Practical Case Study Series #1 – SEN or EAL?

Student N: The slow processor?

Normal bilingual learner or special need?

Student N was a casual admission, i.e., mid-year, and came to our secondary school directly into Year 9. She seemed happy, socially confident and willing to learn, but was extremely slow to acquire English. In the group of students who were all new admissions, she was making the slowest progress out of all of them.

This concerned us as, in the UK system, she was immediately forced into exam classes and needed to acquire English at an accelerated rate. Was she SEN? How would we know? What information was needed to evaluate her? How could we document our decision-making process to ensure validity of approach?

Using hypothesis testing

Frederickson and Cline (1991, 2015 ) suggest a comprehensive framework of information gathering when assessing whether an EAL student is experiencing curriculum difficulties due to cognitive deficiency or to a need for linguistic and socio-cultural adjustment. These are a part of a ‘hypothesis-testing approach’ (Frederickson and Cline 1991, Hall 2001) where possible variables contributing to the educational difficulty are considered and eliminated as part of the assessment of possible SEN in an EAL child.

Get the Frederickson and Cline book.
Get the Deryn Hall book.

These include:

  1. Background: gaining a full picture of the child’s previous educational experience;
  2. Language: conducting a first-language assessment to ascertain whether the child is working at an age-appropriate level in their own language;
  3. Communication Skills: observations of pupil interaction in classroom and play contexts, as well as gathering observational information from community language or religious schools as part of a multi-cultural approach;
  4. Differentiation: checking that appropriate classroom provision for EAL language learning need is effectively used;
  5. Affective Filters: investigating emotional and psychological factors affecting achievement such as past trauma, racist bullying or other environmental stresses; (Frederickson and Cline 1991, Pim 2010)
  6. Testing: considering raw data on national standardised tests but as a minimal part of the whole picture. See note.

Note: standardised tests usually have no construct validity for EAL pupils, i.e., they do not measure what they are designed to test for in pupils who do not understand the test instructions or questions,  and therefore are not a good source of evidence to confirm or deny a hypothesis of cognitive ability.

Get the Chris Pim book.

Frederickson and Cline (1991) stress that only after these questions are explored in turn should the hypothesis that a learning or language disorder be considered. This integrated four-fold approach had been incorporated in the older 2001 SEND Code of Practice: ‘the child’s learning characteristics, the learning environment that the school is providing for the child, the task and the teaching style’.  (DfES, 2001).

In the current 2014 SEND Code of Practice, the need for a wider approach to assessment is explained thus: ‘Identifying and assessing SEN for children or young people whose first language is not English requires particular care. Schools should look carefully at all aspects of a child or young person’s performance in different areas of learning and development or subjects to establish whether lack of progress is due to limitations in their command of English or if it arises from SEN or a disability. Difficulties related solely to limitations in English as an additional language are not SEN.’ (DfE, Section 6.24)

Let’s look at each of the hypothesised question areas in turn for Student N.

  1. Background:

Bilingual Interview – a bilingual interview should get info on:

  • previous schooling experience and any gaps in schooling / when school started
  • (some school systems start at 7 years old – this does not constitute a ‘gap’ in schooling)
  • with who the child is here and some experience of how they came (this is sensitive stuff – don’t dig too deep, and check local legislation about the legality of getting this info – generally your Data Manager should know)
  • what languages they speak and with whom they speak them
  • preferred subjects or school experiences
  • medical needs / allergies / glasses / hearing or eye test ever taken
  • religious practices including main holiday and food needs
  • ambitions – what do they want to do in the future?

Student N was from an area in Bulgaria that was universally bilingual in Turkish and Bulgarian. She was fully educated as per the Bulgarian system, but had some schooling in Turkish and some in Bulgarian. With her family, she spoke both, although different family members had different linguistic strengths and so the ‘home language’ changed depending on with whom she spoke in her family. She wanted to go to university, study business and open her own. She had been tested for hearing and eyesight through her early school years, had no medical issues and considered herself to be Muslim, although not from a tradition that required a head covering.

Partial Conclusion: N’s slowness might be due to her cognitive working in three languages. When we introduced an academic concept, she literally checked through her mental  Bulgarian dictionary for the equivalent and then through her Turkish one to see when a comprehensible match could be found. She would then map her matched concept onto the English word to understand it. If this hypothesis were true, it would account for her slow processing: not because it was taking her three times as long to process one concept, but because she was doing three times the work in checking through her two equally strong languages for a concept match to the third language, English.

This also worked backwards, because if the English concept, say smelting, was found in her mental Turkish dictionary, she would then need to create an equivalent in Bulgarian for herself while also mapping the Turkish to English. This was to enable her to access all her higher order thinking resources across her three languages with equal strength—quite a feat!

2. Language: conducting a first-language assessment to ascertain whether the child is working at an age-appropriate level in their own language

Your choice of first language assessment is based on two things: a) whether you have a competent first language speaker who can faithfully reflect back to you language features as the EAL child answers, and b) what assessments are available to you or your school.

First language problems: You are looking for abnormalities in pronunciation or speech production (stutters or dysfluency), inability to find words, limited vocabulary, inability to do simple math, i.e., the four basic equations, inability to give past or future tenses, difficulty with sequencing known lists (like the alphabet or numbers) and in giving sequential recounts (first this happened, then that happened).

None of these alone will point to a potential problem. A limited vocabulary might be a lack of education or a sheltered life based on the child’s gender. Lack of future or past tenses may be a language feature, as in Thai or Vietnamese. You may be asking questions like ‘what do you want to be when you grow up?’ which the child has never been asked to consider; either because the child has never been asked or because, culturally, its future is decided by others.

A number of these language features together, added to other background information, may help to contextualise the results of a first-language assessment to add to your hypothesis testing.

Next: what test is available? You may have personnel in the SEND department who can do a  simple, non-restricted assessment. Non-restricted means a normative test of ability that staff without specialist qualifications are allowed to administer. This might be something like the Test of Abstract Language Comprehension (TALC) or the British Picture Vocabulary Scale Third Edition (BPVS lll) . You may have an educational psychologist (EP) or Speech and Language Therapist (SALT) to do the testing, although, unless you have started the statutory process for getting a child an Education Health Care Plan (UK), access to these outside professionals is quite limited. You may have someone that the school buys in to do exam special access arrangements. These will not be of much use because they are not designed to measure EAL pupils (no construct validity).  Do not use these as first language assessments for EAL pupils.

To begin, it might be as simple as your bilingual interview conducted in the home language. Your assessment might be a series of short language assessment questions which ask for future and past tense information. It might be a series of pictures that ask the child/student to explain what’s going on, what might have happened before or after the picture or infer what people in the picture might be thinking or feeling? Good resources for this can be found at Twinkl or from any company that produces Speech and Language photo cards, like Speechmark .

Student N was able to give all language features normally, but took her time to experiment  through Turkish and Bulgarian for the future tense abstracted responses, using would, could and if phrases. Clearly, each language provided her with an area of communicative strength.

3. Communication Skills: observations of pupil interaction in classroom and play contexts, as well as gathering observational information from community language or religious schools as part of a multi-cultural approach;

Students should be observed during structured and unstructured times during the school day. It helps to have a club wherein EAL pupils and buddies can play non-verbal games like Connect Four or Jenga, symbol-based snap or Pelmanism, play with clay, do colouring, make pom-poms with knitting yarn / wool or just hang out and interact. Background music helps as well. Gaming on phones works, too, among EAL and native speakers, on something like Fortnight or Musically, although this is dependent on school policy. Note: the game doesn’t matter – the social push to talk about gaming where many kids are playing it and discussing it is the important part.

The point is to provide a supervised environment where students can interact in non-verbal ways. This allows you to observe eye contact, facial expression and other types of non-verbal communication skills. It shows whether an EAL student is observing others and planning how to create interaction or very shy and avoiding interaction. Children are amazing at inventing ways to communicate and responding creatively in non-verbal situations. If you have ever taken your child/children on a camping holiday and watched them effortlessly organise a football game or play amongst other non-English speaking children in a local playground, you know what I mean.

In terms of your hypothesis-testing: a child may be shy and unable to initiate or may be very observant of others, confident and socially adept at using the few words they know to create an interaction. Inability to effectively interact may be a processing difficulty; the deft ability to interact with few verbal resources, may be a sign of great intelligence. Bring your observations to the rest of your evidence to weight it up.

In structured environments, i.e., in class, watch for signs that the child is observing social cues from others in an academic environment. Does the child look at others and try to imitate what they are doing in response to a teacher’s instruction? Do they attempt to catch the teachers’ eye and brighten up to be noticed? Does the child just sit there and close down, i.e., put their pen down, close their book, become withdrawn? Again, these reactions need to be contextualised with information on the child’s natural personality, educational experience and level of trauma upon entry to the country.

Student N was very social, very outgoing and constantly trying to find ways to interact. We didn’t know that her mother had owned a beauty salon in Bulgaria and had trained her in threading. Student N was found with a long line of students waiting to have their eyebrows threaded in the cafeteria one day! Word had spread quickly about her skills!

4. Differentiation: checking that appropriate classroom provision for EAL language learning need is effectively used;

Student N sought the teachers’ attention where she could and always had our EAL team to reinforce her needs to staff. However, we found that staff were taking it upon themselves to ask her linguistic preference for translation. So, for example, her Design teacher found that instructions for Photoshop in Bulgarian worked best while her Science teachers concluded she was strongest in Turkish for complex abstract topics.

We, the EAL team, used our time with her to over-teach vocabulary and help her build a trilingual glossary of cross-curricular terms. Teachers who are willing to experiment are superlatively helpful in your hypothesis testing. With the best bilingual efforts, if a child is not making progress, that is important and key info to bring back to your hypothesis.

5. Affective Filters: investigating emotional and psychological factors affecting achievement such as past trauma, racist bullying or other environmental stresses;

Culture shock is not to be underestimated. Being separated from certain family members while jobs and housing are being set up by one parent, can be devastating. Many children are sent to live with uncles, aunts and cousins, while parents go to the new country to set up a new house and living conditions. Parents and siblings can be separated with months or years while the founding family member works to set up a new economic and household base. I have experienced children as young as 5 being left by parents with family members and not reunited until secondary school age. It takes its toll and children are often traumatised in unknown ways. Sometimes this is acknowledged by the family; sometimes it is not—‘we had to work, it had to be done, they just have to understand’.

Of course, children may develop the necessary resilience, but may not. This situation is common in immigrant families. Occasionally a special need does not become apparent until the child has been put in another context. Often, autistic children, used to a routine, suddenly become more noticeably OCD or non-interactive, whereas, in familiar surroundings, that was ‘just their personality’.

Sometimes, trauma brings its own difficulties, especially where there has been a difficult crossing into the new country: threat of separation, deportation of one parent or family member, detaining by authorities, long and difficult journeys requiring getting across disputed borders to reach transport in more neutral countries.

Children from war-torn regions may not know if remaining family are alive or safe. Unaccompanied minors are the most at-risk. I have known children from Afghanistan whose villages were saving the money to get out one male child at a time (in that culture, yes, males are the priority). Boys were given a copy of the Koran and a dictionary (a hugely prized family or village possession) and told to go to the nearest police they saw at Heathrow to turn themselves in and get taken into social care. Heart-breaking but true. I have seen it.

Those children then may not have the language to tell of their anxieties for those left behind. In these cases, art therapy will be really useful, as it helps the children express some anxieties without the stress of language. Anxiety as an affective filter can block learning in a normally cognitive child as effectively as a learning disorder.

Contextualised background is very important. Art therapists are worth their weight in gold, if you can find one. Otherwise, talk to the art department and see if you can get the child used to a friendly art teacher. It can work wonders in a short time to be given a non-verbal expressive outlet that does not have an exhausting amount of new language involved.

Children will not learn or make any progress until some of that anxiety is relieved. Note: Once the child starts expressing past experiences in drawing, you may need to take the pictures to the Child Protection officer or school senior leadership for further advice or action, if very concerning. Steel yourself and do what is needed according to your school policy.

Student N missed family but had enough cousins in London to help her through. She also went back regularly (something not all students could do) and spoke to her parents regularly. She was nominally homesick, but not enough to affect her processing.

6. Testing: considering raw data on national standardised tests but as a minimal part of the whole picture.

This is a sore spot because schools are expected to be data-driven but have few sources of decent data that apply to EAL children. You may need to fight for the EAL child to be contextualised as opposed to judged by numbers.

CATS: in the UK, the quantitative and non-verbal portions of the CATS are seen as appropriate judges of cognitive ability. They are not. I have sat with a gifted and talented Russian pupil while she started the non-verbal and noted the instructions are still in English. Nor are the examples particularly easy to follow for a student who is not used to on-line standardised tests.

These results with not be worth the 3 hours it takes to do the test UNTIL the child has developed some English. Lesser known is the fact that CATS have a test per year group. It is possible NOT to test in the year the child enters school and to test a year later using the year-appropriate version of the CATS. School management and the data team need to be primed for this. 

Most reading comprehension and ability tests will not show anything other than the child has little English.

If you decided to take an MA at the University of Sweden in Stockholm and took a Swedish IQ test, chances are you would come out as severely learning disabled. This is because you would not have the basic 5,000-word vocabulary of a 5-year-old Swedish child. National standardised tests of ability are useless on EAL children when they first enter the country and for the first year, at least, of their acquisition.

We fought for Student N not to be tested on CATS. Instead we cobbled together a general cognitive profile based on our teaching of her and the background. Using that and feedback from her teachers, we surmised that she was of normal cognitive ability.

Summary: As it turned out, we decided Student N did not have a cognitive processing difficulty. She was simply processing so much language between Turkish, Bulgarian and English that she was slower than the other students. Eventually, she got faster. Currently, Student N is at a university in this country studying business and happy as a clam! Had we labelled her as SEN she may have been limited by our low expectations.

Hypothesis-testing is a necessary and powerful method, but takes time and coordination. Remember to work through the 6 basic variables and contextualise all your data. When in doubt, contact me for help.

Next Case Study: Student E and the mystery of semi-lingualism.

Frederickson, N. and Cline, T. (1991) Bilingual Pupils and the National Curriculum: Overcoming Difficulties in Teaching and Learning. London: University College

Frederickson, N. and Cline, T. (1995) Assessing the Learning Environments of Children with Special Educational Needs: Report of a Workshop. London: Pedagogy Publishing

Frederickson, N. and Cline, T. (2015) Special Educational Needs, Inclusion and Diversity. Third Edition. New York/Open University Press McGraw-Hill

Hall, D. (2001) Assessing the Needs of Bilingual Pupils: Living in Two Languages. London: David Fulton

Pim, Chris (2010) How to Support Children Learning English as an Additional Language. Hyde/LDA-Findel Publishing

DfES (2001) Special Educational Needs Code of Practice. London: DfES

DfE (2014) SEND Code of Practice 0-25. London: DfE