Aspects on Learner-biased Classroom Observational Techniques

This paper was designed to explore the means in the field of observing classroom teaching/learning from both a general, and an “English as a Foreign Language” (EFL), viewpoint. The aim was to browse the relevant literature, and lead to consider observation tools which might serve to research in EFL. This paper summarizes the reading by surveying the field of classroom observation, and then proceeds to evaluate the likely usefulness of a number of selected observation instruments.


Introduction
This paper examines classroom observation as a research activity, with particular reference to observing the learning/teaching of English as a foreign language (EFL) in a specific context.The focus is on EFL because, at a later stage, the author of this paper intends to examine the experience of a group of Chinese students studying English.Although his investigation will be carried out mainly through interviews with the research population, the author would like to add perspective to the study by observing these learners and their teachers in action in real EFL classrooms.This will help verify or perhaps even contradict his interpretation of the interviews, and so serve to add some objectivity to the study.
The author of this paper has browsed interviewing techniques, and gained practical experience through a research paper of real-life interviews.However, he cannot claim previous in-depth knowledge of classroom observation as a research activity.Feeling that it could add a dimension to his proposed research, the author shall examine some of its strengths weaknesses here, as a first step in guiding him to decide on which kind of supplementary instruments might be considered for his purpose.A broad survey of classroom observation as a research activity will be employed, which will hopefully guide him in choosing appropriate observation instruments to pursue areas highlighted as worth investigating after carrying out pilot interviews with the research population.The choice of aspect to investigate will also be influenced by the experience and intuitions gained from some 35 years of learning/teaching English in China.

Research Traditions
In a survey of classroom research specifically related to language learning and teaching, Nunan (1989) refers to four different traditions (to which the author has appended an example of each): 1) Psychometric studies; e. g.Pilliner in Cohen & Manion (1985) 2) Interaction analysis; e. g.Wragg (1970) 3) Discourse analysis; e. g.Narushima (1993) 4) Ethnography e. g.Bailey (1983) In psychometric studies, the researcher investigates the effectiveness of particular methods, activities and techniques by measuring language gain on proficiency tests.In interaction analysis, researchers use systems and schemes for studying classroom behaviours and interactions.
Discourse analysis involves analytical schemes for the linguistic analysis of classroom interactions.
In ethnographic studies, the researcher observes, describes and interprets the classroom in ways similar to those employed by anthropologists when they study unfamiliar cultures and societies.(Nunan, 1989, p. 4) Psychometric studies concerning product or outcome are mainly "quantitative" in approach.Numerical measurement and statistical means are involved, which investigate the quantitative relationships between various classroom activities or behaviour and language achievement.This research might serve to predict trends, but cannot account for the complicated behaviour of individual human beings.
Interaction analysis is strongly influenced by sociology, whereby researchers use the methods of social investigations in classroom observation.Here, observation of the classroom and analysis of the interaction taking place there, serve to investigate social meanings and inferred classroom climate.In this approach, student behavior is regarded as being dependent on classroom atmosphere and on the interaction between teachers and learners.The focus is on the relationship between teachers and students, not on quantitative analysis.Research following this approach still employs some quantitative analysis but the stress is not on mechanical scoring.
Discourse analysis in classroom observation derives from a linguistic perspective, which focuses on the discourse of classroom interaction in structural-functional linguistic terms, not on the inferred social meaning.It is systematic, and includes a dimension for pedagogical aspects, content, speaker, and other functions.Although quantitative analysis is potentially useful in this approach, researchers tend to confine their attention to the appropriate pre-defined categories employed to interpret the discourse.
An ethnographic study derives from sociology and anthropology, which is widely accepted in classroom research.It attempts to explain behaviours from the idea of participants' different understandings, and in this sense might be regarded as an 'objective', qualitative approach.The diversity of purposes, practices, and locations explains why different styles of classroom observation have been developed, and why researchers may adopt a quantitative or qualitative approach (Wragg, 1994, p. 7).The quantitative/qualitative relationship is discussed below (2.2).
However, in reality the distinction between the four approaches outlined above may not always been as clear-cut as the categorization implies, and indeed, each of the four approaches mentioned above might be employed in combination for a particular piece of research,

Quantitative and Qualitative Approaches
Quantitative and qualitative research methodologies are mutually dependent, which offer the possibility of flexibility for researchers.The distinction usually made between quantitative and qualitative research is that the two approaches represent different ways of thinking about and understanding the world around us.The extent to which one is prepared to accept or reject particular methods-quantitative or qualitative-depends on one's view of world (Nunan, 1992, p.77).

Quantitative
In this century, the quantitative approach has been heavily influenced by the nineteenth-century French Philosopher Comte, who claimed that human thought evolved through the stages, theological, metaphysical and scientific.(Wragg, 1994, p. 7) The belief was that social behaviour could be predicted through systematic observation and analysis.Quantitative researchers are interested in facts and their relationship, and in details that can be measured to produce generalizable results.This research often makes use of statistical analysis: in this case it needs to be broad if the results are to be statistically valid (Bell, 1992, p.27).This approach predominated in educational research in the period from 1900 to 1930, and employs methods such as statistical studies, survey studies and experimental studies.
A quantitative approach is generally regarded as being obtrusive, controlled, objective, and product-oriented, and tends to be large scale and time-consuming.It tends to gloss over the complicated nature of individuals.

Qualitative
Compared with a quantitative approach, a qualitative approach tends to be narrow in scale, and focus on individual case studies.It is a process-oriented approach to the study of interaction (Chaudron, 1988, p. 48).Qualitative researchers are interested in individuals and their view of the world.The following types of qualitative approach are often used: symbolic interactionism and phenomenology, and social and cultural anthropology.The qualitative approach used in education research tends to focus on teacher's classroom strategies and learners' adaptations to school, patterns of classroom interaction, learner's perspectives and classroom behaviour, transfer between school, teachers' life histories, and the impact of public examinations on classroom teaching.

Making an Initial Choice of Approach
As the main thrust of the author's research will be conducted qualitatively through interviews, the author shall seek to complement this by obtaining more quantitative data for analysis.
The author of this paper does not intend to pursue a psychometric approach which tends merely to measure learning outcomes, because the process and environment of learning is of more importance in the context of his research.
Carrying out discourse analysis of the English classroom interaction in which the Guangdong University of Foreign Studies (hereafter referred to as GDUFS) learners are involved, is another avenue which does not recommend itself to his purpose.There are practical considerations in recording verbal discourse in the language classroom.Unless the researcher is adept at shorthand, then cassette recordings will have to be made.But teachers are not always happy to have their performances recorded, for they have only the researcher's guarantee that there will subsequently be no breach of confidentiality.Technical equipment can easily break down, especially if the researcher is unfamiliar with it.Inadequate acoustics in normal classrooms can result in recordings that are not clear enough to be transcribed; this applies particularly to group work.However, the most serious drawback is the time taken to transcribe a recording in communicative classrooms where there tends to be a great deal of verbal interaction.A one-hour class can take up to 20 hours to transcribe, and that is before any analysis takes place.This is the time that the author cannot afford, because he will already have his main interviews with the learners to produce as written texts.
The author is also obliged to rule out a supplementary ethnographic study.Thus, by a process of elimination, he has taken the decision to pursue the interaction analysis tradition in supplementing the data obtained through interviewing.This will be carried out through some form of classroom observation which will help me match that data to the reality.However, before doing so, he shall take a realistic look at the strengths and weakness of classroom observation as a research activity.

Classroom Observation
Direct observation would seem to be an obvious and straightforward activity in researching classrooms.The observer sits in on a series of lessons, records what goes on in them, and then analyses the information gathered.The observer sits in either as a participant, taking part in the process, or as a non-participant, observing the action in a detached way.Because of the difficulties of becoming a genuine participant to which he has referred in the author's discussion of ethnographic studies (above), he intends to focus on non-participant observation.However, the author's reading in this area has shown him that classroom observation is far from being an "objective" activity.This does not deny its value, but is an important fact to bear in mind when reading research reports deriving from data gathered through non-participant classroom observation.The author's attention has been drawn in particular to a chapter by Rees (1997) which adopts an awareness-raising-though not destructive-stance in this respect.Below the author edits and summaries some of the points made there.

1) narrow focus
Teaching as a profession, involves very much more than teaching.When we observe actual classrooms, we do not take into account important out-of-class activities which could also contribute to learning.Rees (op.cit.) mentions: planning, reading, homework, student profiling, exam writing, pastoral work and other extra-curricular activities, etc. and he adds other roles of the teacher such as being a friend, disciplinarian, and instilling values.

2) sampling
The question of sampling is a tricky one.Just how many classrooms does an observer need to visit to discover the general characteristics of any one learning environment.

3) variables
What is observed influences the observation, but in reality is often subject to variables which are not under the control of the teacher or the class, such as time of day, day of the week, size of the class, temperature, character of the previous lesson, and so on, all of which can influence the learning/teaching.

4) beyond shared knowledge
There may be knowledge shared by the learners and teacher which is not known to the observer, but may nevertheless influence his/her interpretation of what is seen and/or heard.

5) continuity
Unless the observer has the time and permission to observe any one class relentlessly, then it is difficult to ascertain how one lesson fits with those that precede and those that come after it.Teaching/learning are continuous, long-term processes which cannot always be detected by short-term observations.

6) perspective
There is a natural tendency for observers to concentrate on the role of the teacher, at the expense of that played by the learner, hence "teacher observation" is much more frequently heard than "learner observation".This neglects the equally important role of the learner in the language classroom.Classrooms where real language learning is taking place, are not necessarily characterised by constant teacher intervention.

7) goal
What to observe and why is problematic in the observation of teaching.Rees (personal conversation, 1999) regards this as the Achilles' heel of all classroom observation.There must be good reasons for choosing what to observe, founded on current knowledge of good practice.At the same time, it is wise to consider that-particularly in the history of language teaching'-what is approved of today may be condemned tomorrow.Rees (op.cit.) quotes a memorable extract from Cook (1994) in this respect.
In TEFL, yesterday's criminals become today's respectable citizens with such regularity that it seems almost certain that in this endless alternation, what is outlawed today will be eulogised tomorrow.A gambler would find TEFL a very easy field.The question of what to observe is closely related to the following two factors 1) the wood and the trees: A deliberate decision is usually made by observers to look at aspects of the classroom in analytical detail or in broader perspective.This is usually a matter of purpose, and convenience, and not of right or wrong.However, it should always be borne in mind that individual aspects of teaching examined should never be claimed to represent the whole, and that the whole can easily be lost in a forest of detail; 2) high and low inference: Decisions also have to be made in advance concerning concentrating on high and low inference factors (or a mixture of both) when observing classrooms.Factors which require a low degree of inference from the observer, such as the number of times the teacher moved to the back of the room during the lesson, can be recorded with some certainty, though they all too easily focus on trivia.Important factors, where high inference is required, such as exactly how much is being learned, cannot usually be recorded with such assurance.

1) the good language teacher
Any observation instrument which tries to identify the characteristics of the good language teacher, fails to acknowledge that some teachers teach differently from others with equal success, and that what succeeds in one context may prove inadequate in another.The teacher's performance in isolation cannot guarantee successful language learning.

2) the fragile observer
Observation requires a surprising self-discipline from the observer, who has to remain alert throughout even the most boring of lessons.Not every observer can display this stamina.

3) the egocentric observer
A trap which the observer can easily fall into is to assume that what is interesting/boring for him/her is also interesting/boring for the learners; and equally to assume that what is interesting must also be useful in the language learning context.This is a fallacy.Indeed, the communicative approach, wrongly interpreted, often leads to amusing sessions with no real language learning content at all.

4) the observer as interloper
A common complaint is that the presence of an observer threatens to change the nature of the class being observed, and so defeats the purpose of the observation.Quoting from his own experience, Rees (op.cit.) maintains that this is exaggerated.If the observer learns to be unobtrusive, e. g. by sitting quietly in the back of the classroom, not shuffling papers, nor establishing eye-contact, nor interfering with the lesson in any way, then the learners, with their backs to him, quickly forget that he or she is there.However, the teacher faces the observer, but this one difficulty can to some extent be dealt with by establishing good rapport with him/her beforehand, if possible.Where good rapport has not been established, or the purpose of the observation made dear, then this can lead to very uncharacteristic teacher performances ranging from the spectacular to the inhibited.

5) frames of reference
A problem in using ready-made instruments for observing classrooms is that the person who made them may not share the same frames of reference about teaching/learning as someone else wishing to use them.This explains why ready-made instruments often have to be adapted for use in different contexts if they are to produce meaningful outcomes.

6) tunnel vision
Because of the complexity of what goes on in language classrooms, the observer is obliged to observe only selected aspects of it.The limitations of human vision and hearing, and natural lapses in attention mean that it would be impossible to observe everything.Even watching a videotape of a lesson replayed for a second or third time on a small screen will reveal aspects missed on previous viewings.Of course, learners and teachers too are subject to these same constraints, but this fact reminds us of what a piecemeal activity classroom observation is, and so helps us to keep it in perspective.

7) subjectivity
All classroom observation is by its very nature subjective.Even checklists which require the observer merely to tick off low inference factors can be subject to observer fatigue, temporary loss of attention, and so on.And subjectivity is even involved in compiling any checklist in the first place.This does not mean that classroom observation should be abandoned, for truths can emerge when tackled from different directions.As Bowers (1989, p. 144) reminds us, there is substantial evidence to suggest that no one observational technique is in itself adequate: all techniques have their strengths and their weaknesses.Use of a range of techniques can help to cancel out the weaknesses of each while capitalizing on their strengths.
It is for this very reason that the author shall be using several techniques in his dissertation research.He ends this section with an apt quotation from Bowers which partly summarises the above section: "Whatever you see of a teacher's classroom activity can only be indicative: you will never see enough to know with certainty what kind of teacher he or she is, how representative the sample which you have seen may be of their overall competence and preferences.Moreover, even in what you do see, there is much which remains below the surface: you will observe what the teacher does, but not what the teacher perceives as happening (which may be different).You will not know how what you see ties in with other events (before or after) which add up to the full history of the relationship between this teacher and these pupils."(1989, p.142)

Sources
The author has undertaken a realistic look of the capabilities of observation instruments in classroom research, the next task will be to choose suitable instruments for his purpose, and to adapt them where required.It is astonishing to see that there are a number of instruments that had been published.Those in the classic Simon and Boyer (1975) and its British counterpart Galton (1978) run to over a hundred.He clearly needed to limit his research, and considered the following for ideas (for full details of texts, see the Bibliography): The author also examined a number of individual instruments from various sources shown to him by his colleagues.As these were merely a loose collection which had not been consistently referenced, he is able to present below only the details available to him: 1) Categories for the Puckett system (Puckett, late 1920s) 2) Coding lesson segments (Rees, 1984) 3) Grid for taking field notes (Rees, 1984) 4) Involvement learning in small groups (Wragg, 1994) 5) Language learning questionnaire (Nolasco & Arthur, 1988) 6) Learners' attitudes to the learning process (Forth, 1990) 7) My classroom environment (William Burden, 1999) 8) Observation schedule (Candlin) 9) Observation vocabulary teaching (Shahinda Modnis) 10) Pupil observation (Partington & Luker, 1984, p.48) 11) Pupils' questionnaire on writing (Chzung, 1993) 12) Questions/answers (Rees, 1984) 13) Small group discussion (Turney et al., 1982) 14) The good English teacher (Ministry of Education, Malaysia) 15) What is currently food language teaching practice?(Rees,1984) 16) Your experience of language learning (Forth, 1990) 17) Your views on language teaching (White, 1984) Based on what the author already knew of their language learning experiences from earlier interviews with the GDUFS students (see 1. Introduction), at this stage the author earmarked the following instruments as initially worth considering: 1) Good & Brophy, J. ( 1978

The Instruments for Piloting
The next task was to choose which instruments the author should actually pilot, given the time and resources available to him.He was guided by the following considerations: 1) avoiding abusing the generosity of teaching staff, so needed instruments which he could operate with a minimum of fuss or distraction.This ruled out using a video or tape-recorder, or trying out too many instruments.
2) avoiding all instruments which seemed to directly evaluate teachers' performances.Instead, it was decided to concentrate more on the learners' classroom experience.
3) being sensible to choose instruments which would not require intensive training in order to operate them.

Piloting the Instruments
Before piloting the 6 instruments, the author drafted a letter to be distributed to the teaching staff, explaining his objective, and asking for permission to sit in on their classes.The author circulated this letter only to teachers involved with students within his research, as he hoped to get to know them through this exercise.This might facilitate his using these same instruments with them during his later research.The teaching staff he had approached all agreed to his request.In the event the author discovered that in some cases testing out the instrument did not require the full hour for each which he had anticipated.
Below, the author took each instrument in turn, and comment on his findings.A few of these instruments required some adaptation by him even before use.Where this was the case, it is stated.All the instruments will be found in Appendix.(Good & Brophy, 1978) With only 10 categories, the author found this to be a very easy instrument to operate.The categories are clear, but very general.This requires high inference from the observer, so it would probably benefit from sub-specification to make it more objective.For example, how is "dominating" to be defined, and what is counted as "teacher participation"?As this was a very small class of six students, there was of course less chance of any one student dominating than in a larger one.This practice suggests that instruments need to be adapted for the situation in which they are to be used, and that categories that are too general may increase subjectivity in rating them.As the situation in a Chinese language classroom is quite different, many facets may influence students who are learning a foreign language:

Small-group Interaction
1) The way of performing the students used to have in the middle school.Chinese students have to pass the entrance examination if they want to further their education in universities or colleges.The focus of classroom lecture is to obtain the knowledge they can, not stress on the performance, such as oral presentation.Therefore, lectures are usually textbook-centred.
2) Confucius educational thought which has influenced their view of learning.Chinese students are passive in classroom learning, and always expecting the teacher to instruct them.
3) the teaching methodologies to which they already got used to.Communicative approach in language learning/teaching has proven quite successful in the west, but it has not been popular or acceptable in Chinese.The situation is that students pay a lot of attention to grammar, vocabulary and reading.4) language competency.Furthermore, Chinese students often encounter the barriers, language competency, when they are required to present their ideas orally, they find it difficult to employ proper words.And they frequently hear such comment, 'not comprehensive' which subsequently discourages them to perform naturally in class.Therefore, domination, at this stage, seems far from the reality.
Besides the above points preventing the domination in a language class in China, the normal size of a language class consists of 25 to 30 students.Because of the passiveness, it is unlikely for anyone of them to dominate.(Partington & Luker, 1984) This instrument concentrates on the observation of a single student during a lesson.It is not therefore designed to generalise about the experience of a large number of students at a time.This attracted the author to the scheme, for it could help him to follow up the individual experience of any one student who would seem to merit personal observation as result of his earlier interviews with him or her.

Student Observation
The author tried out the instrument in part of a reading comprehension session at intermediate level, and concentrated on a teenage girl sitting beside the teacher.The author immediately realised that no space had been allotted on the form for him to write these identifying details which may help him do the research in depth.The author would have to add these if he were to use this instrument as part of his research.This subsequently applied to some of the other instruments he tried out.The instrument has 14 main categories which are straightforward and therefore manageable.It also has spaces for unpredicted" other" categories, which gives it some flexibility.Entering simple ticks for each category was not difficult.As he did not have access to the original source of the instrument, he was unclear about the reason for the repetition of "answering questions" which seemed, however, to be referring to writing rather than speaking.
The format of the instrument appealed to the author, especially as he could easily change categories in the "Activity" column, according to what he was looking for.However, though the instrument reveals what the learner is doing throughout the lesson, it does not indicate whether this is the activity expected of him or her by the teacher!For example, if expected to listen to a tape-recording, the student could be talking to another.This could be easily rectified by using a tick for expected participation and a cross for deviant behaviour.Of course, such deviant behaviour could still be contributing to language learning.This practice reminded the author that observation instruments could tell only part of the full story.In the process of a lecture, it sometimes happens that students discuss each other about the questions associated with the learning task, which is not assigned by the teacher.What kind of assessment can we give?The 'absent-mind-less' may be part the learning task.Compared with the lecture offered by western teachers, lectures given by the Chinese lecturer are well organised with strict discipline.Any students peaking in class would be regarded as an offensive behaviour if they discuss things without the approval of the lecturer.A normal language class in China allows one voice, either the teacher or one student, not many.(Rees, 1984) The author of this paper found this to be very boring instrument to operate, but this does not necessarily mean than it is not useful.The scheme looks fairly simple, which led him to no studying it carefully beforehand, so that he had problems during the observation with the interpretation of the category "Use of space".The author subsequently realised that it referred to general disposition of the teacher and learners in each 5-minute segment.To specify this in more detail, the author subsequently decided that it would have been better for him to have added grid reference numbers (rather than just the letters A-H) to the initial sketch, so that he might report, example, that after 10 minutes, the teacher moved from A2 to a position at G3.It is not easy to sum up the main categories in any one 5-minutes period, especially as this cannot be realistically done until the next 5-minute period is already under way.Filling in the 6 categories in longhand as a non-native speaker of English took more time than the author would have expected, and might have caused him difficulties in a fast-paced class.To use this time-sampling instrument effectively would need more practice than he had anticipated.

Grid for Taking Field Notes
This instrument tends to be the trap of subjectivity.When an observer practises it, he may find it hard to avoid subjectivity which may appear in his notes because before he takes notes he has to make the comment on a point which depends upon his own knowledge and understanding.So the subjectivity goes along with the observer's view of the class he/she attends.
The author felt that this instrument would be useful in helping him to make consistent field notes to supplement, for example, a tape-recording of a lesson, but that he would not really find a use for it in his research especially as he preferred some of the other piloted schemes.This trial showed that this instrument should be thoroughly familiar to the observer before using it, and that some instruments look deceptively simple to operate.In particular the author realised that the ability to time-sample does not come naturally, and that lessons do not divide themselves into neat segments for researchers.(Nunan, 1989b) The nature and composition of tasks in the language classroom have gained particular prominence in the era of communicative language teaching.As all language classes are composed of tasks, and as determine the language learner's experience, it seemed commonness to examine a task-based observation instrument.

Tasks in Language Learning
As the author does not intend to make cassette recordings of lessons for his research, every class he observes must be regarded as a one-off, with no opportunity for him to listen to it more than once.So any instrument that he uses must accurately record what he wants it to.The 17 items of Nunan's instrument made this very difficult for him to do.Some of the items, e.g."7.The activities are appropriate to the communicative goals of the task" are high inference, and not easy to assess.The author found that "4.The task reflects the nature of language and learning" and "10.The student and teacher roles are inherent in the task" too vague to assist him in the task of rating them.
The recording was complicated by the fact that tasks in the classroom do not necessarily occur in neat succession, but may suffer external interruptions or the insertion of sub-tasks.And it is far from easy to determine exactly what the task is.In this instance, there was a discussion of homework, which appears to score well on the categories.The reality, however, was somewhat different.The class was composed mainly of female students who are culturally reticent about speaking and playing a prominent role in class.The task was consequently dominated by a male student.So though the task was appropriate for the class, not all the students benefited from it.This instrument might be useful as the basis for a questionnaire to discover students' and teachers' views on what makes a good teaching task.For harmonious teaching, there should be some consensus on this between the two parties.Employing Nunan's instrument made the author realise that though a task in the classroom may be theoretically sound, it is the teacher's role to ensure that it is effective for as many learners as possible.
6.5 How New Words Are Practised (Wajnryb, 1992) This was revealing insofar as there were few occasions in the particular class observed where new vocabulary was being taught.In the few instances that occurred, however, the categories seemed to be viable.
The lesson learned here was the obvious one-the author should have checked beforehand that the instrument he intended to use was appropriate for the particular class.Unfortunately, this is not always possible, and in any case, one of the problems with any classroom observation is such unpredictability.
Another point should be mentioned, which has a very typical Chinese characteristic.The way of learning words in Chinese students is much different from the way which the native speakers acquire the new words.There are two ways for Chinese students' enlarging their vocabulary.1) In intensive reading, students obtain the new word by listening to lecturers, consulting dictionaries and other references.The focus of the process is on explaining the usage of the words, and the comparison between some synonyms as well.Examples of the words will be practised in the process.
2) Learning new words focuses more on the words themselves, less on their contextual meaning.
It takes a long time for the students to memorise the new words.They take pains memorising them mechanically, such as writing each word again and again at beginner's stage without fully understanding the meanings.' Questions/Answers (Richards, 1994) The author likes the graphic presentation of this instrument; it shows clearly, for example, that one student dominated the interaction.He did encounter some difficulty in operating it at speed, and would need some further practice before feeling competent in its use.This would be particularly the case in deciding at speed what is a "reflective" question.

Teacher's and Students
The author should have asked beforehand how many students were likely to be in the class, for he had made boxes for only 10, and needed to add 2 by hand during the interaction.An alternative would have been to make room for many more boxes at the outset by abbreviating the content to S1, S2 for Student one, Student two, and so on, and by quickly crossing out any superfluous boxes as soon as the class had settled down.
The completed instrument does not tell us why one student was so prominent, so a designated space would have to be added to the instrument for additional observer comment.This was in fact the class mentioned above where the dominated his female classmates.What the instrument does not tell us is that part of the problem was the female learners were hesitant in giving responses, and the teacher did not allow them sufficient wait-time.., but this could be the function of a different instrument at a different time.Using this instrument confirmed that any observation instrument cannot paint the whole picture, and needs to be supplemented with further information.And the categories of even the most obvious-looking of instrument must be thoroughly understood and mastered before use.
Another point cannot be neglected that the way of Chinese students' performance in class is always connected with passiveness, because they have been strongly influenced by Chinese Educational System.Students have been led by the examinations which are the key reference to the students who want to further their education at each stage.So it is not really necessary or compulsory for them to notice their own performance.Furthermore, the assessment teachers use is the examinations, no other devices such as oral presentation, group work, writing report or paper etc.

Conclusion
After piloting the instruments mentioned above, the author has to take some points concerning the practice in Chinese circumstances into consideration.
Considering the class size in China, it is not practical to operate the instruments employed in this paper.Class size in China is usually larger, 25 to 30 students in language classes.The size of class has been proved to have great influence on students' and teacher's interaction.
Certain grade students in smaller classes benefited in terms of improved performance on some courses (Word et al. 1990, p.16).As to the language class, the size of class makes great difference accordingly.Smaller size of language class will provide students with more chances to practise, which is very important at the early stage in learning a foreign language.
The interaction between students and teacher is different in small or large class.In addition, another facet also affects on the performance, as mentioned in the quotation of word.This is palpable especially in Chinese language class, the freshmen who enter the university usually show enthusiasm in classroom interaction, and their ebullience would ebb away when they are senior students.At this stage there is no difference how large or small the class size is.
The size of class matters when students are in their first two years study at university as the basic skills, which need interaction between teachers and students, are the stresses that always employ intercommunication.
The author has learned from this research that instruments for observing classroom tend to suggest that classes are much more organised and straightforward than they are in real life.In many ways they are rough tools, which can capture only part of the reality of the classroom.Common-sense categories such as "participating" often need sub-specification if they are to be interpreted meaningfully.Trying out these instruments showed the author how very general and subjective such everyday terms really are as they stand.
One interesting facet revealed by the piloting was the clear difference in the classroom between expected behaviour and actual behaviour; this was something he was always dimly aware of, but had never really brought to consciousness.
Published instruments may be tidied up versions of the original, and not leave spaces where identification details and observer comments can be listed.These may have to be added indeed, it seems that few instruments can be used in new situation without being adapted in some way, either to fit the different context, or to suit the observer's preferences.
It is important, too, to read the background information on any instrument.Picking up an instrument just because it looks interesting, without knowing the ideas on which it was compiled, can be misleading, and lead to doubts about interpreting some of the categories.In fact, it appears that it is unwise to use instruments "cold", for some previous practice or training is required if they are to be operated effortlessly.One must be fully conversant with them.
The author found time-sampling difficult, as it needs considerable skill to sum up quickly the main points of a previous interaction when another one is taking place.This is particularly difficult when summarizing in longhand, either in English or in Chinese symbols.In fact the author discovered that he was happier using instruments that require merely the ticking of categories.
The piloting exercise made the author realise that some advance knowledge of a class is useful to ensure that an appropriate observation instrument is being used.But one can never predict the unpredictability of what goes on in some classes.
The author also became aware that no observation instrument is perfect, for each has drawbacks in operation, and there is often the need to add supplementary handwritten comments, or to explore further with other instruments.
The author was pleasantly surprised to find that his presence in the classroom did not seem to seriously distract the learners; at the same time he discovered that recording what goes on in classrooms can be a mechanical and not a very exciting activity.The study has given him a practical and theoretical insight into classroom observation which he did not previously possess, and he feels that it has made him better equipped to select and operate observation instruments to supplement his research.