BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Predictors of listening comprehension skills in bilingual children
  - Prof Ludovica Serratrice (University of Reading)
DTSTART:20190228T170000Z
DTEND:20190228T183000Z
UID:TALK120148@talks.cam.ac.uk
CONTACT:Julia Heine
DESCRIPTION:\nLike many other English-speaking countries\, the UK is becom
 ing an ever more linguistically diverse society. In England alone in Janua
 ry 2018 21.2% of children entering primary school were classified as havin
 g English as an Additional Language (EAL) and are bilingual to some extent
 . For these young bilinguals English is the only medium of instruction\, a
 nd it is the language in which their language and literacy outcomes are fo
 rmally assessed. Evidence that English proficiency is a strong predictor o
 f academic success in this population is now increasingly being recognised
 . It is therefore imperative to know more about what\, in turn\, predicts 
 bilingual children’s language and literacy proficiency\, not only in ter
 ms of concurrent relationships at a given time point\, but from a longitud
 inal perspective. In this talk will present the findings of a longitudinal
  study on the contribution of memory\, attention\, lexical and grammatical
  skills\, and non-verbal inferencing to bilingual learners’ ability to m
 ake local and global inferences from oral texts.\n\nMethods: 94 EAL learne
 rs with 26 different home languages were recruited in English schools (mea
 n age 5\;8). Children were assessed twice over the school year. At T1 we a
 ssessed their listening comprehension\, memory skills\, vocabulary breadth
  and depth\, morpho-syntax\, comprehension monitoring and non-verbal infer
 encing skills. Cumulative amount of English input was measured through a p
 arental questionnaire at T1. A similar testing procedure was followed at T
 2. The listening comprehension task included three short stories with two 
 literal questions\, three questions on local inferences and three on globa
 l inferences.\n\nResults: At both time points accuracy was significantly g
 reater for text-connecting local inferences than for global inferences req
 uiring reliance on world knowledge. \nGlobal and local inferences accuracy
  at T2 was predicted by vocabulary depth and morpho-syntax at T1. When lis
 tening comprehension at T1 was entered as a mediator in the model\, morpho
 -syntax at T1 still directly predicted a significant amount of variance in
  listening comprehension at T2\, as well as an indirect effect through lis
 tening comprehension at T1. English input did not directly predict listeni
 ng comprehension\, at either times\, but explained a significant amount of
  variance in vocabulary depth and morpho-syntax.\n\nConclusions: Bilingual
  children find it easier to solve local than global inferences\, as has be
 en previously found for monolinguals. The results further confirm the impo
 rtance of both vocabulary and grammar for listening comprehension\, and su
 ggest that vocabulary depth might be a better predictor of inferencing tha
 n breadth. Even when controlling for listening comprehension at T1\, morph
 o-syntax continued to predict listening comprehension six months later\, w
 hile cumulative English input only influenced comprehension indirectly via
  vocabulary and morpho-syntax.\n
LOCATION:GR06/07\, Faculty of English\, 9 West Rd (Sidgwick Site)
END:VEVENT
END:VCALENDAR