04 SES 01 A, New Research on Autism, Aspergers and Inclusion
This paper investigates ethnic disproportionality in the identification of Autistic Spectrum Disorders (ASD), a category of Special Educational Needs (SEN), amongst primary school pupils in England. Disproportionality exists when pupils from an ethnic minority group are more (or less) likely to be identified with SEN compared to pupils in the ethnic majority. Being identified with SEN may facilitate access to additional resources, but it may also carry social stigma. Either way, the over- or under-representation of a particular ethnic minority group has important educational and social consequences.
Ethnic disproportionality in special education has long been a concern particularly in the US (Dunn, 1968) and England (Coard, 1971). Although there is less research in other European countries, in Denmark learners from minority ethnic backgrounds have been reported to be up to twice as likely as their peers to be placed in special education (Berhanu & Dyson, 2012) and in Germany a similar result is reported (Werning, Loser & Urban, 2008). In this paper, we focus on one type of SEN, Autistic Spectrum Disorder (ASD), because this is the most common primary type of need amongst students with Education, Health and Care Plans (EHCP) in England, and because the rate of ASD identification has more than doubled in the 10 years between 2005 and 2014 (Strand & Lindorff, in preparation).
Reviews of research from the US on ethnic differences in ASD prevalence have concluded that little information is available and results are inconclusive (Newschaffer et al, 2008, p240). National surveys of parental reports of autism diagnosis have suggested comparable frequencies in Black and White groups, but significantly lower frequencies for Hispanic children (CDCP, 2006). More detailed data suggest ASD prevalence rates among 8-year olds for White students (0.90%) are somewhat higher than among Black or African American students (0.72%) but in particular are higher than among Hispanic students (0.59%) (CDCP, 2009). Thus the odds for having an identified ASD are about 35% lower for Hispanic students (a relative risk ratio of 0.65:1) compared to the odds for White students. In England Strand & Lindsay (2009) report an analysis of national population data for 6M students aged 5-16 revealing substantial under-representation for Asian groups, with the odds of identification of ASD for Indian, Pakistani & Bangladeshi pupils half the size of the odds for White British pupils. These results do not seem to reflect socio-economic disadvantages, since Mandell et al (2009) have reported that the under-representation of Black and Hispanic groups remained after control for gender, IQ, birth weight and maternal education, and Strand & Lindsay (2012) report that controls for pupils entitlement to Free School Meals (FSM) and neighbourhood deprivation only increased the relative under-representation of Asian groups for ASD.
None of the above studies have taken a longitudinal approach. In this paper we focus on the earliest years at school and our study is, to our knowledge, the first to explore ethnic disproportionality in the emergence of ASD identification in a longitudinal study over ages 4-11 with national pupil level data. Our approach allows us to account for the effects on identification of controlling for a range of socio-economic variables and educational achievement and development in the first year at school.
Specific research questions addressed are:
- What ethnic groups, if any, are over-represented or under-represented for ASD relative to the majority (White British) group?
- Can any over- or under-representation be accounted for by socio-economic variables and early social and educational development, given that some ethnic groups are disproportionately exposed to early risk?
- Are school composition variables associated with ethnic disproportionality in ASD identification?
The study is based on pupil level data drawn from the England National Pupil Database (NPD). The 562,274 students starting Reception class at age 4+ in September 2008 are the focal cohort, and we access their records from the Annual School Census every January between 2008 and 2015. Key data includes: • Ethnic group (18 categories) • Primary type of SEN (12 categories) • Level of SEN (school support or statemented) • Gender (Boy/Girl) • Birth season (autumn, spring or summer born) • Entitlement to a Free School Meal (FSM) • Income Deprivation Affecting Children Index (IDACI) a measure of socio-economic deprivation in the neighbourhood where the child resides (the 32,000 SOAs in England) • Measures of Communication, Language and Literacy (CLL), Problem Solving, Reasoning & Numeracy (PSRN) and Personal, Social and Emotional Development (PSED), drawn from teachers’ ratings at the end of Reception Year. Our dependent variable is whether, and in what year, the pupil is identified with SEN where the SEN primary type is ASD. We utilise Cox’s regression (sometimes also called event history analysis or logit hazard modelling) to identify how the likelihood of ASD identification cumulates over time. This more accurately reflects the likelihood of SEN identification which is not a single-point in time event but instead occurs over time as children age. We identify the Hazard Ratios (HR) of identification for each ethnic minority group against White British pupils. Given the huge size of the dataset (n=560,000+) statistical significance is a very poor indicator of educationally meaningful effects. We identify HRs >1.33:1, or <0.75:1 as evidence of substantial disproportionality. These thresholds indicate the ethnic group is one-third more likely (4:3), or one-third less likely (3:4), to be identified relative to the White British majority in any particular year of a pupil’s primary education. We see how these Hazard Ratios change as we include controls for variables such as age, gender, socio-economic circumstances and educational and social development in the pupil’s first year of school. We also test school level variables, including school type, school size and quintile bands for % FSM pupils and % Asian pupils in the school.
Our analyses reveal the following findings: • Black Other Groups (HR=1.55) were substantially more likely to be identified with ASD than White British pupils, but the risks were not substantially higher for Black Caribbean (HR=1.30) or Black African (HR=1.16) pupils. The largest effect was the under-representation of Asian pupils, where the odds of identification were as low as half the odds for White British pupils (Pakistani= 0.37, Indian= 0.44 and Bangladeshi= 0.73). • The Risk for identification for Boys were fives timer higher than the risk for girls. On their own socio-economic factors had little association with identification (e.g. FSM=0.96). Literacy and mathematics scores at Reception had no substantial association with identification, but a one SD increase in Personal, Social & Emotional development (PSED) score at age 4 was associated with a substantial reduction in ASD identification (HR=0.29). Interestingly after control for PSED score there was a significant negative association with entitlement to a FSM (HR=0.70) and neighbourhood deprivation (HR=0.86), suggesting that pupils from economically disadvantaged families are less likely to be identified than otherwise similar pupils from more advantaged backgrounds. • The above controls accounted for the over-representation of Black Other Groups (HR=1.20) but the odds for all Asian groups remained half the size of those for White British pupils. After adjusting for the pupil level variables, pupils in schools in the top two quintile of % Asian had even greater raised risks (HR=1.34 and 1.48 respectively) suggesting a school composition effect. This will be explored further in the full paper. We conclude that there is evidence of significant under-representation of pupils of Asian heritage among those identified with ASD, suggesting the possibility of substantial unmet need. The causes of ethnic disproportionality in identification of ASD are likely to be varied and the paper will discuss possible explanations and policy implications.
Berhanu, G., & Dyson, A. (2012). Special education in Europe, overrepresentation of minority students. In J. Banks (Ed.), Encyclopaedia of Diversity in Education (pp. 2070-2073). Thousand Oaks: SAGE. Coard, B. (1971). How the West Indian child is made educationally subnormal in the British school system: The scandal of the Black child in schools in Britain. London: New Beacon for the Caribbean Education and Community Workers’ Association. Centre for Disease Control and Prevention (CDCP) (2006). Mental health in the US: Parental reports of diagnosed autism in children aged 4-7 years- US 2003-2004. MMWR Morb Mortality Weekly Report 2006; 55(17): 481-486. Centre for Disease Control and Prevention (CDCP) (2009). Prevalence of Autism Spectrum Disorders - Autism and Developmental Disabilities Monitoring Network, US, 2006. MMWR surveillance Summary 2009; 58(NoSS-10). http://www.cdc.gov/mmwr/preview/mmwrhtml/ss5810a1.htm Corbett, C., & Perepa, P. (2007). Missing out? Autism, education and ethnicity: The reality for families today. London: The National Autistic Society. Dunn, L.M. (1968). Special education for the mildly retarded – Is much of it justifiable? Exceptional children, 23, 5-21. Kogan, M. D., Blumberg, S. J., Schieve, L. A., Boyle, C. A., Perrin, J. M., Ghandour, R. M., et al. (2009). Prevalence of Parent-Reported Diagnosis of Autism Spectrum Disorder Among Children in the US, 2007. Paediatrics, 124(5), 1395-1403. Mandell, D. et al. (2009). Racial/ethnic disparities in the identification of children with autism spectrum disorders. American Journal of Public Health, 99(3), 493-498. Newschaffer, C. J., Croen, L. A., Daniels, J., Giarelli, E., Grether, J. K., Levy, S. E., et al. (2007). The Epidemiology of Autism Spectrum Disorders*. Annual Review of Public Health, 28(1), 235-258. Strand, S. & Lindsay, G. (2009). Evidence of ethnic disproportionality in special education in an English population. Journal of Special Education, 43(3), 174-190. Strand, S. & Lindsay, G. (2012). Ethnic disproportionality in the identification of Speech Language and Communication Needs (SLCN) and Autistic Spectrum Disorders (ASD): 2005-2011. London: Department for Education. Strand, .S. & Lindorff, A. (in preparation). The unequal representation of ethnic minorities for different types of special educational needs: Extent, causes and consequences. Final report to the ESRC. Werning, R., Loser, J. & Urban, M. (2008). Cultural and Social Diversity: An Analysis of Minority Groups in German Schools. Journal of Special Education, 42, (1), 47-54.
The programme is updated regularly (each day in the morning)
- Search for keywords and phrases in "Text Search"
- Restrict in which part of the abstracts to search in "Where to search"
- Search for authors and in the respective field.
- For planning your conference attendance you may want to use the conference app, which will be issued some weeks before the conference
- If you are a session chair, best look up your chairing duties in the conference system (Conftool) or the app.