Dr. Umar Toseeb Hello, welcome to the Papers Podcast series for the Association for Child and Adolescent Mental Health, or ACAMH for short. I’m Umar Toseeb, Professor of Psychology. My research focuses on special educational needs and mental health in childhood and adolescence. All listeners to this and indeed, any of ACAMH’s podcasts, are eligible for a free CPD certificate. Do please visit acamhlearn.org for details of this, together with information on how you can access hundreds of hours of free talks, lectures, interviews, all of which you can also get free CPD certificates for. The web address is acamhlearn.org. That’s a-c-a-m-h-l-e-a-r-n.org. If you’re a fan of our Papers Podcast series, please subscribe on your preferred streaming platform, let us know how we did, with a rating or review, and do share with your friends and colleagues. Today, I’ll be speaking to Dr. Sarah Griffiths, Lead Author of the paper “Special Educational Needs Provision and Academic Outcomes for Children with Teacher Reported Language Difficulties at School Entry,” published in JCPP Advances. Sarah, thank you so much for joining me. Dr. Sarah Griffiths Hi, nice to be here. Dr. Umar Toseeb Do you just want to start by introducing yourself and also, your co-authors? Dr. Sarah Griffiths My name’s Sarah Griffiths and I’m a Senior Research Fellow at University College London in the Department of Clinical Educational and Health Psychology. And my research focuses on neurodevelopmental conditions, including autism and developmental language disorder, and I’m particularly interested in the mental health outcomes of children with these conditions and what determines these, but also, in their educational outcomes, which is what the paper we’re talking about today is about. And I wrote this paper with a team of people, including my colleague, Professor Courtenay Norbury, who set up the Surrey Communication and Language in Education Study, which is known as SCALES, which is what the paper is based on. And SCALES was set up to look at the prevalence of language disorders at the start of primary school and the outcomes for children who have language disorder as they move through education, and I joined the study when the children were in Year Six and we were particularly interested in looking at the social, emotional and mental health outcomes of these children during their transition from primary to secondary school. So, unfortunately, the last phase of data collection for SCALES was in the 2019/2020 academic year and so, it was disrupted by the pandemic. So, partly to get around this issue, we decided to apply to link our data to the National Pupil Database, which is held by the Department of Education and contains records of children’s academic outcomes and special educational needs. So, this allowed us to look at the topic of this paper, which is whether early language difficulties are associated with children’s outcomes on statutory tests and what additional support is available for them as they move through primary school. Dr. Umar Toseeb Excellent, thank you, and so, let’s get into the paper. So, you’ve talked – in the title, you say, “Teacher Reported Language Difficulties.” Shall we talk about language? So, what are the different definitions of the types of language that you’ll be talking about? And we’ve talked about language difficulties and you also mentioned developmental language disorder. So, should we just have a quick round of definitions? Dr. Sarah Griffiths Sure. So, I think I mentioned SCALES was initially set up to identify children with developmental language disorder. This is a really common but often undiagnosed condition which is characterised by difficulties both with expressive and receptive language. So, your receptive language is your ability to understand what other people are saying, and expressive language is your ability to use language to communicate your own thoughts and experiences. And difficulties with receptive language, they often come across as seeming disengaged or forgetful, whereas difficulties with expressive language often come across as not saying very much or being unable to explain something in a coherent way. And for this reason, language disorders are often not recognised as language disorders and may be misattributed to attentional or behavioural difficulties. So, developmental language disorder is the consensus term for persistent difficulties with both expressive and receptive language which are not explained by a known biomedical cause. So, that could be something like autism or Down syndrome, which children who have those conditions often also have language difficulties, but they wouldn’t get a diagnosis of developmental language disorder because it’s seen as part of those other conditions. Developmental language disorder doesn’t – also doesn’t include people who have sensory differences, so deafness or people who’ve had head injury which explain their language difficulties, and these language difficulties have to have a significant impact on your functioning, as well. So, SCALES conducted in-depth testing of both expressive and receptive language and associated cognitive functioning in around 600 children, and they estimated from that that 7.5% of children met the criteria for developmental language disorder when they started primary school. But in this paper, we’re using a broader definition of something we’re calling language difficulties, and the reason for that is that we’re not just looking at those 600 children who had in-depth assessments. We are looking at all of the children in SCALES whose Teachers completed a screening measure for language difficulties. So, we had that data on 7,000 children. So, our language difficulties group consists of children who scored in the top 14% for language difficulties as rated by their Teachers. So, this group is going to contain lots of children who have developmental language disorder, but also children who have language difficulties associated with other conditions that therefore wouldn’t meet the developmental language disorder criteria per se. Dr. Umar Toseeb Thank you. That’s a great introduction to the different types of language difficulties and language functioning there. Let’s give our listeners some context around the English education system. When do children typically start school and when does the identification of various types of difficulties start to happen? Dr. Sarah Griffiths The children start school when they’re five, and in some cases, if a child has a special educational need that’s already obvious to their parents and to people who’ve supported them before they come to school, then they will start school with a special educational need already registered, and in some cases they might actually start in a special setting. So, they’ll go to a special school right from the beginning of school. In other cases, children or parents might notice that the children has special educational needs once they start school, and then a child will have assessment and then at that point, they may be registered as having a special educational need, and that can happen at any point throughout their educational journey. Dr. Umar Toseeb And what kinds of special educational needs might children be identified as having during those primary years? Dr. Sarah Griffiths So, the most common in primary is the one that we are most interested in in this paper, which is speech, language and communication needs. But one of the interesting things we find in this paper is that, that starts off as being quite a common one, but as you go through the school system, it becomes less and less common, even though we know that language difficulties are very persistent. So, sometimes it reflects, like, what’s the most, kind of, salient to the educators at the time. So, in primary school, it tends to be speech, language, communication needs, is one of the most common ones. The other ones are social, emotional, behavioural difficulties, intellectual disability, specific learning difficulties. That’s things like dyslexia. So, there’s a range of different categories that you can get. Dr. Umar Toseeb Yeah, and I think what I’m hearing there and it’s what I understand of the labelling system in general, is that the label a child has is not necessarily reflective of their overall profile of difficulties. It’s just the one that’s been identified or is seen as most impactful on their learning. Dr. Sarah Griffiths Exactly, and so, also, you may have more than one special educational need. Like, you might have autism but you also have language and communication difficulties, and in the data that schools have or that – certainly the data that schools provide to the National Pupil Database, they’re only allowed to list two. So, they list a primary need. So, everyone who’s on the register has a primary need listed, or almost everybody. There are some people who are unidentified, and then they will sometimes have a secondary need listed as well, but in reality, it might be that they get a new diagnosis and this isn’t necessarily reflected in their special educational needs records. Dr. Umar Toseeb And we know that some children will have a special educational need but then they’ll also have an education, health and care plan, so EHCP. Dr. Sarah Griffiths Yes. Dr. Umar Toseeb Can you tell us a bit about what that is? Dr. Sarah Griffiths So, there’s, kind of, two levels of support that children will get. If they’re registered as having a special educational need, they’ll get some, kind of, special educational support in schools, but it’s not very specific. So, the school will get some extra funding based on how many children they have who are on the SEN register, but that funding doesn’t necessarily go towards something specific for that individual child. The enhanced level of support is something called an Educational Health and Care Plan, as you’ve said, and this is funding specifically for an individual child and it comes with a document that lays out exactly what their needs are and what support they should get. Much smaller proportion of children who have this, and it’s a much higher level of support and it will often include being in a specialist setting, so being in a special classroom in a mainstream school or being in a special school. Dr. Umar Toseeb Thank you, and I think – and the way that I try and explain the distinction between – or the relationship between an SEN label and having an EHCP, is, all children with an EHCP have an identified special educational need, but not all children with an identified special educational need will have an EHCP. It’s only a subgroup of those. Dr. Sarah Griffiths Exactly, yeah. Dr. Umar Toseeb And in your paper, you talk about the school entry measures. So, the Teacher reported language difficulties, that is the measure here. And you talk about the need to distinguish between children who have language impairments and other children who have English as an additional language, which is the reason why they might appear to have language impairments. Why is it important to look at those two groups separately? Dr. Sarah Griffiths Yes, thank you. So, I said that our language difficulties group is going to contain children, not just those children who have a specific language difficulties. The other type of child it might contain is children who don’t have proficient English because they speak another language at home. And it’s important to distinguish between those children and children who have English as their first language who are still showing language difficulties, because for children who just have poor English when they start school because they haven’t heard much English before, those children often pick up, like, English very quickly once they start school and they don’t go on to have any additional difficulties in terms of their educational outcomes. Of course, you can have children who have both. So, they speak a different language at home and they have a language disorder, but they’re going to be a real mix in that group of children who speak English as an additional language. So, in our paper, we separated out those two groups of children ‘cause we thought there might be different strength of relationship between language difficulties in a group of monolingual children compared to language difficulties in the group of children who speak English as an additional language, when we think about the relationship with later academic outcomes. Dr. Umar Toseeb Thank you. So, I think we’ve got a really good idea of the context now. Let’s get into your research questions. What were your research questions in this paper? Dr. Sarah Griffiths We had three main research questions. The first was, “Do language difficulties identified at the start of primary school predict children’s results on academic tests throughout primary school?” And these academic tests were SATs tests, which for people from the UK will be familiar. People outside the UK, these are statutory attainment tests that children complete when they’re seven years old and 11 years old. Our second research question was, “What predicts the receipt of special educational provision in primary school for children with Teacher reported language difficulties?” And we were only looking at monolingual children in that group, and we looked at this question both for thinking about what predicts being on the SEN register, and also, for what predicts receiving education in a specialist setting, so that’s in a special school or a specialist classroom in a mainstream school. And finally, we’re interested in, whether “receiving special educational provision was associated with improved academic outcomes for children with language difficulties.” Dr. Umar Toseeb Thank you, and I think we’ve already spoken about the SCALES study, so you’ve already described that to us. So, maybe we’ll just go on to the findings. So, what did you find? So, what did you find for the first research question, so, where you’re interested in the relationship between teacher reported language difficulties at school entry and subsequent academic performance? Dr. Sarah Griffiths Yes. So, we found a robust relationship between teacher reported language difficulties at school entry and scores on academic attainment tests, and this was across the curriculum. So, for the SATs tests that I mentioned, children do reading, writing and maths, and the strength of the association between language difficulties and outcomes on these tests was the same for maths as it was for English, which is interesting, and I think it just shows that language is important across the curriculum, not just in English, which you might think is the most obvious one. And the other thing we found, unexpectedly, is that the relationship was just as strong in children with English as an additional language as it was in monolingual children, and I think this is a little bit puzzling because of the reasons I just spoke about earlier, but this group of children with EAL is very heterogeneous. So, some children will have English as an additional language but they’ll still have English spoken at home. So, it is only going to be a small subgroup of children with EAL who actually haven’t had exposure to English. We weren’t really able to distinguish between those group who hadn’t had any exposure to English and those who probably would’ve had quite a bit of exposure to English because they just have, for example, one parent who speaks another language. They would still end up in that English as an additional language group. So, I think that’s probably the reason that we find the same strength of relationship, but it is a bit harder to explain. Dr. Umar Toseeb I’m not a linguist, so my knowledge of the structure and – of different languages is limited, but could it be that the additional languages the children were speaking in your study were quite similar to English anyway, whereas in other research, the additional languages that were being considered were quite different to English? Could that make a difference? Dr. Sarah Griffiths Possibly. I mean in the context in the UK, we have so many different – I can’t remember exactly how many languages was – were spoken as additional languages in SCALES, but it was, like, a lot. So, it’s a very – as I said, it’s a very heterogeneous group that’s going to contain lots of different possibilities. In other contexts, for example, in Canada where children speak either French or English and they’re, kind of, surrounded – I don’t want to say too much about that context, I don’t know loads about it, but I think depending on where you’re doing the research, the meaning of English as an additional language can change quite a lot. So, you have to, kind of, consider that when you’re thinking about results. Dr. Umar Toseeb Yeah, and a question about causality. So, I get this a lot in my own work. What you are finding is language difficulties predict academic performance. How do we know that that relationship is causal? ‘Cause in my head, it could also be that language difficulties or academic attainment or poor academic performance are both products of some other form of disadvantage. How do we know it’s not that? How do we know it’s language that’s causing these difficulties? Dr. Sarah Griffiths Yeah, I mean we absolutely can’t say for sure. So, as with all observational research, we can’t really say that language difficulties are causing poor academic attainment, and actually, it’s unlikely that that’s the only reason that this group of children are performing more poorly on these assessments. So, for example, we know that language difficulties are more common in children from lower socioeconomic status households. That comes across in this paper as well, ‘cause the – we’ve got a measure of deprivation based on postcodes and that is associated with having higher Teacher reported language difficulties. So, we know there’s this association, so some of the relationship with academic outcomes could be partly due to that. But what it does tell us, which is useful, is that if you have a child who has poor language when they start school, it’s a marker of later poor academic attainment. So, from a, sort of, practical perspective, if you have a child who’s got poor language that Teachers are noticing when they start school, schools need to think about what extra support they could put in place to support language, ‘cause what we do know is that language is really important in terms of understanding what’s going on in the classroom. Even if it’s not the only reason, it’s going to have some impact on your academic attainment, just because it’s so important in actually accessing the curriculum. Dr. Umar Toseeb Yeah, and I think the other thing that I was thinking about is some cognitive processes that might be important for both language learning and also, academic performance. So, things like short-term memory, phonological processing, those kinds of things, which are important for language learning, but also, I imagine would be important for subjects like maths, even where… Dr. Sarah Griffiths Yeah. Dr. Umar Toseeb …you have to be able to comprehend instruction and those kinds of things, and some of the other functions in mathematics might also require those processes. But I think the point that you make about, it’s not necessarily a case of these language difficulties causing these problems, although it might well be, it’s highlighting that the kids who have language difficulties at school entry are likely to go on to have difficulties with other things, as well. Dr. Sarah Griffiths Absolutely, yeah. Dr. Umar Toseeb And then your second research question, where you were interested in whether Teacher reported language difficulties at school entry predicted whether someone got specialist – special education provision. What did you find? Dr. Sarah Griffiths Yeah, so, in this research question we were only looking at the children who were in our language difficulties group, and what we were asking was, “What predicts whether they receive these different levels of support?” And our strongest predictors, as you might expect, were having an existing diagnosis of a developmental or sensory impairment when they started school or when they’re in the first year of school, and our next strongest predictor was Teacher reported academic attainment in that first year. So, the Teachers completed something called the Early Years Foundation Stage Profile, which is a measure of their academic progress in that first year. And that predicted whether they went on to receive these different levels of special educational need support, and also, their language – those within this group, the language – variation in language difficulties also predicted it. So, this is, kind of, reassuring in a sense that it seems like the children who need the support the most seem to be getting it. The other thing that we found in relation to socioeconomic status, which I think is really interesting, is we found a different result depending on whether we were looking at registration for special educational needs or receipt of special education in a specialist setting. So, remember that special education in a specialist setting requires you to have an EHCP plan. So, that’s the higher level of support, and getting this level of support is challenging, and often, we know from talking to parents, it often requires parents to, sort of, fight the system to get this higher level of support. And unfortunately, that is just easier if you have – if you are from a higher socioeconomic status household, because you just have more resources to do that, more financial resources, more education. So, what we found in our paper – so, I can’t really say exactly from our paper for this point, but what we found is that in terms of looking at who gets registered, we found that children from lower socioeconomic backgrounds were more likely to be registered with a special educational need. So, that’s the lower level of support, but in terms of who gets provision in a specialist setting, there was no difference between – in terms of socioeconomic status. I almost think that’s, like, two different things cancelling each other out. So, maybe you’re more likely to need the support if you’re from a lower socioeconomic status household, but maybe you’re less likely to be able to fight for it. So, it’s like – almost like these two things might be cancelling each other out there, although the paper doesn’t really allow me to say that exactly, but that’s, kind of, what I suspect is happening. Dr. Umar Toseeb It’s really interesting. So, I’m just going to unpick the first part of what you said. I think that there’s a piece of work by the Education Policy Institute where they looked at which factors are associated with SEN identification at the national level. And I think what they find is that a lot of the variation in whether a child is identified as having a special educational need, irrespective of whether they need it or not, is at the level of the school and at the local authority, and has very little to do with the child themselves. And I think from that, they infer that some local authorities in schools are just very good at identifying kids and others are less so, and it seems that in your study, in SCALES, which I think is in Surrey, right? Dr. Sarah Griffiths Yeah, that’s right. Dr. Umar Toseeb Yeah. So, it seems that there’s lots of that stuff of the identification might be happening as a result of the child’s needs rather than other things. But in the second part of what you’re saying, it seems that it is the lower SES children who are more likely to get an SEN label, which actually, I think would be not what I would expect. Dr. Sarah Griffiths Hmmm hmm. Dr. Umar Toseeb I would expect there’d be a disadvantage, but then you’re not finding any effect for EHCP, which is interesting. So, it’s really interesting to see that on a national level, we have a certain picture and then, at this local level in one county, you have some different dynamics. Dr. Sarah Griffiths Yeah, exactly. So, I think you’re right there, we should also think about the school, because the measures of SES that we have in SCALES are based on your postcode. So, you’re going to have – even within Surrey, which is obviously one of the more affluent counties, for people who aren’t familiar with the context, even within Surrey, you do have, like, variation in terms of the areas and with – in terms of the schools. So, schools in more affluent areas are probably going to have more resources and probably more able to provide additional support compared to schools in less affluent areas, and that might be some of the reason why you get this difference in EHCPs, but yeah, really interesting to compare with the national data, as well. Dr. Umar Toseeb Thank you, and then your third research question, from what I understand from the paper, it didn’t go to plan, as in you planned to do some analysis and then you did something else. Can you talk us through that? What happened? Dr. Sarah Griffiths So, we were really interested in whether receiving special educational provision, it was associated with better outcomes, ‘cause the idea behind special educational provision is that it’s supposed to be providing teaching in a more optimal way for children’s needs, and therefore, you would hope that this would improve academic outcomes for those children. The complicating factor is that a lot of the reasons why children end up in special education provision are confounded with the outcomes themselves. So, children with greater needs are more likely to end up in special educational provision, and they’re also more likely to do more poorly on these academic attainment tests. So, what we tried to do is something called propensity score matching, which is where you try to find a group of children who receive special educational provision, then you try to match them with children who didn’t on how likely they were to need it. So, you’re trying to find children who have the same kind of background characteristics, the same likelihood of scoring a certain amount on the SATs tests based on their needs, but unfortunately – well, maybe fortunately for these children, but unfortunately for our analysis, we couldn’t find very many children who had a strong propensity to need special educational provision who didn’t actually receive it. So, there just wasn’t a good enough control group for us to do that comparison. So, we weren’t able to do that analysis that we had planned. So, all we’re able to do is show the difference in terms of academic outcomes for those children who did and didn’t have a special educational needs registration. And what we find is that the children who did have a special educational needs registration performed more poorly than those who didn’t. But I would really caution against interpreting anything about the value of special education from that because those children who did receive the special educational provision had greater need. So, of course, they’re going to do more poorly in the end, unless special educational provision was so amazing that it completely cancelled out their difficulties, which obviously, it – we wouldn’t expect it to. So, that’s the reason for the change in the analysis plan there. Dr. Umar Toseeb Yeah, and I think it’s really important, like you say, to highlight there that the children who end up in special education – special provision for special educational needs, tend to have either a higher level of need or a different type of need. Dr. Sarah Griffiths Yeah. Dr. Umar Toseeb So, to just compare outcomes, I suppose is just – just tells you that there’s a difference but not why there’s a difference, and I imagine that’s probably why you wanted to do the propensity score analysis to try and match with those things to begin with. But, you know, sometimes the data doesn’t do what you want it to do, so you have to do the best you can. Dr. Sarah Griffiths Yeah, exactly, and the other thing I think I’d like to mention there is, we are only looking at academic outcomes in this case. And I think you might have some children who maybe they would do better academically in a mainstream setting, perhaps, but the consequence of them being in a mainstream setting is that their mental health really suffers because it’s extremely stressful for them and they actually can’t cope in that setting. So, it’s not just about the academic outcomes, but also, more broadly about what’s best for that child, and this study has – wasn’t able to look at that, and I think most studies that try to do this kind of comparison often don’t look at the, sort of, bigger picture. I think that’s something that’s, kind of, missing. Dr. Umar Toseeb Thank you, and then, thinking about the future, what’s the follow-up work that you’ve got planned that you’d like to tell us about, and is – where are we at with SCALES, as well? Is there a follow-up planned? Dr. Sarah Griffiths Yeah, there’s a couple of things to mention here. So, one of them is the SCALES children are going to be 18 this year, which is very exciting. We don’t have a plan to follow-up all of the children because – partly ‘cause of attrition and the pandemic and them moving schools and things means that it’s harder for us to actually re-contact everybody, but we want to do something similar to what I’ve done in this paper. So, we’re going – we are applying for funding at the moment to run a follow-up study where we look at academic outcomes all the way to the end of school and also, the transition into adulthood, so what happens to them post 18. And we plan to both use linked data, so linked to the National Pupil Database again, but also do some interviews with some of our young people from SCALES who have language and communication difficulties about their experiences of that transition and their experiences of the educational system, and I think that will give us some more insight into what other advantages and disadvantages of being in special educational needs settings and various types of special educational needs support. And the other thing to mention is that, I’m looking also in this dataset at absences and exclusions because we know from speaking to families with – and young people with language disorder, that, especially if language disorder is not recognised, children are more likely to be excluded for behavioural problems. ‘Cause as you can imagine, if you’re in an argument and you can’t explain yourself, you’re maybe more likely to lash out, and if people don’t understand the context, then you could be more likely to be excluded. But also school is extremely stressful for some children with language difficulties and so, school-related anxiety and school refusal is also more common in these young people, and of course, if you’re not actually at school, then that’s going to affect your academic attainment. So, some of the reason for this relationship between language difficulties and academic attainment might be partly due to absences and exclusion. So, that’s another analysis that I’m working on at the moment. Dr. Umar Toseeb Thank you, and I think when you were talking about linked data, it got me thinking that there’s lots and lots of linkage of data that’s happening at the moment to administrative data, lots of very – lots of different studies. Dr. Sarah Griffiths Hmmm hmm. Dr. Umar Toseeb Have you consented or could you also consent to linking, like, health service use or criminal convictions data, that – UCAS data, Inland Revenue, HMRC? Like, there’s so much data that can be linked, and I imagine for this cohort specifically, because we have a DLD sample, and within the England context, it’ll give us so much more information around these children’s journeys as they go through childhood and transition into adulthood. Dr. Sarah Griffiths Yeah, I’m really interested in looking at linking to health data, but it is extremely challenging. I mean, partly because the kinds of things that we’re interested in that might be associated with language disorder are not necessarily the kind of things that exist in those datasets. So, I’d be really interested in looking at, for example, referrals to Child and Adolescent Mental Health Services, but it’s quite difficult to actually find that data and link it, but yeah, that is something I’m interested in doing. Dr. Umar Toseeb Watch this space. Dr. Sarah Griffiths Watch this space. Dr. Umar Toseeb Because I think I’m doing that with ALSPAC for DLD. So, we will find… Dr. Sarah Griffiths Okay, lovely. Dr. Umar Toseeb Hopefully we’ll find out soon. Okay, excellent, that’s been a fantastic conversation. What is your take-home message for our listeners? Dr. Sarah Griffiths My take-home message is that language proficiency is really critical for academic success. Language difficulties are a good marker of long-term or academic attainment later in life, and the implications of this are that we should continue to recognise and support language difficulties throughout children’s academic journeys. I mentioned that one of the, sort of – wasn’t one of our main research questions, but one of the things that came out of this paper was that the recognition of language difficulties tends to drop off as children move through the educational system, particularly when they transition from primary to secondary school. And I think that’s partly to do with the priorities of the schools rather than a change in the children’s needs, ‘cause we know that language disorder is really persistent. Dr. Umar Toseeb Thank you, Sarah. Thank you so much for that. For more details, please visit the ACAMH website, www.acamh.org, and Twitter @ACAMH. ACAMH is spelt A-C-A-M-H, and don’t forget to follow us on your preferred streaming platform, let us know if you enjoy the podcast, with a rating or review, and do share with your friends and colleagues.