THE RELATIONSHIP BETWEEN CLASSROOM PROCESS ANALYSIS AND INTERVENTIONIST STRATEGIES IN FOREIGN LANGUAGE TEACHING

THE RELATIONSHIP BETWEEN CLASSROOM PROCESS ANALYSIS AND INTERVENTIONIST STRATEGIES IN FOREIGN LANGUAGE TEACHING
Antony Peck

Abstract
The aim of classroom process research is to make language teaching more effective. Fundamental, descriptive research makes possible precise accounts of different teachers’ methods. Differences in classroom practice provide scope for experimentation. A teacher in a comprehensive school made principled changes to her usual way of conducting oral practice. The link between descriptive research and interventionist experimentation is made; changes in the teacher’s method are described; and the outcomes of the experimentation are examined.

The need for research
It is a basic thesis of this paper that there is a need for fundamental, descriptive research in foreign language teaching and learning. For twenty years, teachers and researchers have urged that more attention should be paid to basic research in order to make foreign language teaching more effective (Allwright, 1972; Stern, 1983; Brumfit and Mitchell, 1989). It is, above all, studies of classroom teaching process which are required, in order to discover how language teachers actually teach, as opposed to how they say they teach, or how they may be advised to teach by trainers, advisers, or by National Curriculum documentation.

Difficulties in understanding language teaching
Two main difficulties prevent the full understanding of language teaching from being widely disseminated. Teachers work in isolation behind closed doors, and very few people, even in the same institution, have more than a hazy idea of what goes on behind them. The knowledge which teachers themselves have about what they do, and how they teach, is largely instinctive and automatic, since in the pressure of teaching large numbers of students each day, the opportunities for detailed planning or careful reflection are very few. Nevertheless, teachers have a vast store of accumulated observations of what happens in classrooms as the result of their teaching procedures. This awareness is often ‘based on implicit models of what knowledge is and how it is learnt’ (Barnes, in Russell and Munby, 1992:20).

Benefits of experimenting with methods
If knowledge about processes of language teaching can be made precise and explicit, a number of benefits can be expected to ensue. Teachers who become aware of their own techniques of teaching, as a result of research, may become sensitised to the relative effectiveness of different aspects of their own methodology. Those who become aware of how other teachers attempt to attain similar objectives and deal with similar challenges, may see opportunities for varying their teaching method in an attempt to achieve greater effectiveness in their own classrooms. This involves an ability on the part of the practitioner to study how colleagues in other classrooms go about their business, and a willingness to make principled changes to their own practice, and to observe what changes, if any, occur in the performance of their students.
In other words, a potentially direct link is proposed between descriptive research and changes, one hopes improvements, in effectiveness. This may be summarised thus:

[INSERT a/w Peck: Fig. 1]

Case Study
Case Study is the research method chosen to provide the information on which the experiment described here was based. A case study approach to the observation of language lessons makes it possible to study the special and unique features of a teaching situation, or the method used by a given teacher, and to describe this, without prejudice, in the most explicit and vivid terms possible. A case study approach ‘gathers evidence systematically in a "scientific" way … and is concerned essentially with the interaction of factors and events. Sometimes it is only by taking a practical instance that we can obtain the full picture of this interaction’ (Nisbet and Watt, 1978). The product of many observations of foreign language teaching, in the UK and abroad, was a taxonomy of the techniques used by language teachers to stimulate oral practice amongst their students by means of questioning. Each of the observations was verified with the teacher at the end of the lesson, in order to establish its validity.

The research questions
Observations of teachers conducting oral practice showed that there were many, discrete techniques available, but that no teacher, during the observed lessons, used them all. This led to the formation of the following research questions:
Can a foreign language teacher change or add to his/her repertoire of teaching techniques? If so, will observable results ensue?
These questions are derived from proposals for experimentation in the conduct of oral practice resulting from my own earlier observations of foreign language teaching (Peck, 1988).

Background to the experiment
Mary teaches French at a Welsh-medium, bi-lingual comprehensive school. She agreed to take part in an interventionist experiment as part of her teaching of pupils aged 13-14 (see Peck et al., 1990).
The first part of the experiment consisted of an attempt to describe as precisely as possible her existing method of conducting oral practice. Since language teaching goes in cycles consisting of several lessons, it was seen to be necessary to focus on one small segment in order not to overload her with work. Accordingly, it was agreed that one lesson each week would be designated a ‘research lesson’. Part of this lesson, though not necessarily all of it, would feature a substantial amount of oral practice, consisting of question-and-answer. The class was well aware that this lesson had something special about it, and it quickly became known as the ‘York lesson’.
During the first half of the Autumn Term, some seven weeks, the class was taught in the normal way, except for the provisions described above. Mary kept notes about how she had conducted the oral practice during the designated lesson, but did not attempt any experimentation. It is, of course, possible, even likely, that she prepared these lessons with special care during this time, but no experiments were undertaken. Indeed, it was not until the half-term break that experimentation was discussed. The purpose of concealing until this time the subsequent progress of the research project was to throw into the highest profile any changes which she might make to her method, when she was invited to intervene in her normal process of teaching.

The old method
During the first half of the term, Mary kept track of how she used questions and answers, and prepared an inventory of her techniques. Here it is.
1. Questions categorised by type of answer required
a. Mary frequently asked questions where pupils had to reply on their own behalf; that is to say, she did not know in advance what the answer would be. These questions related to a GCSE topic, not a point of grammar. Pupils consequently used different grammatical areas in each answer. Here are some examples:
Comment est-ce que tu arrives à l’école?
Qu’est-ce que tu manges à midi?
Quelle matière est-ce que tu préfères? Pourquoi?
Questions like this might account for about two thirds of Mary’s normal oral work.
b. Another type of question she often used was designed to make pupils practise a particular feature of the language, e.g. a specific verb, such as ‘finir’, with its typical conjugational forms. Here are some examples:
A quelle heure est-ce que tu finis tes devoirs?
A quelle heure est-ce qu’il finit ses devoirs?
A quelle heure est-ce que nous finissons à l’école?
Or: Est-ce que la biologie est plus intéressante que la physique?
Est-ce que les maths sont plus difficiles que les sciences?
Or: Est-ce que Swansea est plus grande que La Rochelle?
Est-ce que la Seine est plus longue que la Tamise?
Est-ce que Paris est plus grande que Londres?
These sorts of questions would typically take up about a third of Mary’s oral teaching time.
c. Some of her questions were intended to check if pupils had understood a text. She put them verbally in Welsh, and pupils responded verbally in Welsh. Here are some examples.
In what class is Marc?
What day is it today?
What is his first lesson?
The above sets of questions could come from a textbook, or could be invented by the teacher. The differences between them tell us something about the syllabus this class was learning, but nothing at all about Mary’s method of using question-and-answer. For that, we must look deeper.
2. Questioning described by method
a. Mary invariably framed her questions so that pupils were required to respond, using a complete sentence. This was most often achieved by asking a question beginning with an interrogative, e.g.
A quelle heure …?
Comment …?
Qu’est-ce que …?
Just occasionally, she used the form: Est-ce que …? but her intention, on those occasions, was that pupils should respond, using a specific grammar point in a full sentence. She would not accept ‘oui’ or ‘non’, by itself.
b. She normally used each question once only. That is to say, she addressed the first question to one pupil, the second to another pupil, the third to another, and so on.
c. Sometimes, she would get the pupils to ask each other questions. Here are some examples, showing how she did this.
How do you ask:
Is there a bank nearby?
What do I do to change money?
Do you accept travellers’ cheques?
Nevertheless, it is true to say that Mary normally put all the questions herself.

The interventionist strategy (Didactic elements)
At half term, the means of intervening in Mary’s normal teaching method was discussed. An analysis of questioning techniques, derived from extensive observations of foreign language lessons in the UK and overseas, was presented for Mary’s consideration. There was no obligation on her to use all the techniques, and in fact she only used about half of them. The choice was left entirely to her, to choose those which she felt accorded with her own, established teaching style. The analysis was presented in the categories described below.
Questions categorised by type of answer required.
a. Formal - Short, abbreviated, but acceptable answers, e.g.
What’s your name? John.
- Complete sentences, demonstrating a command of the structure to be learned and used, e.g.
What’s your name? My name is John.
b. Functional - Questions designed to elicit a grammatically-based answer. The answers to such questions would be known to the teacher in advance, e.g.
Are you cleaning your teeth now? No, I’m not.
Do you clean your teeth in the morning? Yes, I do.
Are you washing your face now? No, I’m not.
Do you wash your face in the morning? Yes, I do.
- Questions designed to elicit a communicative answer. The answers to such questions would not be known in advance to the teachers, e.g.
How do you come to school?
What do you have to eat at midday?
Which subject do you prefer? Why?
Communicative questions, such as those described above, can themselves be sub-divided into discrete types. These also were presented to Mary as possible variations.
(i) Opinion questions. For such questions it would be understood that there was no right answer, and students would be encouraged to express an opinion.
(ii) Inference questions. For these, only part of the information required for the answer would be available, and students would have to speculate in order to respond; or ask elucidating questions.
(iii) Personal questions. A single question of this type could theoretically produce as many right answers as there are students present.
(iv) Group questions. For questions of this sort, students would need to collaborate in pairs or groups in order to provide an answer.
(v) Research questions. Questions like this would require students to go to a source of information, and possibly make notes, in order to be able to respond.
c. Level of difficulty
It was put to Mary that the answers to her questions might also vary according to their difficulty. Thus, some questions might be answered accurately and fluently, while others might be answered, if at all, haltingly and with mistakes.
d. Raising the level of difficulty
(i) First, ask a question beginning with a verb, and which, consequently, can be answered: ‘yes’, or ‘no’, e.g. Is it raining?
(ii) Second, ask questions containing an alternative. This allows the students to hear the answer within the question, and reduces the answering task to choosing and pronouncing, e.g. Is it raining, or snowing?
(iii) Third, ask questions beginning with an interrogative, thus requiring the student to produce an answer without any help, e.g. What’s the weather like?

The interventionist strategy (Pedagogical elements)
e. Type of interaction
(i) The teacher asks the questions.
(ii) Students ask each other the questions.
(iii) Students ask the teacher questions.
(iv) The teacher put the same question to a number of students in turn.
(v) The teacher uses a set of two or three questions, alternating them in random sequence in rapid succession with a number of students.
(vi) The teacher tells student A to ask student B a question, e.g. Ask Peter what time it is.
(vii) The teacher tells student A to ask a question, but leaves him/her to choose who should answer it, e.g. Now you ask a question, Ann.
(viii) The teacher has the students form a chain, e.g. A asks B a question, B asks C the same question, C asks D, and so on.
f. Varying the method
Mary was invited to use a number of the above techniques, provided she felt they were appropriate to her personal style, and the needs and temperament of her class. It was also suggested that some linguistic problems could be satisfactorily dealt with by one type of questioning, whereas another might benefit from questioning of several different varieties.
It was also pointed out that it might sometimes be appropriate to react to each student answer, while at other times a number of answers might be given before she, the teacher, gave any reaction.
If either of the above aspects of experimentation were used, it would be advisable to proceed systematically, if noticeable results were to be observed.

The experimental phase of the project
No pressure was put on Mary to expand her questioning techniques in all directions simultaneously. In the event, she made two major changes, one pedagogical, and the other didactic. She adopted the chain as a standard procedure for maximising the number of questions which were asked and answered in a given space of time, and she encouraged her students to prepare sets of general, personal questions and to put these to other students in the class in the form of interviews. It was impossible to record other, more subtle changes in her teaching, or to detect whether she used her habitual techniques more confidently, purposefully or systematically as a consequence of having reflected carefully about this aspect of her work. Of far greater importance for other teachers who may undertake similar interventionist experiments is the assistance that Mary had in detecting changes in her students as she made changes in her teaching.
The period of experimentation lasted about ten weeks, from late October to mid February. Each week, Mary sent back to York her evaluation of the effects of her teaching, described in a number of precise ways. These evaluations were returned immediately after the lessons, so that trends and movements only became apparent subsequently.

Analysis of questioning by polarity scale
One cannot be precise about language teaching methods and their effects; however, extreme statements are likely to attract from observers general agreement or disagreement. Accordingly, Mary was provided with a number of statements, arranged in extreme opposites, and invited to describe the experimental part of her lesson as tending towards one or the other extreme. These polarities are given below.
1. By answer type
a. Teacher allows short, Teacher requires full,
abbreviated answers. complete sentences.
b. Teacher uses questions to elicit Teacher uses questions
grammatical structures, requiring the communicative
exclusively. use of the language, exclusively.
2. By method of use
c. Teacher asks all the questions. Students ask all the questions.
d. Teacher achieves full coverage Many or most students do
of the class; each student not participate in the lesson.
participates.
e. Teacher uses one question at a Teacher uses each question
time, thoroughly, before once only.
continuing to the next.
f. Teacher uses one technique Teacher uses great technical
of questioning only. variety.
g. Teacher’s use of questions Teacher’s use of questions is
is systematic and thorough, random and unsystematic.
gradually increasing the
cognitive and linguistic
demands on the students.
3. By the observable effects of questioning
h. Teacher uses very difficult Teacher uses very easy
questions exclusively; students’ questions exclusively; students’
answers exhibit many errors answers are fluent and accurate.
and hesitations.
i. Lesson goes at a fast pace; Lesson goes at a very slow pace;
teacher obtains many student teacher obtains very few
responses. student responses.

Descriptors of oral practice
The above polarity scales were combined in a number of ways, shown below, partly because the process of language teaching is so complex that it would be unrealistic to extract any given discrete feature for examination, and partly in order to emphasise the inter-relatedness of all on-going processes in a lesson. It would be tempting to argue that these descriptors also show cause and effect, but here the researcher must be wary, since cause and effect, while not being excluded, cannot be proved. Of greater importance is that the same descriptor, if applied a number of times in succession, can show changes of process taking place, enabling teachers not only to reflect on their teaching with a degree of precision, but also to plan the course of future teaching with some accuracy.
Diagram 1
Comparison of question-programming with students’ fluency and willingness to participate

[INSERT a/w Peck: Diagram 1]

Result 1
The vertical axis is a measure of the care with which Mary constructed a battery of questioning techniques. It is, of course, subjective and global. The direction of the arrow is upwards, and this indicates that, in her opinion, as she became familiar with the various techniques of using questions, she deployed them increasingly systematically, bringing several to bear in each lesson in a principled sequence.
The horizontal axis enables pupils’ willingness to participate and their fluency to be described. Since the direction of the arrow is from left to right, this shows that pupils’ participation became more willing and fluent as the experiment progressed.
Diagram 2
Comparison between intensity of question use, and the pace of the lesson (including student participation)
The result shown by the second descriptor is startling.

[INSERT a/w Peck: Diagram 2]

Result 2
The diagram shows that at the beginning of term, Mary was putting a different question to each pupil, and that the pace of oral practice tended to be rather slow, with pupil participation being somewhat difficult to obtain. As the experiment proceeded, and as Mary began to put the same question(s) to a number of pupils (2b, and her own technique), she noted that the pace of the lesson increased significantly, as did pupils’ contributions.
Diagram 3
Comparison of technical variety and incidence of error

[INSERT a/w Peck: Diagram 3]

Result 3
The horizontal direction of the first arrow, from left to right, indicates that Mary noticed the performance of her class was, as a whole, improving in accuracy. However, since she shows no upward movement along the vertical axis from the first observation to the second, the decline of mistakes has to be accounted for by factors other than technical variety.
The direction of the arrow between the second and third observation is simultaneously upwards and to the right. This shows that a further decline in the incidence of mistakes coincided with Mary deliberately focusing a greater number of questioning techniques than before, on a single grammatical area. One should be wary, however, of seeing cause and effect here, if only because the initial improvement in accuracy was unaccompanied by a conscious methodological change.
Diagram 4
Comparison of who asks the questions, and coverage of the class

[INSERT a/w Peck: Diagram 4]

Result 4
The above diagram shows that to begin with Mary asked all the questions herself, as many teachers do. Her coverage of the class nevertheless increased significantly during the early part of the project. However, the increasing coverage and participation, indicated by the horizontal line, shows that it was caused by factors other than who asked the questions. In other words, the diagram shows no cause and effect. The descriptor shows additionally, however, that as the experiment progressed, she relinquished control of the questioning, and passed it increasingly into the hands of the pupils themselves, using the techniques described in Part 6, and that the high degree of coverage and participation was maintained.

Conclusion
This experiment shows that a teacher was able to intervene in her habitual method of teaching in order to make principled changes of technique, and that the results of so doing could be described. She was able to accommodate these changes within her established, personal style of teaching.
Two factors enabled this to happen: the existence of precise descriptions of language teaching coming from observations of other teachers, and descriptive devices capable of being used by the teacher herself to record important aspects of her work in a reasonably accurate way.