[lg policy] Linguist List Issue: Issues in the Assessment of Bilinguals

[The LINGUIST List]

HFS thought you might be interested in this item from the LINGUIST List
----------------------------------------------------------------------------------------
HFS says ...


----------------------------------------------------------------------------------------
 
Message1: Issues in the Assessment of Bilinguals
Date:09-Feb-2014
From:Liubov Baladzhaeva baladjaeva at gmail.com
LINGUIST List issue http://linguistlist.org/issues/25/25-2391.html 


Announced at http://linguistlist.org/issues/24/24-3923.html 

SUMMARY

This book is the first of the two edited volumes dedicated to uncovering and solving issues in the testing of bilinguals. The contributors to the volumes are committed to demonstrating how bilinguals are different from monolinguals and how this difference is not a deficiency. The current standardized tests used in assessment of both children and adults are, according to the researchers, not adequate for bilinguals as they have been normed on monolinguals. Most existing tests (mainly those that test language development and those used for diagnosing language impairments) use monolinguals as a standard, while normal bilingual development can be deemed abnormal by these tests. This not only hurts bilinguals by classifying them as "language-impaired", but also prevents specialists from diagnosing real language impairments in bilinguals when they are present. The book is both a study and a guidance manual for educators and language specialists that asks them to value and celebrat!
 e language diversity and use it in order to facilitate academic and social success of bilinguals.

Chapter 1: Virginia C. Mueller Gathercole. Assessment of Multi-tasking Wonders: Music, Olympics and Language. 

The first chapter outlines the goals and the composition of this edited volume and provides a short summary of issues raised in the volume and chapters that address them. The main goal of the first volume is to demonstrate why current standardized tests normed on monolinguals are not adequate for testing bilinguals and what might be realistic expectations for the linguistic development of bilinguals. The second volume will offer solutions to the issues raised in the first volume. 

The main reason for the inadequacy of existing tests is that bilingual language development is different from monolingual development in both languages. Currently most school children around the world are expected to achieve a certain level of language proficiency in an official language in a certain amount of time, and failing to do so is considered a problem. However, the norms for these levels and times are based strictly on monolingual development, not taking into account bi- or multi-lingual children. Content areas like math or geography are also built on expectations of monolingual children to achieve certain language mastery. This might prevent bilingual children from excelling in these areas even when they know the material because of insufficient command of the language of instruction. Language deficiencies and impairments are taken very seriously by both parents and educators, and during the school years language acquisition is one of the main means of measuring a !
 child's academic progress. The chapter describes how multilingualism is often underestimated in modern nation-states, and how being multilingual sometimes is seen as an impairment in itself and not an accomplishment. Multilinguals are compared by the author to talented musicians playing many instruments or athletes excelling in different disciplines.

Chapter 2: Virginia C. Mueller Gathercole, Enlli M�n Thomas, Emily J. Roberts, Catrin O. Hughes and Emma K. Hughes. Why Assessment Needs to Take Exposure into Account: Vocabulary and Grammatical Abilities in Bilingual Children.

This chapter raises the issue of norming used in standardized tests. Current tests take a monolingual child as the norm for assessment of language development; however, a monolingual child has much more exposure to her only language than a bilingual has to either of her two languages, and that might lead to later development of some exposure-dependent features in either or both languages. Also, testing a bilingual only in one of her languages might lead to underestimation of the rich and complex language system that a child has. The study described in this chapter investigates the influence of exposure on test performance in Welsh-English bilingual children. It compares monolingual English children and English-Welsh and Welsh-English bilinguals using receptive vocabulary tests. The study demonstrates that in younger children the amount of exposure to a given language conditions the development in this language. The level attained in one language does not necessarily reflect !
 the child's development in other language; therefore, testing young bilinguals just in one language will tell us very little about their overall linguistic development. However, in adolescents who over time gain enough exposure in both their languages, the test results become closer to those of monolinguals and similar performance in both languages can be observed. Another issue the researchers look at is cross-linguistic influence and carryover of grammatical structures. This was tested by using forced-choice picture tasks. Overall, no carryover on the linguistic level was found. 

The authors advocate for a new model of measuring the performance of bilingual children. In such a model there should be two standards of comparison developed: a child should be measured relative to all children in her age group and to children from a similar linguistic background. Two scores obtained using these comparisons would give a comprehensive picture of the bilingual child's language development.

Chapter 3: Shula Chiat, Sharon Armon-Lotem, Theodoros Marinis, Kamila Poli?ensk�, Penny Roy and Belinda Seeff-Gabriel. Assessment of Language Abilities in Sequential Bilingual Children: The Potential of Sentence Imitation Tasks.

The issue raised in this chapter is that testers and educators often do not know a bilingual child's L1 and therefore cannot conduct tests in it. The authors develop a language-neutral test in the form of sentence repetition in the L2 (meaning the official language of instruction). They review evidence that in monolingual children sentence repetition can indicate language impairments. The advantage of using such a task for the testing of bilinguals is that it is relatively less affected by language exposure and experience than other testing measures. Four groups of bilingual children in three different countries were examined: Russian-Hebrew and English-Hebrew in Israel, Russian-German in Germany and Turkish-English in the UK. When compared to the monolingual children, most of the bilinguals performed within the monolingual ranges. However, children whose age of L2 onset was after 2 years old fell below the monolingual range. Socioeconomic background also played an important!
  role in the child's success. One of the limitations of this task is that great differences in socioeconomic background can lead to different results in children with similar linguistic backgrounds. The conclusion was that sentence-repetition tasks with real words are a valid means of measuring L2 proficiency in bilinguals; however it is not yet clear whether they can be used for detecting language impairments, since it first requires norming on bilinguals and this norming should differ according not only to language, but also to socioeconomic background. 

Chapter 4: Netta Abugov and Dorit Ravid. Assessing Yiddish Plurals in Acquisition: Impacts of Bilingualism.

This chapter takes on an understudied issue; namely how a child can be tested in a language that is still developing dynamically. The language in question is modern Hasidic Yiddish spoken in Israel. Due to the contact with Hebrew the language is rapidly undergoing grammatical and lexical changes and sometimes it is not clear what the standard usage is. The authors collected data on plural forms of nouns in adults and then tested children in order to see which forms (in cases where there is more than one form acceptable) they would choose. Children mostly chose the forms with the highest frequency, but they also tended to overgeneralize or adopt some rules of pluralization from Hebrew. This finding is also relevant for testing in a language with less dynamic on-going change. The tests might be based on the older or more formal norm, while children would be sensitive to the new norm, acquiring and producing it. This study also has important implications for assessing children !
 from communities where a non-standard dialect is spoken. Often such children are penalized for using non-standard forms in school settings while their choice only reflects the normal language acquisition of the community language.

Chapter 5: Roc�o P�rez-Tattam, Virginia C. Mueller Gathercole, Feryal Yavas and Hans Stadthagen-Gonz�lez. Measuring Grammatical Knowledge and Abilities in Bilinguals: Implications for Assessment and Testing.

This chapter examines educated bilingual adults: simultaneous or early Spanish-English bilinguals in Miami. Three groups of bilinguals were examined from different home environments: both Spanish and English at home, only Spanish, and early Spanish with addition of English spoken at home. While some tests have shown that bilinguals over time become more similar in performance to monolinguals, even simultaneous adult bilinguals might still differ in some aspects of language knowledge and/or performance. This investigation used two forced-choice picture tasks for testing abilities in English and Spanish. The study showed that all bilinguals, no matter their home linguistic environment, performed in the monolingual range on English tests. That is, by the time they reach adulthood, early and simultaneous bilinguals are able to acquire the official language as well as monolinguals. There were, though, significant differences among them in Spanish: those with less exposure to Span!
 ish performed less well on the tests. Most of the bilinguals also differed from the monolingual Spanish speakers in their performance, even those with the greatest exposure to Spanish. This study, like the one in Chapter 2, found that bilinguals who perform well in one language tend to also perform well in another language. The results of the study confirm that being bilingual is not a disadvantage and does not hinder the acquisition of the official language. 

Chapter 6: Miguel �. P�rez, Cristina Izura, Hans Stadthagen-Gonz�lez and Javier Mar�n. Assessment of Bilinguals' Performance in Lexical Tasks Using Reaction Times.

In this chapter the issue of the sophistication of current tests is raised. The authors argue that tests that only check the accuracy of responses might miss some important aspects of lexical knowledge. They propose using reaction times for testing bilinguals in such tasks as picture naming, visual lexical decisions and word categorizations. This would, for example, allow accessing the difficulty of specific words for an individual, since more complex mental computations require more processing time. Currently reaction times are almost nonexistent in standard testing procedures accessing linguistic knowledge and development. However, given modern technology, tasks using reaction times can be easy to construct and implement. In order to demonstrate the possibility of using reaction times, the researchers conducted an empirical study that was able to assess the effect of the order of acquisition on word processing. The researchers trained monolingual Spanish speakers on new We!
 lsh words in order to test whether order of acquisition influences processing time. The words taught were matched in frequency and predicted phonetic and lexical difficulty for the participants. Early learned words took less time to process than late-learned words, while there was no significant difference in accuracy.

Chapter 7: Rebecca Burns. Assessment and Instruction in Multilingual Classrooms.

This chapter offers practical instruction on how teachers can welcome multilingualism in their classrooms and use other languages in instruction even without actually knowing them. One of the main issues in the education of bilinguals is that often their L1 is not used in formal education in any way, when using it could improve their overall academic success and their L2 acquisition in particular. Public display of home languages at school can also show appreciation for diversity, affirming students' bilingual identity and raising their social status. Considering modern migration patterns, one classroom can contain speakers of many languages, and in many cases it is not possible to provide an educator who speaks every child's L1. However, this chapter shows that not knowing children's L1s should not be an obstacle for using them in the classroom. The chapter describes what kind of resources, especially ones found on the Internet, can be used. For example, the teacher can use!
  posters and other print materials in the home languages. She can employ online dictionaries in order to ensure that the learners understand specific words. Multilingual materials can be used for assignments. The chapter lists specific materials and resources for some languages that might be encountered in US classrooms. One of the issues is that in some US states using languages other than English in classrooms is discouraged or even prohibited. The author gives advice on a way around such prohibitions, for example, sending multilingual materials home with a student. 

However, the author cautions that using some elements of child's L1 in the classroom do not substitute fully for bilingual education.

The chapter's main target group is US teachers educating L2 English learners. The author states that one of the goals is to raise teachers' language awareness..

Chapter 8: Jasone Cenoz, Eli Arozena and Durk Gorter. Assessing Multilingual Students' Writing Skills in Basque, Spanish and English.

This chapter raises the issue of the effect of home language and language of instruction on bi- and multilingualism and compares L1 Basque and L1 Spanish high school students in the Basque country with Basque as the language of instruction. Spanish and English are taught as foreign languages in their school. The authors compare the academic performance of L1 Basque and L1 Spanish speakers in all three settings, assessing their writing skills in Spanish, Basque and English. For these two groups the level of exposure to Spanish is relatively similar, since Spanish is the main language of the community outside school; however their exposure to Basque differs significantly, as Spanish speakers do not encounter much Basque outside school. The participants wrote essays in each of three languages describing pictures that were provided by researchers. The essays were then scored for content, organization, vocabulary, language use and mechanics. While both groups were similar in thei!
 r scores for essays in Spanish, L1 Basque students outperformed L1 Spanish students on both Basque and English essays. While better results on Basque are expected for the L1 Basque group, as they had more exposure to the language than L1 Spanish students, it is not clear why they perform better in English as well. The study demonstrates that even in the case of multilinguals in the same school settings different linguistic development patterns can be found, influenced by many factors.

Chapter 9: Stephen J. Caldas. Assessment of Academic Performance: The Impact of No Child Left Behind Policies on Bilingual Education: A Ten Year Retrospective.

This chapter addresses the impact of the "No Child Left Behind" (NCLB) policy on the academic performance of US children. The policy charges schools with the responsibility for raising the academic performance of underachieving student groups, including children with home languages other than English (called English Language Learners -- ELLs). The author provides a history of language education policies in the US and compares test results in Math and Reading from before and after the NCLB policy was instated in 2001. According to the author, the policy strongly encourages acquisition of English without assigning importance to child's L1. The main focus of the study is Spanish-English bilinguals, since they are the majority of ELLs in US schools. The study found that NCBL actually negatively affected this population with the gap between them and monolinguals increasing after the policy's adoption. The issue of defining English learners and how it affects testing policies is a!
 lso raised. Currently ELLs are defined as such when they underperform on the standard tests. Thus by definition ELLs are never able to achieve the same results as the main population. This prevents educators from seeing the real results of all children with home languages other than English, as those of them who perform well on the tests are not counted as ELLs. Another issue is that under the NCLB policy schools are encouraged to produce good test results and they train their students to pass tests instead of teaching the material, which ultimately lowers the quality of education. The author suggests that language policy in school should be based less on politics and more on empirical research, since the data demonstrate how the current policies may not be beneficial either for the children or for the schools.

Chapter 10: Virginia C. Mueller Gathercole. Summary of Issues Surrounding the Assessment of Bilinguals and the Way Forward to Solutions.

The last chapter once again summarizes the issues raised in the volume. The author concludes that current test measures used in academic settings should be revised as they are not adequate for testing bilinguals and multilinguals. 

EVALUATION

This edited volume raises an important issue of testing bilinguals in educational settings. The issue is not only theoretical, as the tests in their current form may both hurt bilingual students' academic opportunities and prevent specialists from noticing language impairments in bilingual children. The volume is of great value for researchers, educators, and education policy makers. It should definitely be used in teachers' education programs around the world to raise their awareness of multilingualism. While all the studies in this volume are interesting and valuable, on the whole, there is an issue of coherence. The logic of transition from chapter to chapter is not clear, and two chapters seem a bit too specific and out of place: Chapter 7 that provides a practical guide to US teachers and Chapter 9 that reviews the No Child Left Behind policy. First, these two chapters, unlike the others in this volume, do not present empirical studies. Second, they both focus strictly !
 on US education, while the other chapters offer a more general approach. Since the volume does not address a US audience exclusively, it is not clear why such specific chapters should be included with no attempt to generalize from them. Finally, the chapter addressing teachers would make a better contribution to the second volume that offers solutions to the problems of educating bilingual students. Organizing the studies in several thematic sections prefaced by theoretical introductions might have improved the coherence of the volume. 

Liubov Baladzhaeva is a PhD student at the University of Haifa. She is interested in multilingualism, language acquisition and attrition. 


Also you can take a look at it by visiting 
http://linguistlist.org/issues/25/25-2391.html

Read other LINGUIST List posts:
http://linguistlist.org/issues/index.cfm

Get your own free subscription to The LINGUIST List:
http://linguistlist.org/LL/subs-index.cfm



_______________________________________________
This message came to you by way of the lgpolicy-list mailing list
lgpolicy-list at groups.sas.upenn.edu
To manage your subscription unsubscribe, or arrange digest format: https://groups.sas.upenn.edu/mailman/listinfo/lgpolicy-list