Research highlight

By Natalie Smith

The literacy skills that are acquired in childhood are the foundation on which future academic success is built, so it is important that all children become successful readers. Research has provided a wealth of information about the development of skilled monolingual readers, but this evidence base is lacking when it comes to children learning EAL, particularly in a UK educational context.

Research has suggested that children learning EAL often fall behind their monolingual peers on measures of listening and reading comprehension (McKendry & Murphy 2011), though research within the UK is not extensive. Hutchinson et al. (2003) report that children learning EAL in the UK perform, on average, one year behind their monolingual peers on measures of reading comprehension. This apparent pattern of relative underachievement perhaps suggests that the literacy needs of children learning EAL are not fully understood and/or are not being met.

My research, therefore, aims to understand which specific skills underpin reading comprehension for both monolingual and EAL children. In my longitudinal field work I use standardised and bespoke measures to explore a range of linguistic and cognitive skills important to reading comprehension. In the first and third year this will include measures of vocabulary, grammatical knowledge, working memory, phonological awareness, listening comprehension and speed of lexical access (how quickly words can be brought to mind). I am working with children in Y2 and Y4 and will follow the development of all participants into Y4 and Y6. My aim is to develop an evidence base that will allow insight into the differences and similarities between English-speaking monolingual children and children who use English as an additional language in two different year groups in the UK.

One of the key issues concerning this line of research, however, is with the term ‘EAL’. The term conflates a very wide range of young people, making it difficult to apply findings from one study to the wider population. For example, there are huge variations in factors such as first-language (L1) proficiency, socio-economic status, the writing system of the L1, parents’ English-language proficiency, the L1 of peers within a school, exposure to linguistic and literacy practices in the home. The list goes on. Irrespective of these differences, these children are all categorised as ‘EAL’. This raises further questions about how to approach EAL research. Should research focus on learners of a specific SES or from just one language heritage so that findings can then be applied to this specific group of learners? Or should individual studies try to incorporate a range of learners with a range of languages to reflect the multilingualism of many classrooms?

To echo the point that many researchers are making, we need to push for research in this field and we need to collaborate with each other to develop a holistic and coherent understanding of our EAL children. I would be very interested in your experiences and thoughts!

*** The full-length blog post was originally posted on the EAL Journal Blog ***

By Lydia Gunning

Funded by a Leeds Academic Research Scholarship (LARS) and the Bradford Institute of Health Research (BIHR), my project works closely with schools associated with the Born in Bradford cohort (BiB) (https://borninbradford.nhs.uk/). This group is particularly interesting for linguistic study due to both the high numbers of bilinguals in the community and the notably low SES within the city as a whole. However, despite the interesting make-up of this group, and the substantial effects these two variables have been shown to have on language development, there has yet to be any research conducted looking at the language profile of this cohort. This, in part, is likely due to limited resources available that can comprehensively assess language development. Many language assessments currently in distribution either do not allow for a complete assessment of language (focussing only on a specific area), are not standardised in the UK, do not allow for the assessment of primary school aged children, or do not give standardised scores for monolinguals and bilinguals alike. Given that approximately a quarter of children in any particular class in the UK identify as being bilingual, a tool that could quickly screen for any language difficulties, while taking into consideration the number of languages a child might speak, would be highly valuable.

Thus, to sufficiently assess the language development of BiB children, or children similar to those in the BiB cohort, my research aims to develop a fully comprehensive language battery that can overcome the barriers mentioned above. Consequently, the battery can be used to screen for atypical language development in both monolingual and bilingual populations, but will also help further distinguish differences between abilities of these two groups. Once finalised, the battery will allow assessment of core language skills, such as receptive language (the ability to understand language) and expressive language (the ability to produce language), as well as phonological awareness (the ability to manipulate language). It will also identify any deficits in pragmatics; the social use of language.

Measuring a bilingual’s language competence becomes increasingly difficult due to the abundance of research suggesting that the amount, type and context of exposure to language can all affect proficiency. Consequently, exposure to languages must be taken into consideration when studying bilinguals and this is especially true if bilinguals are to be studied alongside their monolingual peers. Although there are various language exposure questionnaires already in distribution that might allow measurement of the extent to which language ability relies on linguistic exposure, many of these rely on parental report, which often has feasibility issues within research. A secondary aim of the project is thus to develop a way of quantifying a bilingual’s language experience without relying on the, often elusive, parental report. For this reason, we developed the Leeds Bilingual Exposure Questionnaires (LeBEQ). Comprising of two separate questionnaires, one to be completed by the child themselves, the other to be completed by their teacher, the LeBEQ attempts to determine the extent of a child’s bilingualism, based on exposure to the languages in question. Factors included were those that research has repeatedly shown to affect a bilingual’s language development, such as length of exposure, cumulative exposure, contexts of use, birth order and language used in various activities (reading, playing games etc.).

Though the pragmatic components of the battery are still being developed, and the data for the LeBEQ and core language components is still being collected and analysed, my research eventually attempts to shed light on questions such as: (1) precisely how, and to what extent, linguistic experience affects language development, (2) how this might be complicated by an overlying layer of poor SES and (3) the effect these variables have on a child’s resultant academic attainment.

By Marta Wesierska

Many of the children in UK primary schools are identified as EAL learners. Coming from various language backgrounds (many different first languages – “L1s”) and with varied ability levels of English (their second language – “L2”) the language and literacy development of these children is likely to vary compared to their monolingual native speaking English peers.

Furthermore, assessment of this group of children usually focuses on their abilities in English, overlooking their performance in their native tongue. This is partly due to a lack of assessment measures in languages other than English and researchers with the necessary language abilities to administer them (although efforts are underway to create such tools, for example the recent Receptive Vocabulary Assessment App).

In my research, I have aimed to broaden the understanding of the performance of EAL children within the areas of pre-literacy, oral language, and reading from the perspective of their first and second languages. To achieve this, I have been working with a group of Polish EAL pupils and two control groups of their monolingual Polish and English peers.

The findings of my PhD research have identified areas of strengths and weaknesses in this sample of children in both languages. One of the most striking findings is the EAL group’s substantial weaknesses in oral language in both English (L2) and Polish (L1). This is particularly relevant due to the impact of oral language skills on reading comprehension in later grades. Therefore, one of the implications for teachers of EAL pupils is continued attention to L2 oral language to close this performance gap between them and their monolingual peers and to improve subsequent reading comprehension in the L2. First language is still relevant however: one of my studies has shown L1 phonics to be a significant predictor of both L1 and L2 decoding and L1 decoding to significantly predict L2 reading comprehension in the EAL sample. Therefore, L1 phonics intervention could potentially be used to support reading comprehension in L2, in this population of children. To sum up it is important that both teachers and practitioners are aware of a child’s EAL profile and abilities in all languages spoken in order to reduce the risk of misidentification and inform instruction.

by Emily Oxley

Emily has recently completed a PhD investigating vocabulary acquisition in children with more than one language. She conducted a systematic review of word learning interventions for children with English as an additional language, and carried out an experimental word learning study comparing EALs and children who only speak English.

What is a systematic review?

A systematic review is a very thorough way of searching the current literature without bias. You decide upon search terms and enter these into databases of research papers. From there you end up with thousands of results that you filter according to an inclusion and exclusion criteria. This gives you a much more manageable amount of papers to read in depth and include in your review.

How many papers did your review include?

After screening my results, I was left with a total of 18 studies to review. 17 of these studies were conducted in the USA and one in the UK. All studies included children with EAL from a variety of language backgrounds. The most common language spoken was Spanish.

What were the results of the studies?

The results of the studies suggested that explicit vocabulary training in context can produce word learning gains for EALs. When new words are explicitly taught within a curriculum, EALs can learn at the same rate as their peers who only speak English. Storybook reading alongside adult led questioning and discussion can lead to vocabulary growth with younger children. However explicit vocabulary teaching in this review showed larger comparative vocabulary gains than storybook related interventions. Two studies showed no intervention effects across measures; implicit acquisition through television viewing and sign supported English instruction.

What are the conclusions to the study?

The study provided more evidence that children with English as an additional language start school with a lower English vocabulary than their monolingual peers. Early interventions are recommended in first years of schooling for EAL children so that they will not fall further behind. Most interventions in this review have shown to reduce the vocabulary deficit.

Were there any limitations?

Yes- because most interventions in this review were carried out in the USA, it may not be possible to replicate their techniques in the context of the United Kingdom. Due to the homogenous sample of first languages in the USA, many of the interventions were able to use the first language to help bridge the gap. Language backgrounds of children in the United Kingdom are diverse so this causes additional constraints. In addition, it is not generally possible to know a child’s underlying L1 vocabulary in the UK, whereas in the USA there are standardised Spanish first language vocabulary measures which can be used.

What are the future directions?

This review has shown that we are really in need of randomised control trials for school aged EAL children in the United Kingdom. There is so much still to learn about children with English as an additional language, and although research in this field is growing in the UK, it is still in its relative infancy compared to the USA

THE PROJECT

“Referential communication and executive function skills in bilingual children” is a research project that was funded by the Leverhulme Trust between September 2012 and May 2015.  The team comprised Cecile De Cat (principal investigator), Ludovica Serratrice (co-investigator), Sanne Berends and Furzana Shah (research assistants).

The aims of the project were

1. to extend previous findings on the relationship between key executive function skills (cognitive flexibility, inhibitory control and working memory) and language experience to bilingual children who have unbalanced exposure to two languages;
2. to gather new information on the role played by language proficiency, bilingual experience and SES on the above subset of executive function skills and on referential abilities;
3. to develop our understanding of the linguistic and non-linguistic contextual variables affecting children’s referential choices (visual context, awareness of differences in perspective between speaker and listener, and linguistic factors affecting a referent’s prominence).

We targeted a highly heterogeneous group of children, in terms of socio-economic status and bilingual experience.  There was a total of 28 home languages in our sample: Punjabi (21%), Urdu (17%), Arabic (9%), Spanish (6%), French (8%), Bengali, Cantonese, Catalan, Dutch, Farsi, Greek, Hindi, Italian, Kurdish, Mandarin, Marathi, Mirpuri, Nepalese, Pashto, Polish, Portuguese, Shona, Somalian, Swedish, Tamil, Telugu, Thai, Tigrinya (each of the latter representing less than 5% of the data).

There was a total of 174 children in the final sample (including 87 monolinguals).  The children were between 5 and 7 years of age at the time of testing.  The amount of bilingual experience varied substantially across children, ranging from very little experience in the home language (in the “almost monolingual” children) to clear dominance of the home language (in children for whom English was the weaker language).

MAIN FINDINGS

1. A bilingual advantage in cognitive (executive function) skills was only found in the inhibition task.  The main child-related predictors of performance were age, socio-economic status, self-monitoring, and amount of home language experience. Using a novel method of analysis, we identified a critical threshold of home language experience for the bilingual advantage (correcting for age, socio-economic status and self-monitoring). Most of the children above that threshold came from households in which both parents spoke the home language with their children all the time. See De Cat, C., Gusnanto, A., & Serratrice, L. (2017). Identifying a threshold for the executive function advantage in bilingual children. Studies in Second Language Acquisition, 1-33. https://doi.org/10.1017/S0272263116000486 or for open access: Paper available here.  This paper received the Albert Valdman award for outstanding publication in Studies in Second Language Acquisition for the year 2018.
2. The language proficiency of bilingual children varies considerably (developmentally and across individuals). We measured different aspects of EAL children’s proficiency in English (focusing on interpreting and repeating complex sentences, and on aspects of vocabulary knowledge). Our analyses show that children’s performance in English proficiency tests is influenced by the amount of experience of English they have had over their lifetime. Using a new method of analysis, we found that children between the ages of 5 and 7 who had been exposed to English for 60% of the time over their lifetime had “caught up” with their monolingual peers. Passivity in the home language does not have a “protective” effect on school language proficiency. Children’s environment (as reflected by the socio-economic profile of their family) also has an impact. See https://osf.io/f5q98/ for the full paper, or the summary of our findings on OASIS. 
3. Children’s ability to communicate information effectively is influenced by the same factors in bilingual and monolingual children. In 5- to 7-year olds, the ability to take the perspective of the listener into account depends on whether the relevant information is presented visually or verbally. At that age, children are generally not able to adapt to the fact that their interlocutor is not able to see what they can see, but those with better inhibition skills are able to take into account the information previously shared verbally with their interlocutor.  English proficiency was also a strong predictor of performance in our tasks, which put some bilingual children at a disadvantage. But bilingual children were as informative as monolinguals when proficiency was controlled for. See Serratrice, L. and De Cat, C. (2019). Individual Differences in the Production of Referential Expressions: The Effect of Language Proficiency, Language exposure and Executive Function in Bilingual and Monolingual Children. Bilingualism: Language and Cognition.