topic of assignment: ethnography. submitted by Umair Ijaz.
(roll Number 1 BS English 4th semester) submitted to: Sir Waseem Akhtar. date of submission: 12-06-2012.

OUTLINE
In this chapter, I shall define ethnography and describe its central characteristics and principles. I shall also look at the key research concepts of reliability and validity as they relate to ethnography, and will discuss the importance of context to ethnographic inquiry. In the final part of the chapter, I shall highlight some of the 'central concerns of this topic by contrasting psychometry and ethnography, The chapter seeks to address the following questions:
• - What do we mean by ethnography?
• - What are the key principles guiding ethnographic research?
• - How might one deal with threats to the reliability and validity of this type of research?
• - Why is context important to ethnographic research?
• - In what ways does ethnography contrast with psychometric research?

• Definition:
Ethnography involves the study of the culture/characteristics of a group to real-world rather than Laboratory settings. The researcher makes no act to isolate or manipulate the phenomena under investigation, and insight generalizations emerge from close contact with the data rather than from theory of language learning and use. it is a qualitative type of research. Ethnography is provided by LeCompte and Goetz (1982). They use ethnography shorthand term to encompass a range of qualitative methods including study research, field research, and anthropological research. LeCompte and Goetz argue that Ethnography is defined by the use of participant and non-participant observation, a focus on natural settings, use of the subjective views and belief systems of the participants in the research process to structure that research,and avoidance by the investigators of manipulating the study variables. Wilson (1982) identifies the roots of ethnography in anthropology and sociology although there is also a strong tradition in research into animal behaviour (see, for example, Martin and Bareson 1985). Chaudron (1988) identifies ethnography as one of the four major traditions in applied linguistic research, although he does not devote a great deal of his book to research carried out within this tradition. He characterizes ethnographic research as a qualitative, process-oriented approach to the investigation of interaction, and points out that it is a rigorous tradition in its own right, involving 'considerable training, continuous record keeping, extensive participatory involvement of the researcher in the classroom, and careful interpretation of the usually multifaceted data.
Watson and Ulichny (1988) identify several key principles of ethnographic research. These include the adoption of a grounded approach to data the use of 'thick' explanation, and going beyond description to analysis interpretation and explanation. They point out that ethnography involves interpretation, analysis, and explanation not just adoptation.
Explanation takes the form of grounded ; theory which, as we have seen, is theory based in and derived from data and arrived at through a systematic process of induction'(The most complete treatment of grounded theory is to be found in Glaser and Strauss Their two other key principles are 'holism' and 'thick' explanation.

Principles of ethnographic research
Ethnography has suffered somewhat from being applied rather loosely to any research that is not a formal experiment, giving rise, in some quarters, to the suspicion that the tradition and its practitioners lack rigour. However, as Chaudron (1988) and others have pointed out, true ethnography demands as much training, skill, and dedication as psychometric research.
Wilson relates the tradition to two sets of hypotheses about human behaviour. These are:
1. the naturalistic ecological hypothesis: The naturalistic ecological perspective has, as its central tenet, the belief that the context in which behaviour occurs has a significant influence on that behaviour. It fellows that if we want to find out about behaviour. we need to investigate it in the natural contexts in which it occurs, rather than in the experimental laboratory. Arguments in favour of field research as opposed to laboratory research are supported by studies of particular phenomena which come up with different findings according to whether the research is conducted in a laboratory or in the field. For example, Bellack, Hersen, and Turner (1978) found that subjects performed in a role-play situation very differently from the way they performed in real-life social situations where the same behaviours were observed. It has also been observed that parents and pupils respond differently to questions according to whether they are posed in school or at home.
The dilemma, as we saw in Chapter 1, is that In dealing with the problem of generalisability (an issue of external validity)and placing the research in the field, one increases the possibility of a threat to the internal validity of the research, because intervening variables may make it imoossible to ascribe a causal relation between the variables under investigation. However, not all ethnography is out to ascribe such causal relationship and so the problems which beset the quantitative researcher in a field setting become unimportant.
2. the qualitative-phenomenological hypothesis: The second hypothesis identified by Wilson is the qualitative phenomenological hypothesis. This principle also throws-ethnography into stark contrast with psychometry because it questions the belief that there is an objective reality which is independent of the subjective perceptions of researchers and their subject. Rather than subscribing to a belief in external 'truth', ethnographers believe that human behaviour cannot be understood without incorporating into the research the subjective perceptions and belief systems of those involved in the research,both as researchers and subjects.
The Psychometrician's belief that the task of research is to identify, describe, and explain external, objective reality has been imported into the social sciences from the natural sciences. Most people accept the notion of external reality and objective facts in relation co the natural world - we can See seeds germinating, flowers growing, day following night. For those brought up in a Western educational tradition, it also seems reasonable to assume that there are mechanisms and principles that similarly govern human behaviour and that it is the task of the researcher to identify those principles. It can therefore come as something of a shock to encounter writers such as Wilson questioning the idea of an objective reality, and suggesting that the methods, procedures, and assumptions governing the physical sciences may not be appropriate for investigating human behaviour. (it is interesting to note that at present in physics, the hardest of the hard sciences, the notion of objective reality is being questioned by some researchers).
Watson-Gegeo and Ulichny (1988) identify similar defining characteristics as Wilson, that is, the importance of context and subjective perception to the research enterprise. They highlight in Particular the contextual characteristic which focuses the research in real situations and settings where people actually live and work rather than in laboratory, or simulated settings. Within this context, the research focuses on the cultural meanings revealed by the behaviour of the subjects under study.

strong and weak view
Van Lier (1988) also identifies cultural description as a central of ethnography. He presents two views on ethnography, the 'weak view and the 'strong' view. The weak view, which according to van Lier currently prevalent in applied linguistics, sees ethnography as essentially inferior to psychometry, as it consists of unstructured and unsystematic observation. The principal virtue of ethnography, according to the weak view, is may throw up questions or hypotheses which can subsequently be test in a formal experiment. in other words, it is essentially a ground clearing ation rather than a valid tradition in its own right. The strong view, to which van Lier himself subscribes, sees ethnography as a valid research paradigm its own right: ethnography is theory building, and thus the core often humanistic approach to social science. in this sense it can be traced naturalistic approaches to social science. (vanLier 1988: 54)

Characteristics of ETHNOGRAPHIC Research characteristics: Gloss
• contextual: The research is carried out in the context in which the subjects normally live and work.
• unobrrusive: The researcher avoids manipulating the phenomena under investigation.
• longitudinal: The research is relatively long-term.
• collaborative: The research involves the participation of stakeholders other than the researcher.
• interpret: The researcher carries out interpretive analyses of the data.
• organic: There is interaction between questions/hypotheses and data collection/interpretation.

importance of context perhaps the only main thing which distinguishes ethnography from other research techniques is the "context". in ethnographic research, the researcher studies the phenomena in natural settings and not in artificial environment of laboratory. the investigation of language and learning by Heath, which is arguably the major contemporary ethnographic investigation of language development to have been since the early sixties, the researcher goes to the field and collects data from the field directly. she lived her life like the community members of the community under study. perhaps this thing make this research difficult and somewhat dubious because the person have to devote a long time as participant observer and sometimes he has to marry in that community and make family in order to win the confidence of the local members. long participation may alter the thoughts of the researcher but the researcher who is focussed and take precautionary measures can make his research objective. so, it would be pertinent to write that particular context is the pivot point of the ethnographic research.

The reliability and validity of ethnography
The major criticisms leveled at ethnography by proponents of quantitative research concern the reliability and validity of such research. Most of these criticisms stem from the fact that ethnographies are based on the detailed description and analysis of a particular context or situation. Because of the quantity of data yielded in these studies, it generally impossible to include anything but a small amount of the data in a published account of the research. This makes it difficult or outside either to analyze the data themselves or to replicate the study Whether or nor internal and external validity are problematic will depend on the scope of research, and the researcher's purposes. if the researcher is not attempt establish a causal relationship between variables the issue of internal validity will be less problematic than if such a relationship is being sought. The initial consideration in relation to external validity is to what extent fill from a study carried out in a particular site can be generalised.
LeCompte and Goetz deal firstly with reliability,which they define extent to which studies can be replicated. They argue ethnography may battle attempts at replication', in comparison with laboratory experiments. strategies against threats to reliability following are the strategies which can be applied to minimize the chance of errors. these strategies are the way in which the threats to reliability are minimized and they gave the research a certain path to its success. they are as follows: strategies against threats to external reliability External reliability, that is, the replication of the research enhanced if the ethnographer is explicit about key aspect of the research. These are:
1. the status of the researcher.
2. detailed description of subjects.
3. the choice of situations and conditions.
4. the analysis constructs and premises.
5. the methods of data collection and analysis.
Attending to researcher status position requires research about the social position they hold within the group. LeCompte and Goetz make the point that in one sense, I exactly replicate the finding of another because, even i context could be found, the second researcher is unlike same status in the second social situation. A related problem parallel informants in the second research context. This is of major proportions wherrit is considered that 'the extent edge is gathered is a function of who gives it' (LeCompte). It is therefore imperative for researchers to describe extremely carefully. The social situation and conditions need to be described explicitly. Definitions of constructs and premises are also crucial and social contexts of a prior study, replication may remain important.

Strategies against threats to internal reliability
Strategies identified by LeCompte and Goetz against threats to internal reliability include the use of low inference, multiple researchers, participant researchers, peer examination and mechanically recorded data. they are briefly explained as follows:
1. INFERENCE DESCRIPTORS
INFERENCE descriptor describes behaviour on which it is easy for indicate observer to agree. For example in classrooms "wait time' (the time waits between asking a question and then following up and use of factual questions would be examples behaviours. High inference behaviours, on the other hand,requiring the observer to make inferences about the observed behaviour. observer to infer unobservable mental states from behaviour. the problem here is that it is the high inference behaviour at the often of most interest.
2. MULTIPLE RESEARCHERS/PARTICIPANT RESEARCHERS the most effective way of guarding against threats to internal reliability is the use of multiple researchers. in much research this is not feasible, a research team consisting of several members can be extremely,particularly given the extended nature of much ethnographic. an alternative is to enlist the aid of local informants to validate the ethnographer. For example, an ethnographic study could involve teachers in reviewing and validating the researcher's conclusion.
3. PEER EXAMINATION
Peer examination involves the corroboration by other researchers working in similar settings. According to LeCompte and Goetz, this can proceed in three ways. Firstly, researchers may utilize outcomes and findings from other field workers in their report. Secondly, findings from studies carried out concurrently may be integrated into the report. This provides a form of cross-side validation. Finally, provided sufficient primary data are included in the published report, these may be used for reanalysis by the researcher colleagues.
4. MECHANICALLY RECORDED DATA
The final strategy researchers can employ to guard against threats in internal reliability is the use of mechanically recorded data (for example, in the form of audio or video recordings). This strategy allows for the preservation of the primary data. However, it must be remembered that these devices do not preserve all of the data, but only those data selected by the researcher for preservation.

some limitations
While the foregoing strategies are valuable, they may not all be practical for someone carrying out an ethnographic investigation with limited time and resources. The use of multiple sites and researchers in particular could prove to be extremely expensive. As 1 have already indicated, the use of low inference descriptors is also a problem, because high inference behaviours and constructs are often of greatest interest. (Ethnographies in language classrooms have investigated phenomena such as motivation, interest, power, authority, and control, all of which are high inference behaviours).

care and explicitness - mendatory for a successful ethnographic research
The over thrust of LeCompte and Goetz's suggestions for guarding against threats to reliability can be summarized in two words care and explicitness. Basically, what they are suggesting is that if one is careful in the collection and analysis of one's data, and if one is explicit about the way the data were collected and analysed, then one can reasonably claim reliability for one's investigation.

key points related to reliability here key points related to both internal and external reliability are given in terms of questions. if all these the researcher sattisfies all those conditions specified in these questions then his research will be more successful and less dubious. these questions are as follows:
Internal reliability:
Does the research utilize low inference descriptors?
Does it employ more than one researcher collaborator?
Does the researcher invite peer examination or cross-site corroboration?
Are data mechanically recorded?

External reliability:
Is the status of the researcher made explicit?
Does the researcher provide a detailed description of subjects?
Does the researcher provide a detailed description of the context and conditions under which the research was carried out?
Are constructs and premises explicitly defined?
Are data collection and analysis methods presented in detail?

difference between internal and external reliability Study has internal reliability if independent researchers, on analyzing the data, come to the same conclusions as the original investigators. Here the key difference is that external reliability refers to the replication of the original study, while internal reliability concerns the reanalysis of Original data by independent researchers. Such reanalysis is due to the fact that ethnographers rarely use standardized instruments such schedules. validity
LeCompte and Goetz, who argue that internal validity is one of the strengths of ethnographic research (a view questioned by a number of other researchers, including Beretta 1986a). Internal validity relates to the extent to which an investigation is actually measuring what it purports to measure. External validity, on the other hand, poses the question: To what extent can research outcomes be extended to be groups? Dealing with threats to the external validity of their research can be the most difficult methodological task confronting ethnographers. LeCompte and Goetz argue that the claim of ethnography to high internal validity derives from the data collection and analysis techniques employed:
First, the ethnographer's common practice of living among participants and collecting data from long periods provides opportunities for continual data analysis and comparison to refine constructs and to ensure the match between scientific categories and participant reality. Second, informant interviewing, a major ethnographic data source, necessarily is phrased more closely to the empirical categories of participants and is formed less abstractly than instruments used in other research designs. Third, participant observation, the ethnographer's second key source of data, is Conducted in natural settings that reflect the reality of the life experiences of participants more accurately than do contrived settings. Finally, ethnographic-analysis incorporates a process of researcher self-monitoring . that exposes all phases of the research activity to continual questioning and reevaluation. (p. 43)
For the researcher wishing to generalize beyond the context in which the data were collected, external validity is particularly problematic, as the procedures (such as the use of subjects randomly assigned to experimental and controlled conditions) are generally irrelevant. This external validity of ethnographic research is threatened by effects that reduce its comparability. key ppoints related to validity The researcher can guard against this threat by focusing on the key points. these key points are given as follows:

Internal validity :
Is it likely that maturational changes occurring during the course of the research will affect outcomes?
Is there bias in the selection of informants?
Is the growth or attrition of informants over time likely to outcomes?
Have alternative explanations for phenomena been rigorous examined and excluded?
External validity:
Are some phenomena unique to a particular group or site therefore non-comparable?
Are outcomes due in part to the presence of the researcher?
Are cross-group comparisons invalidated by unique historical experiences of particular groups?
To which extent are abstract terms and constructs shared all different groups and research sites.

facctors affecting the credibility of cross-group comparison LeCompte argue that the credibility of cross-group comparisons can be affected by particular factors: selection effects, setting effects, history effects,and construct effects.
Selection effects can have a bearing on the external validity of an ethnography if the constructs under investigation are specific to a single group because there has been a mismatch between the chosen group and structure for investigation. In discussing setting effects, LeCompte point our that the very act of investigating a group, culture, or setting have an effect which renders cross-group comparisons invalid. Then sion brings to mind Labov's (1972) 'observer's paradox' (Labov that the aim of sociolinguistic research is to find our how people behave they are not being psychometrically observed, but the data can be obtained through systematic observation). Cross-group comparisons may Also be considered invalid by the unique historical experiences of groups and culture. Finally, construct effects can refer either to the extent to which abstract and concepts are shared across different populations, or to the extent which explanations are regarded as valid across groups. main theme most researchers have drawn principally upon the work of LeCompte and Goetz to address the various factors which may threaten the external reliability and validity of ethnographic research. While the adoption in ethnographic approach to research can pose formidable problems of reliability and validity,these problems are not insurmountable, and there are practical steps one can take to guard against them, although these may involve additional time and resources.

integrated approach ethnography contrast markedly with the experimental method in its assumption,methods, and attitudes to evidence. In principle there is no reason why research programs should not integrate psychometric and ethnographic methods of investigation, and, indeed, several calls have already been made for such integration. In practice, however, integrated approaches seem to be almost non existant. I have already argued that this reflects the fact that two approaches are underpinned by very different conceptions of social reality ,truth and the nature of evidence. The different conceptions of 'truth', 'reality' and 'evidence' held by some language researcher is one reason for the growing attention being paid to the ethnographic techniques for gathering and analyzing language data. in finding alternatives to formal experiments has also been stimulated a skepticism over the ability of psychometry to produce definitive answer that some researcher expect,(Ellis 1990a: 67) Ellis advances two reason for this skepticism which are as follows:
1. in the first place, the relationship between instruction and learning is Extremely complex. it is not a linear relationship, and is no one-to-one relationship between teaching and learning. experimental research can therefore only provide us with an understanding of individual pieces of the language learning Jigsaw, but not the whole ppuzzle.
2. second, according to Ellis, class rooms do not exist simply to provide cannon for research. the relationship between findings from a formal experiment conducted under laboratory conditions, and class room practice is complex and indirect.
Action in the class room can never be just a question of implementing a condition derived from research. it is always a process of negotiation the teachers overall educational ideology, the learners expectations and preference and local constraints that determine what is feasible. There is no single pedagogical solution which is applicable to all classrooms.(Ellis 1990a: 68)

examples of ethnographic research some examples of ethnographic research are given below:
1. ethnographic study carried out in the sixties by Geoffrey (1968). The purpose of the study was to describe complexities in the contemporary urban classroom. The data collection procedure study was deceptively simple - Smith, a university-based researcher when trained as a psychologist, spent a term sitting in Geoffrey's classroom as participant observer. His database consisted of extensive field notes and he observed, as well as interviews with the researcher and students. These were used to construct a rich descriptive and interpretive picture of the complexities of an urban classroom in the sixties.
2. In language teaching, a similar investigation was carried out by Freeman in (1992). Freeman became a participant observer as a French in a foreign language classroom, and his database included lesson transcripts, field notes, interviews with the teacher and students. The analysis consists of discussed and interpretive work on the database. Freeman concludes from his investigation that:
The process of evolving a shared understanding of what to learn and how to learn is at the heart of what makes [the teacher's] classes Work. It takes place against backdrop of constant social interaction and is intimately tied to sharing authority and control. [The teacher] has been able to make public the process of creating and internalizing the language precisely because she allows the talk activity in her class to be largely self-regulated. Students come to control themselves in their interactions; that control-goes hand-in-hand with authority over.

CONCLUSIONethnography is an alternative research adition to psychometry. ethnography was underpinned by very different values and assumptions those of psychometry. Two beliefs in particular have guided this research tradition. These are the importance of context to human behaviour, and the centrality of the subjective belief systems of those involve research to the processes and outcomes of research.
Ethnographic research has sometimes been criticised for its failure to against threats to reliability and validity, and there is an unending discussion of the practical steps which can be taken to guard such threats. While reliability and validity are critical to psychometry, researchers have argued that they are nor necessarily the appropriate criteria against which ethnography should be judged. some researchers are of the view that research needs to be reliable. Internal validity is important in explaining rather than descriptive research (and is therefore important if one at Watson Geegeo and Ulichny's structure that ethnography should be explanatory as well as descriptive). External validity is only an issue for research wishing to make claims of generality beyond the research sTress where' data were collected. The ethnographic tradition was illustrated by a detailed summary research by Heath into language learning and use in rural communities i United States. this summary provides some of the types of questions which ethnographers ask, and the sorts of all in which they engage. In principle research method or methods one employs should be determined by the questions which one wishes to investigate, rather than by any adherence to one tradition rather than another. In practice, however, peer that a preference for one particular tradition determines the questions one considers worth asking in the first place. For anyone seriously interested in contextualised research into language learning and use, these two are the mmost important researches:
1. Heath's (1983) investigation of language and learning is essential reading.
2. LeCompte and Goetz (1982) provide an articulate and considered analysis of threats to the reliability and validity of ethnographic research, as well as suggestions on ways in which these threats may be overcome.