On Power Relation in the Design of Language Research Project and the Analysis of Data

How to deal with the relationship between the researcher and the ‘researched’ is crucial thing in design of language research project and the analysis of data. The paper mainly focuses the influence of power relations between the researcher and the ‘researched’ people. To do social research ‘on, for and with’ subject does not only benefit the researched, but also benefits the researcher. In the empowering research, the relationship between the researcher and the researched is like a part of process of social interaction and like the relationships among the researched.


Introduction
'Any social researcher who has undertaken fieldwork must at some level be aware that power relations exist in this context as in others' (Cameron et al, 1997:145).
It is generally believed that the relationship between the researcher and the 'researched' is odds relationship.The researcher always controls the research from the beginning to the end, moreover, acquire what she/he needs.Is this the correct way to do social research?In the 1990s, Cameron andher colleagues (1992, 1997) point out there are three kinds of relationships between the researcher and the 'researched': ethical, advocacy and empowerment.To understand these three kinds of relationships, one needs to know the three key words: on, for and with.The ethical model is to do research on, the advocacy is to do research on and for, and empowering research is on, for and with (Cameron et al, 1997).Before we explain these three kinds of relationships, we need to define the 'researched'.Generally speaking, the 'researched' means the people who the researcher studies and the researched subject which the researcher studies.The people who the researchers research can be divided into two kinds: the informant and the participant.The informants, like its literal meaning, are the people who only provide the material related to the research subject.The participants refer to the people who not only offer the related information but also participate in the research.
According to my understanding, in ethical research, the relationship between the researcher and the informants (here people only offer the information.) is an imbalanced one.The informants play a passive role in the research project.The researcher is the only beneficiary.However, the advocacy model shows that the informants not only provide the researched material but benefit from the research, i.e. 'it places the researcher in the position of an advocate, engaging in research not only on social subjects, as in the ethical framework, but also for them' (Cameron et al, 1997:148).Comparing the third model with the first two models, the major difference between them is that the participants are part of the major difference is the participants who are part of the research in the empowering research, like what Cameron et al (1992: 56) stated 'this involves more than the researcher entering into dialogue in order to find out local points of view-it entails informants taking part in the construction of theories about their own experience'.I have mentioned above that the 'researched' also means the research subject.Normally, the researcher chooses his/her subject through observation from life, or from reading the related articles.As a social researcher she/he needs to prove his/her idea, i.e. hypothesis, via doing the research.When the researcher conducts the research project, sometimes the process of the research project will make the researcher change the subject due to some facts and other specific reasons.This paper will mainly focus the influence of power relations between the researcher and the 'researched' people via the design of a languages-related research project as well as the analysis of its findings.

Power relation on the design of research project
Research design is a process of 'translating a very general research aim or purpose into specific, concrete questions to which specific, concrete answer can be given' (Cohen et al, 2000:75).We need to think lots of particular questions over before designing the research project, for example, what kinds of data are required?How will the data be gathered?Who will undertake the research?Here we will mainly focus on why think about the relationships between the researcher and the researched and how do they affect?

Why consider the power relation?
The aim of the language research is to find some special language phenomena and their related problem.This kind of research project is conducted between/among people, i.e. the researcher and the researched.Generally speaking, the ideal relationship between the researcher and the researched is equal and honesty.However that ideal is hard to realize.The empowering research would be the best one, the researched is empowered not powerless.Moreover, 'often, the researched can exert power over researcher by virtue of what they know that researchers do not' (Cameron et al, 1997:156).This is not good, in my view, if the researched is too powerful, the research could be out of the control.It can not achieve the purpose of the social research.By the same token, if the researcher is too powerful, the result of the research will lose its objectivity.
Therefore, when we consider how to design the research project, we should pay more attention to the relationship between the researcher and the researched, because it will influence the result's validity and reliability of the result.Both validity and reliability is the key point to effective research, especial in quantitative and qualitative research.'It is suggested that reliability is a necessary but insufficient condition for validity in research; reliability is a necessary precondition of validity' If a piece of research is invalid and unreliability then it is worthless (Cohen et al, 2000:105).

The influential aspects in the design of research project
When we design a language-related research project, the following points should be taken into consideration.

Obtaining the participants consent
BAAL guidelines (1994) describe 'Relationships with informants should be founded on trust and openness.They should be informed about all aspects of research that might reasonably be expected to affect their willingness to participate.'This is very essential point for doing the research.Before we start our research project, the first thing is to get the participants real consent whether in oral form or written one.If the participants are children under sixteen years old, the researcher must obtain both children's and their parents' permission.In addition, the researcher should respect people's decision not to participate.
At the same time, the researcher should offer the participants some information about this research.That will help them to make a decision.But how much is it necessary to give them?How does the researcher tell them, direct way or indirect way?For example, if the researcher interests in people using profanity.Can they tell them directly?Therefore, personally speaking, there is hard to define.

Doing the research with the participants
As the social researcher, like Cameron et al point out (1992:23) we might let the participants to be active co-operators, i.e. the participants should not be treated as objects and their thought should be respected about the research by the researcher.Sometimes the researcher may share 'knowledge' about the procedures of research with the researched.

Considering the participants' age, gender, and characteristics
According to my own research in China, I found the participants' age, gender and characteristics also influence the relationship between the researcher and the researched.I will use my example to explain this point later.
In the anthropologist Charles Briggs' case (1986), some of his informants were senior to him, they treat him not like the researcher to answer his questions, but take the chance to educate him as a parental role (Cameron, 2001:146).The relationship between the researcher and the researched was changed due to the participants' age.'As Briggs point out, an illuminating analysis of the resulting data would need to take account of the fact that informants saw what they were doing not as "answering an expert's questions" but "educating a young and ignorant newcomer"' (Cameron, 2001:146).

Choosing the suitable place/environment
The place/ environment, where the research will be taken place, should not give the participants uncomfortable feeling while using interview or questionnaire to conduct the research.For example, if the place is the researchers' office, that will make the participants feel more pressure, while in a pub or club, which will be too noisy for interviewing or tape-recording.As the participants are also the co-operators of the research, the researching place can be decided by both of them, not one side.

Making sure the tools working
Here the tools mean the audio-recorder or even the video camera, etc.If the researchers use them during their research, it would be better to check first.It can avoid some problems during the research even when analyzing the results.
There are still some aspects influencing the design of the research project.I would like use my own example to explain why I think above five points are needed to be considering firstly.I had done a research about seeking the motivation of Chinese students to learn second language.I chose thirty male students and thirty female English-major students in my department from my university.Their ages are from eighteen years old to twenty-one years old, i.e. all of them were born in 1980s (At that time, China had been doing the reform and opening the gate).Some of them came from cities, some from rural areas.I chose questionnaires and interviews(in English) as means of conducting the research, because the questionnaires can collect a large number of data to do comparison easily, but it is hard to know the participants' reasons for choosing the answer.Considered about this points, I also chose the interviews, 'interviews enable participants -be they interviewers or interviewees -to discuss their interpretations of the world in which they live, and to express how they regard situations from their own point of view' ( Cohen et al, 2000:267).
With the help of my department, I told all of undergraduate students (450) in my department I need them to help me with a research project.I told them the aim of this research, and what they need to do as part of this research.Firstly, under their permissions, I let all of them to complete the questionnaire.Secondly, considering the students' age, gender, characteristics (extroversion and introversion) and family background, I chose sixty students to do the interviews under their consent.These students were from first year, second year and third year, because the fourth year students needed to do their teaching practice, I didn't chose them.The place for us to have an interview was chosen by the researched students and me.Their classroom as our working place, because they felt relaxed and comfortable in their classroom.But one thing I could not control was the noise from outside, and that made the quality of some recording data not very well.To make sure the data was corrected; I made notes and also used the recorder-pen to record interviews.The interview we did was kind of 'institutional talk', but the atmosphere made it looked like 'ordinary' talk.Because the good 'quality of data is dependent on the quality of the relationship build up between the interviewer and interviewee' (Arksey & Knight, 1999:101).So during the interviews, I asked questions and also answered some questions.I was not the complete dominator; I was also a participant like my students.After the interviews, I gave some gifts to my students for thanks.From this research, I found it was very interesting that the students' gender, character and family background had great influence with the interviews and also the results.For instance, one of girls from Qingdao, whose character was easy-going, she was a good co-operator during the interview.She not only answered my questions but also asked lots of related questions about our research subject.The interview with her was out of my control for a short while.

The influence of power relation on the analysis of research's findings
When we design the research project, we need to think the power relation between the researcher and the researched.Moreover, we also ask to consider the power relation between them during the process of the analysis of its finding.There are several sides affect the analysis research's findings.

'Confidentiality and anonymity'
The participants have the right to ask the researchers to keep their responses confidential for personal reasons.Their request should be respected.And the researcher also can deal with them by anonymity.Furthermore it will be good for asking the participants again after the researchers finish their process of data-collection, sometimes.'But it is important to let informants know that it is not always possible to conceal identities completely, and that anonymity can sometimes be compromised unintentionally' (BAAL guidelines, 1994).

Validity and reliability
Generally speaking, the research's results cannot be of hundred percent of validity and reliability because it is impossible.Still take my own example--interview, to achieve the greater validity is to reduce, to the greatest extent, the amount of bias.'The source of bias are the characteristics of the interviewer, the characteristics of the respondent, and the substantive content of the questions' (Cohen et al, 2000: 121).For reliability, one way to control it is to have a highly structured interview, i.e. have same modality, same words and same questions for every participant.

The researchers
When the researchers analyze their research's findings, can they identify their findings truthfully and completely?The researchers have their own aim about their research, something they want to prove it, something not.That is to say, the researchers themselves, personally speaking, including their character and gender, could influence the analysis of research's findings.Moreover, even the researchers realize this point; it is hard to avoid it, because the researchers themselves are human beings, in other words, the researchers' emotion must influence their estimation and they cannot be real objective person for their research.For example, I recorded the interviews during my research, transcribing would be the first stage of analysis and interpretation.'A transcription is supposed to be a full and faithful written/ graphic representation of spoken material' (Cameron, 2001:32).Obviously the researcher should write down every word they heard from their tape-recorded data.There is no standard way for transcribing talk, so different researchers work at hearing them accurately and represent them differently.Therefore, the researcher may analyze the data without complete objectivity.During my period of transcription, I transcribed twice and the result was not the same.I think the reason could be my listening ability and my impression for students.

The researched
There are two aspects about the researched as the researcher need to consider when they analyze the research findings.
The first point is the researched emotional factor, and the second one will be discuss later.
Every human being has the emotional period of highs or lows.Under the researched consent, sometimes the researched even cannot cooperate well with the researchers due to their physical or emotional limitations.To put another way, the researched will not understand the aim of the research in normal way.Even they will offer incorrect information.That will indirectly affect the result of research.

Feedback
BAAL guidelines (1994) indicate 'wherever possible, final project reports should be made available in an accessible form to informants and informants should have the right to comment on them.'As mentioned above, the researched are the co-operators with the research.Therefore, 'if knowledge is worth having, it is worth sharing' (Cameron et al, 1997:134); that is to say, the researcher should talk to the researched and share with them.It calls on 'feedback' technique, which is to express the findings to the researched in order to obtain more informed consent for the research subject about the researched or their community.It is a good way to continue the dialogue between the researcher and the researched (Cameron et al, 1997:157).
The way of taking feedback can be formal or informal.The researcher can hold a meeting to share his/her findings with the researched.Or the researched finds some questions via the research which he/she attended to ask the researchers.The researcher's answer can make the researched know the findings; in addition, their questions will illuminate the researcher's new thought.
At last, the feedback will be good information for the researchers when they consider publication and dissemination of the research findings (Cameron et al, 1992:42).

Conclusion
To do social research 'on, for and with' subject does not only benefit the researched, but also benefits the researcher.In the empowering research, the relationship between the researcher and the researched is like a part of process of social interaction and like the relationships among the researched.
Personally speaking, as a language teacher, what I deal with is all language and students.Thus it may be concluded that the empowering research is the suitable one to be adopted.To sum up, I would like use a number of guidelines formulated by Cameron et al (1992:128) to end my paper.
1) Ask questions that interest the researched group or are generated by them.
2) Be open about your agenda and negotiate at all stages.
3) Make the knowledge and perceptions of the researched group count; do not impose an 'expert' framework unthinkingly.4) On the other hand, share information and analytic tools; the group may reject them but it is wrong to assume from the outset they do not want to know.

5)
Present what you learn from research in such a way that the researched group will find it accessible.'