30.161, Review: Applied Linguistics; Computational Linguistics; Language Acquisition; Text/Corpus Linguistics: Möller (2017)
The LINGUIST List
linguist at listserv.linguistlist.org
Fri Jan 11 20:26:55 UTC 2019
LINGUIST List: Vol-30-161. Fri Jan 11 2019. ISSN: 1069 - 4875.
Subject: 30.161, Review: Applied Linguistics; Computational Linguistics; Language Acquisition; Text/Corpus Linguistics: Möller (2017)
Moderator: linguist at linguistlist.org (Malgorzata E. Cavar)
Reviews: reviews at linguistlist.org (Helen Aristar-Dry, Robert Coté)
Homepage: https://linguistlist.org
Please support the LL editors and operation with a donation at:
https://funddrive.linguistlist.org/donate/
Editor for this issue: Jeremy Coburn <jecoburn at linguistlist.org>
================================================================
Date: Fri, 11 Jan 2019 15:26:41
From: Nausica Marcos Miguel [marcosn at denison.edu]
Subject: Language Acquisition in CLIL and Non-CLIL Settings
Discuss this message:
http://linguistlist.org/pubs/reviews/get-review.cfm?subid=36412557
Book announced at http://linguistlist.org/issues/29/29-379.html
AUTHOR: Verena Möller
TITLE: Language Acquisition in CLIL and Non-CLIL Settings
SUBTITLE: Learner corpus and experimental evidence on passive constructions
SERIES TITLE: Studies in Corpus Linguistics 80
PUBLISHER: John Benjamins
YEAR: 2017
REVIEWER: Nausica Marcos Miguel, Denison University
SUMMARY
This monograph presents a comprehensive analysis of the knowledge of passive
constructions by 16-17 year-old L1 German speakers who are learning English as
an L2 at the Gymnasium level (i.e., high school leading to university
studies). Given the positive results that have been attributed to Content and
Integrated Language Learning (CLIL) instruction, the question this study
addresses is what factors contribute to its success. Therefore, learners
following either the CLIL approach in their classes or only an English as a
Foreign Language Approach (EFL) are compared. This analysis brings together
different factors that can influence the success of learners in CLIL programs
such as the program’s selectivity of participants, cognitive and social
learner individual differences, and the teaching materials utilized. Results
showed that learners in CLIL programs demonstrate greater knowledge of passive
constructions than learners in EFL programs. Möller posits that the
selectivity of CLIL, individual learner variables, and this method of
instruction contribute to these results.
The results of this study are relevant for researchers, language
practitioners, and materials developers. Specifically, this book will be of
interest to researchers working on ways to combine SLA (Second Language
Acquisition) and Learner Corpus Research (LCR) methodologies. Moreover, this
study links research on teaching materials to SLA. For researchers planning to
create their own corpus, this book is also an excellent resource since it
discusses in depth how two corpora – an L2 learner corpus (ScooLE) and a
textbook corpus (TeaMC) - were created, normalized, and analyzed.
Chapter 1 introduces the methodology and goals of this study. By using
methodologies from SLA and LCR, this monograph provides ample data to explore
CLIL success. Passive constructions are analyzed in the Secondary-Level Corpus
of Learner English (SCooLE) created for this project. This corpus consists of
essays written by students in one of the following categories: CLIL+, CLIL0,
and CLIL-. CLIL+ refers to learners who received CLIL instruction (n=158);
CLIL0 refers to learners to whom a CLIL option was not available (n=182); and
CLIL- refers to “learners who deliberately did not take part in a CLIL option
or dropped out” (n=56). The learners who contributed to the corpus also took a
test about knowledge of passive constructions, a background questionnaire, a
psycholinguistics test, and a motivation questionnaire.
Chapter 2 situates CLIL in the German context, and, more precisely, in the
Baden-Württemberg school system by providing the reader information about that
instructional context as well as by summarizing previous studies of CLIL
versus non-CLIL learning outcomes carried out in Germany and Austria with
German L1 speakers. These studies have examined language awareness and
proficiency in the four skills (writing, reading, speaking, and listening) as
well as grammatical, lexical, and general L2 competence. Overall, CLIL
learners outperform non-CLIL learners. However, Möller points out that the
differences between learners cannot be conclusively attributed to CLIL. She
draws this conclusion because the selectivity of CLIC, learner linguistic
background, socio-economic variables, and cognitive variables have been
analyzed, but have not been taken into consideration conjointly in the
methodological design.
Chapter 3 discusses at length the structural and lexicogrammatical properties
of passive constructions. Structural properties include the way passive
constructions combine with modal auxiliaries and catenative verbs, the use of
different tenses, agenthood, and the type and number of verbs that accept the
passive. Meanwhile, lexicogrammatical properties include the “varying degrees
of adjectivalness” of passive structures (e.g., central and non-central
passives). Frequency effects in the use of passive because of textual genre
and individual differences are also reviewed. This chapter concludes with a
discussion of the research on acquisition of passive constructions among
English L2 learners.
Chapter 4 explains in detail what kind of data is examined in the study and
how it was collected. This chapter is key to understanding the rest of the
monograph. First, it explains how the SCooLE corpus (258,586 tokens) was
compiled from argumentative essays by CLIL+, CLIL- and CLIL0 learners. The
chapter carefully describes how the corpus was normalized (automatically and
manually), how errors were dealt with, and how it was POS-tagged with
Tree-Tagger and CLAWS so that the recall rates for be Ved were above 98%.
Second, the chapter illustrates how the learners, apart from writing two
argumentative essays, completed: 1) a profile questionnaire; 2) a psychometric
test to measure overall cognitive skills with subscores for verbal, reasoning
and concentration ability (PSB-R 6-13, Prüfsystem für Schul- und
Bildungsberatung für 6. Bis 13. Klassen- revidierte Fassung); 3) a
standardized questionnaire to measure their motivation for their performance
(FLM 7-13, Fragebogen zur Leistungsmotivation für Schüler der 7. Bis 13.
Klasse); 4) an experimental task where they transformed sentences from active
into passive; and 5) a rating reliability of their responses in their
experimental task.
Third, to compare data from SCooLE with relevant reference language varieties,
the chapter outlines how the Teaching Materials Corpus (TeaMC) and a subcorpus
of argumentative essays from A-level students from the Louvain Corpus of
Native English Essays (LOCNESS) were prepared for this study. The TeaMC
(970,494 tokens) was compiled for this project and consists of continuous
texts from textbooks for EFL and CLIL. The TeaMC compilation and normalization
is fully documented in this chapter.
Now that the data collection has been discussed, Chapter 5 deals with how the
data was analyzed. First, the chapter revisits in detail the querying of the
passive constructions in the three corpora. Next, the variables described in
Chapters 1 and 2 are revisited, with some being eliminated. Those that are not
eliminated are further described. Finally, the data is analyzed using the
following four methods: 1. a quantitative analysis, 2. a structural analysis,
3. a lexicogrammatical analysis, and 4. an error analysis. The section on
methodology for the quantitative analysis explains why mostly non-parametric
tests were chosen for the statistical analysis (Table 5.3.). This section also
discusses the multiple regression modelling used to explore the influence of
the educational context. The section on methodology for the structural
analysis presents constructions that refer back to those introduced in the
literature review in Chapter 3. There is no statistical analysis of this
structural data, apart from frequency analysis, as the author analyzed
concordances from ScooLE without linking them to individual learners. The same
applies to the lexicogrammatical analysis of the corpora. Nevertheless,
individual frequency scores of production of the passive in SCooLE were
included in the quantitative analysis.
Chapter 6 analyzes in detail the EFL and CLIL teaching materials by first
reminding the readers that the official curriculum of Baden-Württemberg sets
the use of the passive as one of its learning goals. Next, the EFL and CLIL
materials included in the TeaMC corpus are separately explored for the
frequency with which they use the passive voice. Moreover, the structural
characteristics and lexicogrammatical features of the passive constructions
found in each set of materials are closely scrutinized. The chapter also
includes a genre-based analysis of passive constructions in the materials.
Since the two series of EFL textbooks included in TeaMC offer insights into
the kind of instruction learners received, there is a supplementary analysis
of their grammar sections, exercises, and vocabulary lists. As expected, and
consistent with the official curriculum, all tenses are present at some point
from year 7 to year 10. Passives with complex-transitive verbs and impersonal
passives are not taught. The grammar exercises focus mainly on forming passive
sentences, choosing between passive or active, and activities that combine
both kinds of exercises. Whereas one series includes more activities and
focuses mostly on learning the form of the passive, the other series presents
more varied, but fewer, tasks. The rest of the analysis (i.e., frequency,
structural and lexicogrammatical analysis) do not present other main
differences between the series.
When comparing CLIL and EFL materials, CLIL+ learners are exposed more to
passives with be Ved and central passives than the non-CLIL learners.
Moreover, more structures were presented in EFL as lexical chunks.
Chapter 7 explores the information provided by the operationalized learner
variables. This analysis supports the idea that entrance in CLIL+ programs is
indeed more selective. First, although not all participating schools offer the
same options to their learners, the official state curriculum indicates that
CLIL instruction is for learners “with above-average cognitive capacities and
motivation” (p. 228). Secondly, the analysis showed, for instance, that
students who repeated years did not tend to be part of CLIL+. Other factors
showcased a more motivated CLIL+ group, as exposure to English in their free
time was significantly higher for them and they felt more competent regarding
grammatical knowledge. The CLIL+ group also obtained significantly higher
results for cognitive abilities. Nevertheless, many of the personal variables
were not different between the groups (e.g, age, gender, L1, etc.).
Chapter 8 quantitatively measures the influence of the educational setting.
First, the comparison of the three groups shows that the CLIL+ strands wrote
significantly more passive forms than the non-CLILC strands in the essay that
did not prime passive with its prompt. However, the essay with a prompt
utilizing passive primed all participants to use more passive.
A multiple regression is carried out to answer the main question of this
study: what factors contribute to CLIL success. Albeit to a small degree,
variables related to the selectivity of the CLIL system and variables related
to the instructional setting contribute to the production of passive. That is,
selectivity is not the only contributing factor to CLIL success.
The quantitative analysis is followed by a structural analysis and a
lexicogrammatical analysis of the SCooLE corpus and the subcorpus of LOCNESS,
as well as an error analysis of the annotations performed during the
normalization of the corpus. In general, ratios in the SCooLE were similar
across instructional settings. Differences in the use of tenses and auxiliary
verbs show probable influences of instructional settings, while differences in
conceptions of agenthood show probable influences of both instructional
settings and L1. The lexicogrammatical analysis shows effects of learned
lexical chunks. Moreover, non-CLIL learners made more errors than CLIL
learners.
Chapter 9 focuses on the experimental task that complements the corpus data.
In the task, participants were given 12 prompts including constructions with
one object (4), two objects (4), prepositional phrases (2), complex-transitive
(1) and impersonal (1). Four scores are given to each learner based on
correctness of: 1. Sequence of constituents, 2. Passive auxiliary, 3.
Participle, and 4. Overall correctness. CLIL participants outperformed the
non-CLIL groups in these scores. The regression analysis of educational
setting and learner variables suggests that variables relevant to the
selectivity of the program and instructional setting can explain most of the
learners’ results. Selectivity variables tend to have the most explanatory
value.
As in the other chapters, the author continues analysing the data according to
its structural and a lexicogrammatical properties. This is an opportunity for
the reader to see learner examples that exhibit the probable effects of
instructional materials and L1.
Chapter 10 summarizes the previous chapters and includes some points for
further research.
EVALUATION
This monograph achieves its main goal; namely, to explore a variety of factors
influencing success in CLIL programs. In so doing, it confirms that
selectivity, while an extremely important factor, is not the only factor
influencing learners’ performance. Corpus and experimental data give a clear
view of the passive voice constructions learners can produce at this stage.
Another main contribution of this monograph is its regard for teaching
materials, which tend to be neglected in SLA research. For example, knowing
the instructional sequence of the passive forms that are taught, and not
taught, allows for a better understanding of learners’ proficiency
development. For instance, those forms that are barely taught, and are also
less frequent in the utterances of native speakers, have similar frequencies
in the L2 learners. Furthermore, textbook use was not consistent by the
participants. For instance, CLIL+ learners had access to both EFL and CLIL
textbooks, whereas non-CLIL learners had only access to EFL textbooks. The
richer input, both in quantity and in teaching approaches, for the CLIL
learners is an advantage that might have been missed, had the author not
addressed teaching materials in this study.
Although the results are relevant for CLIL and EFL programs, this book is not
for teachers, unless they have a solid background in linguistics. That is,
this book is not for those seeking pedagogical recommendations about how to
effectively teach passive voice. Nevertheless, one compelling pedagogical
recommendation of the author is to avoid mixing books from different
publishers as each publisher might have sequenced the passive in a certain way
(see Chapter 6); and this may negatively affect learners. However, given that
textbooks do not necessarily mirror teachers’ practices (e.g., Shawer, 2010),
this recommendation should be further explored including data from teachers’
perspectives. As Möller indicates, learners’ development also depends “on the
didactic capacities of teachers, as well as their linguistic competence, their
teaching styles, etc.” (p. 366).
Overall, this monograph suggests new questions related to teaching practices
and perspectives, and for pedagogical interventions-based studies to see how
the passive voice could be more effectively taught in CLIL and EFL settings.
For instance, future research might follow up on the finding that an increased
number of years in EFL education negatively correlated with correct responses
in the experimental task for CLIL0, but positively correlated for CLIL+. On
the contrary, primary school instruction following an immersive-reflective
lessons approach correlated positively with correct use of passive. Thus,
successful instructional settings deserve further scrutiny to improve
practices and increase students’ accuracy and motivation.
As for additional information regarding the instructional settings, it might
have been helpful to have more information about the schools from which the
participants were recruited. For example, it was not clear how many schools
participated, where they were located, and what kind of program each one of
them had. Additionally, the socio-economic construct was measured according to
the size of the city where the school was. This did not help in establishing
the individual socio-economic status of each learner. Thus, the reader does
not have enough information to understand how socio-economic factors might
contribute (or not) to the success of CLIL learners. This leaves some
questions for further studies.
Moreover, further studies might explore whether the analysis of oral data
could provide more examples for the get Ved structure analyzed in this study.
As the author explains, the oral features of this form make it different than
the be Ved structure. As for the TeaMC corpus, more complex statistical
analysis could potentially be performed. It does not seem necessary to know
who the author of each text is to be able to analyze a textbook as a whole
unit (see page 183).
In brief, this is an informative study that helps the reader better discern
not only how CLIL is understood in Germany, and, more specifically, in the
Baden-Württembergish context, but also what factors contribute to its success.
REFERENCES
Shawer, S. F. (2010). Classroom-level curriculum development: EFL teachers as
curriculum-developers, curriculum-makers and curriculum-transmitters. Teaching
and Teacher Education, 26(2), 173-184.
ABOUT THE REVIEWER
Nausica Marcos Miguel completed her doctoral degree at the University of
Pittsburgh in Hispanic Linguistics with a focus on Second Language
Acquisition. She has taught language courses in Spanish, English, and German.
She is an Assistant Professor at Denison University, where she teaches Spanish
language courses and linguistics. She is interested in learning and
instruction of morphology and vocabulary, and in materials research.
------------------------------------------------------------------------------
***************** LINGUIST List Support *****************
Please support the LL editors and operation with a donation at:
The IU Foundation Crowd Funding site:
https://iufoundation.fundly.com/the-linguist-list
The LINGUIST List FundDrive Page:
https://funddrive.linguistlist.org/donate/
----------------------------------------------------------
LINGUIST List: Vol-30-161
----------------------------------------------------------
More information about the LINGUIST
mailing list