disorder. Autism Research. 2016; 9; 4: 480-490. 10.1002/aur.1524. 26418313</bibtext> </blist> <blist> <bibtext> Roberts JMA, Adams D, Heussler H, Keen D, Paynter J, Trembath D, Marleen W, Williams K. Protocol for a prospective longitudinal study investigating the participation and educational trajectories of Australian students with autism. British Medical Journal Open. 2018; 8; 1: e017082. 10.1136/bmjopen-2017-017082</bibtext> </blist> <blist> <bibtext> Rosenberg JM, Beymer PN, Anderson DJ, Van Lissa CJ, Schmidt JA. tidyLPA: An R package to easily carry out latent profile analysis (LPA) using open-source or commercial software. Journal of Open Source Software. 2018; 3; 30: 978. 10.21105/joss.00978</bibtext> </blist> <blist> <bibtext> Rimfeld JM, Malanchini PN, Hannigan DJ, Dale CJ, Allen JA. Teacher assessments during compulsory education are as reliable, stable and heritable as standardized test scores. Journal of Child Psychology and Psychiatry. 2019; 60; 12: 1278-1288. 10.1111/jcpp.13070. 31079420</bibtext> </blist> <blist> <bibtext> Rutter M, Bailey A, Lord C. (SCQ) The social communication questionnaire. 2003; Western Psychological Services</bibtext> </blist> <blist> <bibtext> Solari EJ, Grimm RP, McIntyre NS, Zajic M, Mundy PC. Longitudinal stability of reading profiles in individuals with higher functioning autism. Autism: the International Journal of Research and Practice. 2019; 23; 8: 1911-1926. 10.1177/1362361318812423. 30866651</bibtext> </blist> <blist> <bibtext> Stone LL, Otten R, Engels RCME, Vermulst AA, Janssens JMAM. Psychometric properties of the parent and teacher versions of the Strengths and Difficulties Questionnaire for 4- to 12-year-olds: A review. Clinical Child and Family Psychology Review. 2010; 13; 3: 254-274. 10.1007/s10567-010-0071-2. 20589428. 2919684</bibtext> </blist> <blist> <bibtext> van Rooij ECM, Jansen EPWA, van de Grift WJCM. Secondary school students' engagement profiles and their relationship with academic adjustment and achievement in university. Learning and Individual Differences. 2017; 54: 9-19. 10.1016/j.lindif.2017.01.004</bibtext> </blist> <blist> <bibtext> Wei X, Christiano ERA, Yu JW, Wagner M, Spiker D. Reading and math achievement profiles and longitudinal growth trajectories of children with an autism spectrum disorder. Autism: the International Journal of Research and Practice. 2015; 19; 2: 200-210. 10.1177/1362361313516549. 24449604</bibtext> </blist> <blist> <bibtext> Whitby PJS, Mancil GR. Academic achievement profiles of children with high functioning autism and Asperger syndrome: A review of the literature. Education and Training in Developmental Disabilities. 2009; 44; 4: 551-560</bibtext> </blist> <blist> <bibtext> Wickham HWickham H. Programming with ggplot2. ggplot2: Elegant graphics for data analysis. 2016; Springer. 10.1007/978-3-319-24277-4</bibtext> </blist> <blist> <bibtext> Zaidman-Zait A, Mirenda P, Szatmari P, Duku E, Smith IM, Zwaigenbaum L, Vaillancourt T, Kerns C, Volden J, Waddell C, Bennett T, Georgiades S, Ungar WJ, Elsabbagh M. Profiles and predictors of academic and social school functioning among children with autism spectrum disorder. Journal of Clinical Child and Adolescent Psychology. 2021; 50; 5: 656-668. 10.1080/15374416.2020.1750021. 32324064</bibtext> </blist> </ref> <aug> <p>By Dawn Adams; Matt Stainer; Kate Simpson; Jessica Paynter and Marleen Westerveld</p> <p>Reported by Author; Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib25" firstref="ref2"></nolink> <nolink nlid="nl2" bibid="bib18" firstref="ref3"></nolink> <nolink nlid="nl3" bibid="bib20" firstref="ref4"></nolink> <nolink nlid="nl4" bibid="bib28" firstref="ref5"></nolink> <nolink nlid="nl5" bibid="bib36" firstref="ref6"></nolink> <nolink nlid="nl6" bibid="bib26" firstref="ref8"></nolink> <nolink nlid="nl7" bibid="bib14" firstref="ref10"></nolink> <nolink nlid="nl8" bibid="bib39" firstref="ref13"></nolink> <nolink nlid="nl9" bibid="bib15" firstref="ref18"></nolink> <nolink nlid="nl10" bibid="bib27" firstref="ref23"></nolink> <nolink nlid="nl11" bibid="bib42" firstref="ref24"></nolink> <nolink nlid="nl12" bibid="bib10" firstref="ref25"></nolink> <nolink nlid="nl13" bibid="bib17" firstref="ref27"></nolink> <nolink nlid="nl14" bibid="bib12" firstref="ref34"></nolink> <nolink nlid="nl15" bibid="bib29" firstref="ref35"></nolink> <nolink nlid="nl16" bibid="bib34" firstref="ref37"></nolink> <nolink nlid="nl17" bibid="bib31" firstref="ref38"></nolink> <nolink nlid="nl18" bibid="bib32" firstref="ref40"></nolink> <nolink nlid="nl19" bibid="bib35" firstref="ref45"></nolink> <nolink nlid="nl20" bibid="bib13" firstref="ref49"></nolink> <nolink nlid="nl21" bibid="bib37" firstref="ref50"></nolink> <nolink nlid="nl22" bibid="bib33" firstref="ref54"></nolink> <nolink nlid="nl23" bibid="bib19" firstref="ref56"></nolink> <nolink nlid="nl24" bibid="bib21" firstref="ref57"></nolink> <nolink nlid="nl25" bibid="bib41" firstref="ref58"></nolink> <nolink nlid="nl26" bibid="bib11" firstref="ref59"></nolink> <nolink nlid="nl27" bibid="bib23" firstref="ref80"></nolink> <nolink nlid="nl28" bibid="bib40" firstref="ref86"></nolink> <nolink nlid="nl29" bibid="bib38" firstref="ref87"></nolink> <nolink nlid="nl30" bibid="bib24" firstref="ref88"></nolink> <nolink nlid="nl31" bibid="bib16" firstref="ref89"></nolink> <nolink nlid="nl32" bibid="bib22" firstref="ref90"></nolink> <nolink nlid="nl33" bibid="bib30" firstref="ref95"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1460676
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Profiles and Longitudinal Growth Trajectories of Teacher-Rated Academic Skills and Enablers in Autistic Children and Adolescents
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Dawn+Adams%22">Dawn Adams</searchLink> (ORCID <externalLink term="http://orcid.org/0000-0001-8001-0126">0000-0001-8001-0126</externalLink>)<br /><searchLink fieldCode="AR" term="%22Matt+Stainer%22">Matt Stainer</searchLink><br /><searchLink fieldCode="AR" term="%22Kate+Simpson%22">Kate Simpson</searchLink><br /><searchLink fieldCode="AR" term="%22Jessica+Paynter%22">Jessica Paynter</searchLink><br /><searchLink fieldCode="AR" term="%22Marleen+Westerveld%22">Marleen Westerveld</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Journal+of+Autism+and+Developmental+Disorders%22"><i>Journal of Autism and Developmental Disorders</i></searchLink>. 2025 55(1):267-283.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Springer. Available from: Springer Nature. One New York Plaza, Suite 4600, New York, NY 10004. Tel: 800-777-4643; Tel: 212-460-1500; Fax: 212-460-1700; e-mail: customerservice@springernature.com; Web site: https://link.springer.com/
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 17
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2025
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Audience
  Label: Education Level
  Group: Audnce
  Data: <searchLink fieldCode="EL" term="%22Elementary+Education%22">Elementary Education</searchLink><br /><searchLink fieldCode="EL" term="%22Secondary+Education%22">Secondary Education</searchLink>
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Autism+Spectrum+Disorders%22">Autism Spectrum Disorders</searchLink><br /><searchLink fieldCode="DE" term="%22Elementary+School+Students%22">Elementary School Students</searchLink><br /><searchLink fieldCode="DE" term="%22Secondary+School+Students%22">Secondary School Students</searchLink><br /><searchLink fieldCode="DE" term="%22Student+Motivation%22">Student Motivation</searchLink><br /><searchLink fieldCode="DE" term="%22Learner+Engagement%22">Learner Engagement</searchLink><br /><searchLink fieldCode="DE" term="%22Interpersonal+Competence%22">Interpersonal Competence</searchLink><br /><searchLink fieldCode="DE" term="%22Study+Skills%22">Study Skills</searchLink><br /><searchLink fieldCode="DE" term="%22Individualized+Education+Programs%22">Individualized Education Programs</searchLink><br /><searchLink fieldCode="DE" term="%22Elementary+School+Teachers%22">Elementary School Teachers</searchLink><br /><searchLink fieldCode="DE" term="%22Secondary+School+Teachers%22">Secondary School Teachers</searchLink><br /><searchLink fieldCode="DE" term="%22Student+Evaluation%22">Student Evaluation</searchLink><br /><searchLink fieldCode="DE" term="%22Individualized+Instruction%22">Individualized Instruction</searchLink><br /><searchLink fieldCode="DE" term="%22Learning+Trajectories%22">Learning Trajectories</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1007/s10803-023-06186-1
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0162-3257<br />1573-3432
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: In non-autistic children, academic skills are associated with academic enablers (motivation, engagement, study/interpersonal skills), but few studies have explored these in autistic children. This study identified profiles of academic skills and enablers in autistic students and explored the trajectory of each profile over time. Teachers completed the Academic Competences Evaluation Scales for autistic children in primary and secondary educational settings annually for 5 years. Latent profile analysis identified six profiles in the primary/younger cohort and seven in the secondary/older cohort. Whilst some profiles showed relative stability across skills and enablers, others profiles were more variable. The profiles remained stable and significantly different from each other over time, with no profile x time interactions identified. Autistic children may show variability across their academic skills and enablers. This highlights the importance of understanding each individual student and their profile of strengths and challenges when planning supports.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2025
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1460676
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1460676
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s10803-023-06186-1
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 17
        StartPage: 267
    Subjects:
      – SubjectFull: Autism Spectrum Disorders
        Type: general
      – SubjectFull: Elementary School Students
        Type: general
      – SubjectFull: Secondary School Students
        Type: general
      – SubjectFull: Student Motivation
        Type: general
      – SubjectFull: Learner Engagement
        Type: general
      – SubjectFull: Interpersonal Competence
        Type: general
      – SubjectFull: Study Skills
        Type: general
      – SubjectFull: Individualized Education Programs
        Type: general
      – SubjectFull: Elementary School Teachers
        Type: general
      – SubjectFull: Secondary School Teachers
        Type: general
      – SubjectFull: Student Evaluation
        Type: general
      – SubjectFull: Individualized Instruction
        Type: general
      – SubjectFull: Learning Trajectories
        Type: general
    Titles:
      – TitleFull: Profiles and Longitudinal Growth Trajectories of Teacher-Rated Academic Skills and Enablers in Autistic Children and Adolescents
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Dawn Adams
      – PersonEntity:
          Name:
            NameFull: Matt Stainer
      – PersonEntity:
          Name:
            NameFull: Kate Simpson
      – PersonEntity:
          Name:
            NameFull: Jessica Paynter
      – PersonEntity:
          Name:
            NameFull: Marleen Westerveld
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 0162-3257
            – Type: issn-electronic
              Value: 1573-3432
          Numbering:
            – Type: volume
              Value: 55
            – Type: issue
              Value: 1
          Titles:
            – TitleFull: Journal of Autism and Developmental Disorders
              Type: main
ResultId 1