Three Distinct but inter-related Classroom Research Paradigms:

[English Language Professors]

New trends in language teaching have resulted in a move towards
research in the language classroom. A brief overview of classroom
research reveals three distinct but inter-related research paradigms:
classroom-centered research, classroom process research, and
qualitative research, respectively.

1. Classroom-Centered Research

By definition, classroom-centered research is simply research centered
on the classroom as distinct from other research types. This research
type usually views classroom as the "object" of research, and not
simply the "setting" for research. Classroom-centered research should,
however, be taken as a cover term for a whole range of research
studies on classroom language learning.

Research on classroom language learning is basically done by either
observation or introspection, or even a combination of both.
Observation necessarily implies keeping a record of what goes on in
the classroom. To this end, different techniques are available to the
researcher. The use of audio-tape recordings, video-tape recordings,
and so on could be enlisted as some of these techniques. Even a
trained observer can handle the job of doing the observation.

A second approach to classroom-centered research is introspection.
Allwright uses the term to refer to research techniques that involve,
for instance, asking people to answer questions rather than asking
them to allow themselves to be observed in action. In any case,
introspection always calls for self-reporting of some kind. The use of
questionnaires or interviews can be viewed as a good means of
eliciting introspective data. A fairly recent development is, however,
the use of diary keeping.

A third approach is the use of what Allwright calls "triangulation."
Multiple viewpoints, at least three, may be necessary if we are to
understand what actually goes on in classrooms. Allwright further
argues that, in practice, triangulation means a combination of
observation and introspection. This calls for a good number of
observers and introspects.

Modern classroom-centered research is deeply rooted in the attempts of
teacher trainers (in the fifties) at responding to the need to provide
student teachers with adequate feedback on their teaching. At that
time, the finding out of what constituted good effective teaching was
of prime importance. Teacher trainers believed that only through such
investigations would it be possible to train effective teachers. To
this end, it was necessary to develop the tools of classroom
observation. Some scholars as Flanders (1960) chose to use direct
observation. Some others developed what they called observation
schedules. Early on, the researchers realized that the application of
their findings to teacher development was actually premature. This
finding shed light on the fact that the enormous complexity of
classroom behavior is so varied that it could never be simply reduced
to a few categories (Dunkin and Biddle, 1974).

Classroom-centered research is almost a new trend in the field of
language teaching profession. One reason why the language teaching
profession realized the importance of this research type so late might
be the fact that language teachers have been enjoying a period of
euphoria and unprecedented confidence in the methods they used in
their classrooms. This preoccupation with methods could be eloquently
called "the era of global methodological prescription" to borrow a
phrase from Allwright. That is, instead of realizing what actually
happened in the classroom, the training of language teachers was
informed by the issue of which of the major methods to prescribe. It
was not until very late in the 1960s that the scholars in this
profession began to realize, in the light of research done by Scherer
and Wertheimer (1964), Smith (1970), Otto (1969), and others, that it
no longer made sense to imagine that any one method would prove in
some absolute way superior to its rivals.

This urged some scholars to move a step down the hierarchy of approach-
method-technique. They, therefore, began to carry out the so-called
small-scale research projects at the level of technique. As Carlsson
(1969) mentions, in Sweden the GUME (Gothenburg English Teaching
Method) Project was an offspring of such an orientation toward
technique and away from method. Politzer (1970), in the US, video-
taped a number of language classes, recorded the frequencies with
which certain techniques (e.g. structural pattern practice) were used,
and correlated these frequencies with learner achievements in
different classes. Politzer (1970: 42) notes that "the very high
complexity of the teaching process makes it very difficult to talk
about bad and good teaching devices in absolute terms.

It was soon apparent that the level of technique was not a reliable
place to dwell in. Therefore, it seemed inevitable to retreat even a
further step back into the kept-in-the-dark arena of classroom
processes. To sum up, two moves were involved: (1) movement from
prescription to description; and (2) movement from technique to
classroom processes. These two moves, when taken together, called for
an effort to find ways of describing classroom processes to find out
what actually happened in language classes.

Classroom-centered research has been divided into two distinct
branches. On the one hand, some scholars have drawn on a sociological
outlook on education and have tended to look at language lessons as
socially constructed events. The teacher in this approach is no longer
viewed as the all-knower and the only available source of knowledge.
Classroom activity is viewed as a collective endeavor toward the
production of learning opportunities. On the other hand, some other
scholars including Gaies (1977) have viewed the language classroom as
a setting for classroom language acquisition. The teacher in this
approach is viewed as a source of input. The role of teacher talk in
classroom language acquisition is the main focus of investigation.
These two approaches should, however, be viewed as complementary
rather than mutually exclusive practices.

The twin moves from "prescription to description" and "technique to
process" have resulted in a move from teacher training to something
more like fundamental research. This, no doubt, has resulted in a
revision of the tools of classroom-centered research. Observation
schedules have been modified to be appropriate to the obvious
complexities of language teaching. Moskowitz (1971) produced the most
widely known and used modification of a general educational schedule
and called it FLint (Foreign Language Interaction System). FLint was
actually the expanded form of Flanders' (1960) so-called FIAC. FLint
was used as a research tool in deciding what constituted effective
teaching. Fanselow (1977), in an important contribution to this area,
after modifying Bellack's (1978) pioneering analytical system,
produced an observation schedule which was called FOCUS. The FOCUS was
primarily developed with language teacher training in mind. It is,
however, a good descriptive system applicable to almost any human
interaction.

Classroom-centered research should be viewed in the light of three
important issues. First, the two viewpoints (i.e. focusing on the
interactive aspects of classroom behavior, and focusing on the process
of classroom language acquisition) should be taken together as being
complementary. Second, the study of classroom language acquisition as
opposed to natural language acquisition is still in its infancy
period. And, third, there is a controversy over research methods. That
is, some researchers prefer objectivism while others accept subjective
methods as valid.

2. Classroom-Process Research

Stephen J. Gaies (1983) draws on a new dimension on classroom-centered
research. Besides the ethonomethodological trend in sociolinguistics
which set out to investigate the kind of interaction that went on in
language classrooms, a second dimension of research concerns itself
with the investigation of classroom processes. It gives priority to
the direct observation of second language classroom activity.
Classroom-process research is primarily concerned with the
investigation of the numerous factors that shape language acquisition
in language classes. Classroom-process research aims at describing the
linguistics and instructional environment which second language
learners encounter in the classroom and how that environment might
differ from what is available outside the classroom.

It is commonly believed that classroom-process research is based on
several shared premises. It is vital to summarize these premises at
the outset:

A. There has been, as Dick Allwright argues, a perceptible trend
away from global categorization of second language classroom
instruction. Classroom-process research rejects any univariate
classification of second language instruction as simplistic.

B. The emphasis is on describing as fully as possible the
complexity of second language instructional environment. Classroom-
process research tries to identify the variables that shape classroom
language instruction. In so doing, it generates hypotheses. Therefore,
classroom-process research is considered to be hypothesis generating.
It does not directly lead to empirically validated applications;
rather, it is directed more at the clarification of those factors
which must be taken into account in any given assessment of what goes
on in language classrooms.

C. The priority of direct observation of classroom activity is
another premise which unifies classroom-process research. The main
source of data in this research type is wholly or substantially the
classroom itself. Teacher talk functions as the major source on input
for learners' language acquisition.

D. The major trends in classroom-process research include second
language classroom language (classroom input), error treatment, and
patterns of classroom participation.

Perhaps one of the most interesting aspects of classroom process
research is the investigation of the nature of classroom input. The
language classroom provides what might be almost totally inaccessible
outside the class--a native speaker (or a really proficient non-native
speaker) who is delegated to interact with learners and to provide
them with linguistic input. Gaies (1977), in an investigation of the
syntactic features of ESL classroom teachers, revealed that the
subjects' classroom speech was syntactically less complex on a number
of variables. The subjects of this investigation were observed to
drastically fine-tune their classroom speech to the level of their
learners' proficiency.

Hamayan and Tucker (1980) carried out another investigation in two
French immersion schools and three regular French schools in Montreal.
The subjects of their study were teachers of the third and fifth grade
level classes. They examined the speech and teaching behavior of these
subjects. The linguistic aspects of the speech they studied included
indirect questions, contractions, reflexives, and subjunctive (usually
viewed as the most complex syntactic aspects of any language). These
researchers found strong correlation in the frequency with which these
structures occurred at the two grade levels and in the two school
systems. They also found that the frequency with which students used
these structures in story retelling tasks correlated strongly with the
frequency of occurrence of these structures in the speech of their
teachers. This reveals the old claim that classroom linguistic input
shapes learners' linguistic production.

More recently, another important aspect of classroom-process research
has come into vogue. Attention has shifted from the nature of input to
the nature of interaction between native speakers and second language
acquirers. A prominent figure in this connection is Krashen (1978,
1980) who argues that, through interaction, second language acquirers
gain access to the so-called "optimal input" (i.e. input which is
likely to lead to further acquisition). Long (1980b) claims that the
modified input available to second language acquirers through
interaction between native speakers (here, the teacher) and the
learners is the necessary and sufficient condition for second language
acquisition to take place.

Long and Sato (1983) examined the forms and functions of ESL teachers'
classroom questions. They hypothesized that questions in and outside
the classroom tend to serve different interactional functions. They
classified questions as belonging to either the "display type"
category (i.e. questions which are intended to elicit information
already known to the questioner) or the "referential type" category
(i.e. questions intended to provide referential or expressive
information unknown to the teacher). The most striking point in their
findings was the observation of the total absence of display type
question in data gathered in naturalistic settings outside the
classroom.

Roughly at the time when language classroom process research began on
a large scale, interaction analysis predominated in educational
research. A good example of such interaction analyses is the study
carried out by Flanders (1970) (already mentioned in previous
reviews). Seliger (1977) in another study classified learners into the
two categories of "high input generators" and "low input generators"
on the basis of a numerical count of classroom participation. The
findings of this study show that "high input generators" (i.e. the
students who are more active in classroom interaction) tend to be more
field independent.

Sato (1981) studied the patterns of turn-taking in university-level
ESL classes. This study is an excellent illustration of how classroom-
process research may serve to refine our understanding of patterns of
participation. She compared nineteen Asian and twelve non-Asian
learners and found that Asian learners initiated significantly fewer
turns than did their non-Asian counterparts. In addition, the Asian
students were less often called upon by their teachers.

Schinke (1981) has examined patterns of participation in all-English
content classes. She examined the experience of LEP (Limited English
Proficiency) learners who had been mainstreamed. She found that LEP
learners had significantly fewer interactions with their teachers than
their non-LEP classmates. Teacher-LEP interaction was also revealed to
be functionally quite different from the type of interaction in which
non-LEP learners and the teacher engaged. The LEP-teacher interaction
was mainly concerned with classroom and lesson management.

A third dimension in classroom-process research is the investigation
of error treatment patterns in language classrooms. This research type
is mainly concerned with the investigation of the role of corrective
feedback in classroom language acquisition. Errors have been viewed as
windows to the nature of language acquisition process. They are seen
as overt reflections of learners' internalized knowledge of the
language. One major step forward in this connection is the abandonment
of a "global" or all-out approach to error correction in the
classroom. Fanselow (1977), for instance, found that 22% of the errors
committed by students received no treatment of any sort.

A second trend in research on error treatment has focused on the
nature of the so-called corrective feedback. In other words,
researchers have started to study the type of error treatment which is
provided by teachers. An important finding in this connection is that
when teachers treat errors in the second language classroom, they do
not necessarily provide overt corrections. They rather prefer to
provide implicit or indirect feedback. In a study by Catheart and
Olsen (1976), twenty-one teachers of adult and university ESL classes
responded to a questionnaire which asked for their preferred and most
frequently used error treatment strategies. The students in these
teachers' classes were also asked to respond to the same
questionnaire. The only striking discrepancy between the teachers' and
the students' preferences was the students' wish to be corrected much
more frequently than their teachers actually did. I personally like to
be corrected overtly on the spot.

On the whole, studies concerning error treatment reveal that the type
of corrective feedback provided by teachers depend on a number of
factors. The first of these factors concerns the type of linguistic
error committed by the learners. Studies on error treatment reveal
that, based on their linguistic type (i.e. their phonological,
lexical, syntactic nature), errors are treated differently. The type
of classroom activity during which an error occurs is yet another
factor that plays a major role in the treatment of errors. As
mentioned earlier in the discussion of the study carried out by
Hayaman and Tucker (1980), the level of instruction also plays a major
role in the type of corrective feedback provided by the teacher.
Finally, the teachers' individual styles are also significant
determinants of error treatment.

These studies on the whole reveal that error treatment in language
classrooms is often inconsistent and ambiguous. Chaudron (1977), for
instance, has pointed out that error treatment usually consists of not
a single teacher response, but rather of an exchange or cycle of
verbal moves. Feedback has also been categorized into two types: (1)
negative cognitive feedback (i.e. feedback that focuses attention on
an error), and (2) positive affective channel feedback (i.e. feedback
that encourages the learners to make further attempts at
communication).

There are alternative approaches to classroom-process research. These
alternative approaches are known by a variety of headings, among which
are anthropological, qualitative, and mentalistic research. According
to Cohen and Hosenfield (1981), the chief virtue of classroom-process
research is that it allows for the investigation of aspects of
classroom language learning which more conventional external
investigation cannot get at. A comparison of conventional and non-
conventional methods of classroom-process research reveals that:

A. Conventional classroom observation provides insufficient
accounting of learners who are reluctant to participate orally in
class.

B. Direct external observation cannot provide accurate insight
into learners' conscious thought processes. This, in turn, does not
allow for any direct examination of the means by learners change input
into intake.

C. Quantitative research requires the pre-selection of variables
to be observed and measured. It cannot, for instance, identify
individual or psychological variables of the classroom experience.
This is, as Schumann and Schumann (1977) argue, best guaranteed
through qualitative research based on learners' diaries.

In brief, current classroom-process research has two major dimensions:
First, it reveals previously unexplored aspects of classroom processes
in which teachers and learners are engaged. The other important
dimension of classroom-process research is that it may ultimately
enable us to develop and test hypotheses about second language
teaching and learning which reflect better than has been done in the
past the complex activity which we seek to understand.

3. Qualitative Research

Research in applied linguistics has taken on two major forms. Most
traditional research projects have drawn on quantitative research
methodology in which the researcher sets out to investigate already
hypothesized variables. More recently a new trend in second language
research methodology has come into vogue. Qualitative research has
made a significant gain in terms of visibility and credibility in
recent years. We should, however, admit that the purposes,
assumptions, and methods of qualitative research are still debated,
misunderstood, and/or ignored by some applied linguists.

Lazaraton (1995) draws our attention to the state of the art of
research in applied linguistics. She distinguishes between the two
major research methodologies of quantitative research and qualitative
research. The term "quantitative methods" is employed by Lazaraton to
include the application of descriptive and/or inferential statistical
procedures in research. In reviewing the related literature on
qualitative research, Lazaraton express dissatisfaction in the face of
the fact that there is no qualitative-research-specific text of any
kind. She, however, observes that some scholars have devoted some
pages of their books to a short-sighted discussion of qualitative
research topics.

Nunan (1992), for instance, in Research Methods in Language Learning
states that "two alternative conceptions of the nature of research
provide a point of tension within the book" (pp. xi-xii). Galguera
(1993) reviews Nunan's book in Language Learning journal, and argues
that Nunan displays a bias toward his stated preference for non-
experimental research despite his attempts to provide a balanced and
objective view. Johnson (1992), in his book Approaches to Research in
Second Language Learning, strives for balance and objectivity in the
presentation of six research approaches (i.e. correlational, multisite/
multimethod/large scale, survey, ethnography, experimental, and case
study).

Jacob (1987) notes that qualitative-quantitative dichotomy leads one
to conclude that only two methodological alternatives are available to
the educational researcher. In fact, Denzin and Lincoln (1994)
distinguish six interpretive paradigms and perspectives that guide the
research process: positivism/post positivism, constructivism,
feminism, ethnic models, Marxist models, and cultural studies models.

Qualitative research has its roots in a number of traditions in
different disciplines. Holistic ethnography is, for example, a
qualitative research tradition that dates back to anthropology.
Ethnography of communication is rooted in both anthropology and
sociolinguistics. Cognitive anthropology has been widely used in both
linguistics and anthropology. Other traditions include discourse
analysis, phenomenology, ecological psychology, symbolic
interactionism, heuristics, ethnomethodology, and hermeneutics which
are rooted in linguistics, philosophy, psychology, social psychology,
humanistic psychology, sociology, and theology, respectively.
Hermeneutics has also been practiced in philosophy and literary
criticism.

Henning (1986: 701) attempts to provide a definition of quantitative
research as opposed to qualitative or "anecdotal research." Brown
(1991) carefully shuns the term empirical when discussing statistical
research stating that "there are other non-statistical studies that
could be called empirical (e.g. ethnographies, case studies, etc.),
since, by definition, empirical studies are those based on data (but
not necessarily quantitative data)" (p. 570).

Lazaraton, in a discussion of what hinders the development of
qualitative research, draws our attention to the shortcomings of
qualitative research. The first problem with qualitative research is
that, to date, there is no exact definition of what constitutes
qualitative research. Besides the problem of definition, a fair amount
of controversy exists about the scientific rigor of qualitative
research. The rigor arguments seem to encompass two related issues:
(1) that quantification of qualitative data is not only possible but
also desirable, and (2) that quantification of qualitative data is
necessary in order to make generalizable claims to and about other
contexts; hence, the problem of generalizability.

Positions as to the importance of the quantification of qualitative
data are two-fold. On the one hand, people like Henning take a strong
position maintaining that "without some recourse to quantitative
methods, some marriage of words and numbers, it is inconceivable that
the investigation of language acquisition will ever be said to belong
to the realm of scientific inquiry" (1986: 702). Adopting a similar
but somewhat weaker position, Chaudron (1988) argues that "Process-
oriented qualitative researchers explore the intersubjective and
context-dependent nature of classroom events as they occur, noting the
regularities and idiosyncracies in the events" (p. 49).

The fact that some qualitative researchers themselves employ or
recommend quantification further complicates the situation. Watson-
Gegeo (1988) claims that in a hypothesis oriented mode, qualitative
research may involve "quantification in the form of frequency counts,
tests of significance, or multivariate analyses of patterns and
themes" (pp. 584-585). All these shortcomings (i.e. lack of a precise
definition, the problem of generalizability, and the tendency towards
quantification) have given qualitative research the ill-state it is
experiencing today.

Perhaps the most frequent criticism leveled against qualitative
research is that the results obtained through qualitative research
methods are not generalizable to other contexts. In defense of
qualitative research, Lazaraton argues that:

A. Generalizibility in research is more than a matter of
counting. Quantification of any set of data does not ensure
generalizability to other contexts, nor does a large sample size:
Population characteristics must be carefully considered when selecting
a sample from which to make statistical inferences.

B. Even meeting the most stringent criteria does not guarantee
meaningful interpretation of results. Donmoyer (1990) agrees with this
point and argues that "Even statistically significant findings from
studies with huge, randomly selected samples cannot be applied
directly to particular individuals in particular situations" (p. 181).

C. Critical theory has made a significant contribution to our
profession in that we have begun to question the meaning of concepts
that we take for granted. Matters of research methodology are not just
abstract, epistemological issues about the way we view the world: They
are also issues of legitimacy and power.

There are, of course, some other factors that limit the application of
qualitative research methodology. According to Watson-Gegeo (1988),
one reason ethnography is not more widely used in SLA studies is that
it views language learning from a socialization rather than language
acquisition perspective, crediting context and culture for much of
what happens in the learning environment. Because many of the studies
that use elicited, experimental data rarely consider these factors, it
is understandable why the approach has not been more widely adopted.
Second, training is another factor. Although there are books and
materials available for self-study and reference, it is not an easy
task to train oneself in any research methodology. Finally, anyone who
has completed a qualitative research project is familiar with the
sheer size of the resulting document. In spite of all this
argumentation in favor of qualitative research, Lazaraton believes
that, no matter which research methodology a researcher may draw on,
he should do his utmost to do quality research (italics ours).

References:

Allwright, D. (1988a). Observation in the language classroom. London:
Longman.

Allwright, D., & Bailey, K. M. (1991). Focus on the classroom: An
introduction to classroom research for language teachers. Cambridge:
Cambridge University Press.

Allwright, R. L. (1983). Classroom-centered research on language
teaching and learning: a brief historical overview. TESOL Quarterly.
17, 191-204.

Bailey, K. M. (1990). Diary studies in teacher education programs. In
J. C. Richards and D. Nunan (Eds.). Second language teacher education.
New York: Cambridge University Press.

Bailey, K. M. (1985). Classroom-centered research on language teaching
and learning. In M. Celce-Murcia (Ed.). Beyond basics: issues and
research in TESOL. 96-121. Rowley, MA: Newbury House.

Bailey, K. D. (1982). Methods of social research. New York: The Free
Press.

Bartlett, L. (1990). Teacher development through reflective teaching.
In J. C. Richards and D. Nunan (Eds.). Second language teacher
education. New York: Cambridge University Press.

Blum, R. E. (1984). Effective schooling practices: a research
synthesis. Portland, Ore.: Northwest Regional Educational Laboratory.

Catheart, R. L., & Olsen, J. W. B. (1976). Teachers' and students'
preferences for correction of classroom conversation errors. In
Fanselow and Crymes: 41-53.

Chaudron, C. (1977). A descriptive model of discourse in the
corrective treatment of learners' errors. Language learning. 27,
29-46.

Chaudron, C. (1988a). Second language classrooms: research on teaching
and learning. Cambridge: Cambridge University Press.

Cohen, A. (1990). Language learning: insights for learners, teachers,
and researchers. New York: Newbury House.

Doyle, W. (1977). Paradigms of research on teacher effectiveness. In
L. S. Shulman (Ed.). Review of research in education. Vol. 5. Itasca,
Ill.: Peacock.

Dunkin, M., & Biddle, B. J. (1974). The study of teaching. Washington
D.C.: University Press of America.

Fanselow, J. F. (1977b). The treatment of learner error in oral work.
Foreign language annals. 10, 583-93.

Flanders, N. A. (1960). Interaction analysis in the classroom: a
manual for observers. Ann Arbor: University of Michigan.

Flanders, N. A. (1970). Analyzing teacher behavior. Reading, MA:
Addison Wesley.

Freire, P. (1970). Pedagogy of the oppressed. New York: Seabury Press.

Gaies, S. J. (1983). The investigation of language classroom
processes. TESOL Quarterly. 17, 205-17.

Gaies, S. J. (1983b). Learner feedback: an exploratory study of its
role in the second language classroom. In Seliger and Long: 190-213.

Gebhard, J. G., Gaitan, S., & Oprandy, R. (1990). Beyond prescription:
the student teacher as investigator. In J. C. Richards and D. Nunan
(Eds.). Second language teacher education. New York: Cambridge
University Press.

Hamayan, E. V., & Tucker, G. R. (1980). Language input in the
bilingual classroom and its relationship to second language
achievement. TESOL Quarterly. 14 (4), 453-468.

Jacob, E. (1987). Qualitative research traditions: a review. Review of
educational research. 57, 1-50.

Johnson, D. M. (1992). Approaches to research in second language
learning. New York: Longman

Lazaraton, A. (1994). [Review of the book Approaches to research in
second language learning]. TESOL Quarterly. 28, 208-209.

Lazaraton, A. (1995). Qualitative research in applied linguistics: a
progress report. TESOL Quarterly. 29 (3), 455-472.

Long, M. H., & Sato, C. J. (1983). Classroom foreigner talk discourse:
forms and functions of teachers' questions. In Seliger and Long:
268-86.

Long, M. H. (1983). Training the second language teacher as a
classroom researcher. In J. E. Alatis, H. H. Stern and P. Strevens
(Eds.). GURT '83: Applied linguistics and the preparation of second
language teachers. Washington D.C.: Georgetown University Press.

Moskowitz, G. (1967). The FLint system: an observational tool for the
foreign language classroom. In A. Simon and E. G. Boyer. Mirrors for
behavior: an anthology of classroom observation instruments. Section
15, 1-15. Philadelphia: Center for the Study of Teaching at Temple
University.

Moskowitz, G. (1976). The classroom interaction of outstanding foreign
language teachers. Foreign language annals. 9, 125-43 and 146-57.

Nunan, D. (1990). Action research in the language classroom. In J. C.
Richards and D. Nunan (Eds.). Second language teacher education. New
York: Cambridge University Press.

Nunan, D. (1992). Research methods in language learning. Cambridge:
Cambridge University Press.

Porter, P. A., Goldstein, L. M., Leatherman, J., & Conrad, S. (1990).
An ongoing dialogue: learning logs for teacher preparation. In J. C.
Richards and D. Nunan (Eds.). Second language teacher education. New
York: Cambridge University Press.

Sato, C. (1981). Ethnic styles in classroom discourse. In M. Hines and
W. Rutherford (Eds.). On TESOL '81. 11-24. Washington, D.C.: TESOL.

Schinke, L. (1981). English foreigner talk in content classrooms.
Ph.D. dissertation, Northwestern University.

Schumann, F. E., & Schumann, J. H. (1977). Diary of a language
learner: an introspective study of second language learning. In Brown,
Yorio and Crymes: 241-9.

Seliger, H. W. (1977). Does practice make perfect? A study of
interaction patterns and L2 competence. Language learning. 27, 263-78.