ERIC ED367162: Second Language Acquisition Research in the Language Classroom.

26 Pages·1991·0.44 MB·English

by  ERIC

DOCUMENT RESUME
ED 367 162
FL 021 884
AUTHOR
Nunan, David
Second Language Acquisition Research in the Language
TITLE
Classroom.
PUB DATE
91
NOTE
26p.; In: Sadtono, Eugenius, Ed. Language Acquisition
and the Second/Foreign Language Classroom. Anthology
Series 28; see FL 021 883.
PUB TYPE
Information Analyses (070)
Speeches/Conference
Papers (150)
EDRS PRICE
MF01/PCO2 Plus Postage.
Classroom Environment; Classroom Techniques; *Data
DESCRIPTORS
Collection; *Educational Environment; Educational
Research; Foreign Countries; *Language Research;
*Research Methodology: Second Language Learning;
*Second Languages: Statistical Analysis
ABSTRACT
The state of second language acquisition research,
particularly that which is classroom-oriected, is examined in a
review of 50 empirical investigations undertaken over the last 25
years. The studies were analyzed according to the following
dimensions: the environment in which the data were collected
(classroom, naturalistic, simulated classroom, or laboratories);
rationale of the research; research design and data collection; type
of data collected (quantitative or qualitative); and type of analysis
conducted (grammatical, statistical, and/or interpretive). It is
concluded that while there have been considerable advances in the
field, future research would benefit by incorporating: more
contextualized (classroom-based rather than classroom-oriented)
research; an extension of the theoretical bases of research agendas;
extension of the range of research tools, techniques, and methods,
adopting and adapting them where appropriate from content classroom
research; re-evaluation of the distinction between process-oriented
and product-oriented research; and a more active role for classroom
practitioners in applied research.
(MSE)
***********************************************************************
Reproductions supplied by EDRS are the best that can be made
from the original document.
*********************************************************************
wfb
SECOND LANGUAGE ACQUISITION
RESEARCH IN THE LANGUAGE CLASSROOM
David Nunan
U S. DEPARTMENT OF EDUCATION
-PERMISSION TO REPRODUCE THIS
Office of Educahonet Research and improvement
MATERIAL HAS BEEN GRANTED BY
DUCATIONAL RESOURCES INFORMATION
CENTER tERICI
s clocomeni has been ,e0,0Ouced as
Nrse-N
eiyed horn Ihe Pelson 0, organizahon
r
originating 4
M,nor chang ss have been made to improve
leoroduchon qualoy
Pooms of v.ew or Of_ mons staled ,r1 ms 00c o
meet do not necessarily represent official
TO THE EDUCATIONAL RESOURCES
OE RI position or pohcy
INFORMATION CENTER (ERICr
cD0
too
716
2
5.
RESEARCH
SECOND LANGUAGE ACQUISITION
CLASSROOM
IN THE LANGUAGE
David Nunan
Abstract
critical look at the current state of SLA
The purpose of this paper is to take a
which is intended to provide directions
research, focusing in particular on research
based on an analysis of 50 empirical
for classroom teaching. The study is
and reported in the literature over the
investigations which have been carried out
last 25 years.
Introduction
the ways in which any learner,
Second language acquisition (SLA) refers to
language. The learning may take place in
child or adult learns a second or foreign
second or foreign language scttings. The
tutored or untutored environments, and in
undergone tremendous growth over the last
field of second language acquisition has
is
devoted exclusively to the subject, and it
twenty years. There are now journals
Ph.D. programs in SLA.
even possible to undertake
increasing scrutiny by those involved in
SLA research is coming under
and materials writers are interested in
language pedagogy. Curriculum developers
into optimal sequencing orders for
insights which the research can provide
best
with those pedagogic tasks which can
acquisition, while teachers are concerned
facilitate acquisition in the classroom.
language acquisition is synonymous with
For many language teachers, second
claims of relevance to pedagogy.
Krashen (1981, 1982) whose work contains strong
under attack from a variety of sources.
In recent years, however, this work has come
unanswered) critic is Gregg (1986) who, in
Perhaps the most trenchant (and, as yet,
a celebrated review wrote:
which a radio interviewer tries to get
There is a Monty Python review in
explain her new theory about the
Miss Ann Elk, a dinosaur expert, to
and hawing, false starts and
brontosaurus. After a great deal of hemming
1
3
Brontosauruses were very
general time wasting, we are finally told this:
then very thin again
thin in the front, much, much thicker in the middle and
that, as a theory, this is rather
at the end. Most of us would agree
Elk before she can tell
unsatisfactory (indeed, the interviewer shoots Miss
it was not meant to be taken seriously as a
us her second theory). But then
theory.
is meant to be taken
Reading The Input Hypothesis, which evidently
Hypothesis is the latest in a
seriously, brings Miss Elk to mind. The Input
much repeats what he
series of books and articles in which Krashen pretty
is, he offers 'what I call,
has said in all his other books and articles; that
language acquisition' (p.vii).
perhaw audaciously, a theory of a second
'audaciously'; and in fact
(There are perhaps more fitting words than
simply of second language
Krashen usually drops the article and talks
complex error of suggesting
acquisition theory, a location that makes the
language theory exists, and that his
that his theory is a theory, that a second
theory is it.) (Gregg 1986: 116-7).
about deriving pedagogical
While many SLA researchers are cautious
Given the controversies and
implications from their research, others are not so coy.
themselves, to what extent should the
disagreements among the researchers
value? This question was one which
practitioner take these claims at their
invited to carry out a review of
preoccupied me to the point that, when I was
Second Language Acquisitior, I decided to
research for a special issue of Studies in
research later in
point of departure. I tall summarise this
use the question as my
the paper.
Background
Lightbown suggests that SLA studies
In an important review of SLA research,
learners' use of their second language
"are designed to investigate questions about
language acquisition and use" (Lightbown
and processes which underly second
distinction which is commonly drawn between
1985: 173). Her statement reflects a
tradition.
research and the process-oriented
the product-oriented tradition in SLA
describe and predict the stages through
The aim of product-oriented research is to
second language. Process-oriented research, on
which learners pass in acquiring a
which may
identifying those pedagogic variables
the other hand, is aimed at
facilitate or impede acquisition.
2
'J.
instructional
The variables may relate to the learner, the teacher, the
treatment/environment or some form of interaction between these.
following:
Questions which classroom SLA research seeks to address include the
facilitate second
What types of classroom organisation and grouping patterns
language development?
What task and activity types facilitate acquisition?
What are the characteristics of teacher talk (including questions, amount, error
of this talk for
feedback, instructions, directions) and what are the implications
acquisition?
of acquisition?
Does formal instruction make a difference to the rate and/or route
achievement?
What affective variables correlate with second language
acquisition?
What type of input facilitates comprehension and, by implication,
and, by implication,
What interactional modifications facilitate comprehension
acquisition?
similarities and
One major strand of SLA research is that which has focused on
of the classroom. It
differences between input and interaction inside and outside
classroom and naturalistic
has been observed that there are clear differences in both
types of teacher
settings in terms of patterns of interaction, language functions,
and Long, 1991
questions and so on. (See, for example, Pica 1983; Larsen-Freeman
the two settings and the
for a summary of similarities and differences between
implications of these
possible consequences of these for acquisition.) The
should resemble real life
differences, and the extent to which classroom interaction
interaction, are still being debated (van Lier 1988).
in the SLA research
Long (1983; 1986) has argued that there is no evidence
which learners acquire
literature that classroom instruction can alter the order in
and in fact, all the
particular morphosyntactic feature of the target language,
literature suggests that
evidence is to the contrary. However, his review of the
In the first instance,
formal instruction does seem to be advantageous in three areas.
Focusing learners on form
formal instruction can facilitate processes of acquisition.
If instructed learners are compared with those
can help make target features salient.
in the early stages of
picking up the language naturalistically, it is found that
In
acquisition the instructed learners are, paradoxically, making more errors.
morphosyntactic items
particular, they make 'oversuppliance' crrors, inserting
loving a dog' or 'The girls
where they arc not required, saying things such as 'He
loves her dog'.
3
5
.-
their
teachers that one of the effects of
Initially, it looks like a disaster for
kinds of errors that even naturalistic
work is that learners make new
they seem to slide back down to
learners don't make, but in the long run
if you look
these things over time. Whereas
the appropriate suppliance of
still doing
later in their development, they are
at naturalistic learners much
affected
and deletion. So I think the processes are
an awful lot of omission
instruction. (Long 1987:295)
in a healthy way by formal
that instructed
acquisition processes, Long argues
In addition to facilitating
Finally,
quickly than uninstructed learners.
learners develop their skills much more
end up being much
it seems that instructed learners
in terms of ultimate attainment,
instruction.
those who do not receive formal
more proficient than
The Study
invited to survey
of this paper, I was recently
As I indicated at the beginning
language acquisition.
classroom-oriented research in second
the current state of
of the claims made in
subject to critical scrutiny some
This study was carried out to
teachers, curriculum
which purport to be relevant to
the literature. Fifty studies
analytical dimensions.
analysed according to five
developers and teachers were
There were:
they collected in
the data were collected (Were
The environment in which
1.
classroom, in
environments outside the
classrooms, in naturalistic
simulated settings, or in laboratories?)
principally to
(Were the studies carried out
The rationale of the research
2.
pedagogy?)
inform those concern with
the data collected
method of collection (Were
The research design and
3.
What methods were used?)
through experiments or not?
qualitative
(Did the study yield quantitative or
The type of data collected
4.
data?)
and
(What grammatical, statistical,
The type of analysis conducted
5
out?)
interpretive analyses were carried
Results
of the survey, see Studies in Second
(For a detailed description and analysis
Language Acquisition, 1991, 13, 2.)
Environment
the fact that, of the fifty
The most surprising outcome of the survey was
in language classrooms which were
studies, only fifteen were actually carried out
learning rather than research. Twenty
constituted for the purposes of teaching and
in laboratory, simulated or naturalistic
eight of the studies were carried out
in mixed environments (that is,
environments, and seven studies were carried out
non-classroom
collected in classroom, and some in
some of the data were
environments.)
Design
collected through some form of
Design refers to whether or not the data were
only two of the fifteen classroom-based
experimentation or not. Not surprisingly,
Roughly half (thirteen of the twenty-
studies utilised some form of experiment.
experiment, as did the studies carried out
eight) non-classroom studies involved an
the seven).
in 'mixed' environments (three out of
observation,
the studies. These included
There were eight methods used in
introspection, questionnaire, and case study.
transcript, diary, elicitation, interview,
with
employed data collection method,
Elicitation was the most frequently
obtain their
form of elicitation procedure to
exactly half of the studies using some
of
studies which obtain their data by means
data. Classified under this heading arc
The use of such
picture, diagram, standardized test, ctc.
a stimulus, such as a
order
research since the original morpheme
devices has been a feature of SLA
When
of the Bilingual Syntax Measure.
studies obtained data through the use
it is important to consider the extent to
evaluating research utilizing such devices,
(scc,
of the elicitation devices employed
which the results obtained arc an artifact
5
7
implications
for example, Nunan 1987 for a discussion on the dangers of deriving
cautious in
for SLA from standardised test data). One needs to be particularly
Ellis (1985) has
making claims about acquisition orders, based on elicited data, as
pointed out. (Sec also, Eisenstein, Bailey and Madden, 1982).
refers
'Observation' can be either focused or unfocused. Focused observation
of language and
to studies in which the investigator looks for specific aspects
for classifying
behaviour, usually with the assistance of an observational instrument
while non-
the behaviour being investigated. From the tables, it can be seen that
classroom
classroom investigations tended to utilise some form of elicitation,
('Transcripts'
studies were more likely to utilise observation or transcript analysis.
initially to some form
refers to the analysis of interactions which are not subjected
of categorisation, but which undergo interpretative analysis later.)
information
'Questionnaires' are defined as instruments in which prespecified
oral responses.
is collected from informants through either written or
A closed
Questionnaires can be either closed or relatively open-ended.
which require
questionnaire solicits data which can be readily quantified (e.g. those
questionnaire enables
zubjects to circle the appropriate response), while an open
questionnaires which
subjects to provide a free-form response. Constructing
designed to
unambiguously elicit accurate responses is difficult, and questionnaires
complication of
obtain information about language learning have the additional
sometimes being mediated through the learner's first language.
According to their
Interviews can also be relatively closed or open-ended.
the learner's first or target language.
purpose, they may be conducted either in
written, discursive
The term 'diary' is used as a form of shorthand to refer to
contain &cc-form accounts of
accounts of teaching or learning, and which therefore
learners, teachers or outside
the learning/teaching process. They may be kept by
researcher/observer.
cognitive psychology
The use of introspective methods has a long history in
analysis in Ericsson and
(scc, for example, the use of verbal reports and protocol
in second language
Simon 1980; 1984), but has only recently made its appearance
in Faerch and Kasper
research. The emerging status of thc method is reviewed
(1987).
6
8
questionnaires, interviews, diaries and
From the study it was found that
SLA research. This may
introspection are infrequently used in classroom-oriented
view introspective and self-report
reflect the suspicion with which SLA researchers
data.
Type of Data
studies according to the type of
Initially, it had been intended to classify the
virtually all of the studies were based on
data collected. However, it was found that
being either discursive (e.g. transcripts) or
qualitative data as defined by Grotjahn,
observation schemes etc.) This analysis
nominal (e.g. functional categories,
useful or insightful and was discarded.
therefore did not add anything particularly
the common-sense notion of
(In the event, it may have been better to use
Grotjahn's more stringent notion of
numericality to define 'quantitative' rather
interval scale.)
whether or not the data are measured on an
Type of Analysis
and interpretative were conducted,
Three types of analysis, linguistic, statistical
in Tables 1, 2 and 3.
and the results of this analysis are set out
Table 1
Classroom-based Studies: Type of Analysis
5
Functions
Linguistic
4
Complexity
3
Morphosyntax
Suprasegmentals
1
Lexis
1
2
Correlation
Statistical
2
Chi square
(-test
1
U test
1
9
Interpretive
7
9
CS.
,7
Table 2
Laboratory, simulated and naturalistic studies: Type of Analysis
Morphosyntax
Linguistic
9
8
Functions
2
Complexity
2
Quantity
Correlation
7
Statistical
Chi square
5
6
t-test
2
Factor analysis
2
F-ratio
2
Cronbach's alpha
2
ANOVA
ANCO VA
1
6
Interpretive
Table 3
Mixed studies: Type of Analysis
6
Linguistic
2
Morphosyntax
Chi square
3
Statistical
2
t-test
Correlation
I
3
Interpretive
Not surprisingly, classroom studies tend to be more interpretative and make a
reinforces the
more limited use of statistics than non-classroom studies. This
picture which emerged in relation to the design and methods issues already
discussed. It should be pointed out that the notion of 'interpretative analysis' is a
relativistic one. All studies, even those employing a true experimental design and
8
l 0