• 沒有找到結果。

This chapter discusses the research design of this study

N/A
N/A
Protected

Academic year: 2021

Share "This chapter discusses the research design of this study"

Copied!
11
0
0

加載中.... (立即查看全文)

全文

(1)

Chapter Four Methodology

This study starts off with a brief review of the history of interpreting as to when and how it was done. The present study hopes to sort out important facts and clues from past works for research direction on the basis of the literature review so accumulative debates and empirical studies regarding directionality are an important part of examination for any particular point of interest related to the subject of working into A as a standard. This study is also interested in revealing the actual extensiveness of working into A as acknowledged and put into practice in the worldwide field of interpreting today thus a supplemented qualitative questionnaire survey (the Internet survey) to either support or refute the research question that this study has come to according to the literature review. This chapter discusses the research design of this study. It starts with a section on the research framework, research method and research instrument employed by this study.

4.1 Research Scope

Working into A and into B each has its advantages and drawbacks and directionality has long been an issue of debate over the years. As reviewed in Chapter Three, working into A has its distinct advantages as claimed, however, strict practice of this direction may not be possible in various parts of the world due to the market demand for AB retour. In addition, whether working into A indeed assumes the standard role as it is seemingly claimed in the literature is another point of interest. To answer these questions, this study intends to gather and evaluate the opinions and facts from members in the field regarding the issue of AB retour. Data collected from targeted groups are analyzed and compared for results. The objective is to unveil

(2)

either any pattern or inconsistency, if there is one, regarding the actual practices and views about interpreting direction.

4.2 Research Question

This study attempts to gather and evaluate opinions and facts of practice from members in the field concerning the issue of directionality. Based on the literature review in Chapter Two and Chapter Three, this study assumes that:

l Working into A as a standard may be a claim well-noted but not as many actually support or practice it.

4.3 Research Method

Literature review and a qualitative survey were this study’s research method while the former takes on a heavier portion than the latter. According to Neuman (2003), qualitative research often depends upon “interpretive or critical social science” and qualitative researchers focus on “detailed examination of cases” and often present “authentic interpretations that are sensitive to specific social historical contexts.” (p. 139) The research tool adopted in this study is analysis of secondary data (literature) and questionnaire. Survey research can be applied to various sizes of groups. Its aim is to reveal facts or phenomenon among the targeted group/sample instead of focusing on any particular case study. It also generates the highest external validity among all types of empirical research (Chang 2004).

This study adopts sample survey. The way to conduct sample survey is to select several representative samples out of the population through either random or

(3)

non-random sampling method in order to reveal certain facts or attitudes. The followings are some of the reasons for this study to adopt sample survey according to Chang (2004, P. 375):

l Low cost and less demand of human resources

l Research time is considerably reduced due to fewer subjects involved and the results are valid in terms of timeliness.

l Quality research data also due to fewer subjects involved.

In terms of types of survey research, four general categories are identified:

questionnaire survey, Internet survey, visiting survey and telephone survey (Chang 2004). This study makes use of Internet survey to administer questionnaires to the respondents through the convenience of electronic mail. According to Chang, an Internet survey can be conducted through ways such as setting up personal Web server, BBS and electronic mail (p. 280). Chang further specifies that Internet survey poses the advantages such as (1) the lowest cost, (2) The fastest collection of data which is already in the form of electronic files and requires no additional manual work for key-in, (3) the widest sample distribution which virtually covers all over the world and (4) the best anonymity of respondent’s identity (p. 381). The one drawback of Internet survey as Chang cautions is that it may have problems in its sample selection; for examp le, the sample may be composed of Internet surfers who may not be located again for additional questioning.

(4)

4.4 Administration of the Questionnaire

This study makes use of several strategies to encourage response rate. First, there are only three questions contained in the questionnaires for all respondents; the respondent is only required to check one of the provided responses for each question although some questions may also require a subsequent short answer. Instruction as to how to make response and definition of the key terms are stated on top of the questionnaire. Other precautions are also taken to encourage the return of the questionnaires such as stating the identity of the research institute and offering incentives (Chang 2004). In the email (the introduction letter), the respondent is informed that the survey is concerned with a thesis research. On the first page of the questionnaire, the research institute, in this case the name of the researcher and the affiliation, are specified. What is more, the names and professional titles of the research supervisors are both listed clearly on the top of the questionnaire to raise a sense of trust out of the respondent. The researcher also offers to send the survey results via email upon respondent’s request as an additional incentive to increase response rate. A list of all other invited respondents is also attached to the email (except for the pretest) so that the respondent knows who else is involved in this worldwide survey and that this is a chance in which their voice can be heard. Yet perhaps the most effective way to encourage response rate, as Chang (2004) suggests and this study certainly applies, is to foster a sense of altruism by informing the respondents that their participation is an important contribution to the research. While Chang (2004) also cautions that it is difficult to control who actually fill out the questionnaire in a mail survey, all questionnaires in this study are sent straight to each and every respondent’s personal email box instead of having someone else (i.e. the department’s secretary, etc.) to forward it to the targeted respondent. The results

(5)

showed that all returned questionnaires were directly replied through the personal email boxes of the targeted respondents as originally sent. Note that some of the questions may pose a problem to respondents who lack the knowledge to reply. This means that the respondent does not know the answer “but someone in the selected household does”. One of the solutions to deal with such problem is to “identify and interview the household member who is best informed.” (Fowler 1993, p. 87; Chang 2004)

To fulfill research confidentiality and anonymity, the respondents in this study cannot be linked to their responses in any ways and the respondents are assured of it

“as almost all surveys promises.” (Fowler 1993, p. 90) The respondents are also informed to state their answers in details so that the researcher can better avoid having to contact the respondent twice for of an ambiguous answer they made the first time.

4.5 The Respondents and the Selection

In order to obtain the required data, this study intends to send out questionnaires and collect data form those who are in the position to reply. The selected respondents come from three different populations: 1). Directors of interpreting schools/institutes, 2). President/Director of related organizations and 3) conference interpreters. The selected respondents are the key persons who have the knowledge and are in the position to answer the questions. The focus is not how many respondents are involved in this survey but who and what they represent. It is not the quantity but the quality of the responses of those who have the knowledge to reply that is of top importance in order to build up the linkage between the facts gathered from literature in this study and the real world practice.

(6)

The questionnaires go out to a total of 17 different countries across four continents. It is the hope of this study to invite more respondents from a wider geographical scope. Two key criteria are taken into account in respondent selection as follows:

1. Questionnaires must be sent to as many names via e-mail as obtained within the limited timeframe. For this study to remain neutral on the subject, the survey should include the respondents whose standpoints and practices are clearly against/not in favor of the research question.

2. For the rest of the invited respondents, their views and actual interpreting practices should remain unknown to the researcher until their response. In other words, this study attempts to avoid being put under the suspicion that the survey mostly invites those who do not support strictly interpreting into A as previously identified and thus favors the research question for results.

4.6 Questionnaire Design

The three questions contained in the questionnaire are based on the research question derived from the literature review in this study. Several important points are also taken into consideration in the design of questions as proper measures (Fowler 1993). First, inadequate wording should be avoided. For example, questions that are incomplete or those with improper optional wording or poor wording. This study therefore makes sure that all questions are formulated in clear and complete sentences with one question asking for one definite answer only. Next, in the attempt to ensure

“consistent meaning to all respondents”, the way to do it is to keep away from “poorly defined terms” (p. 74), thus key terms such as A language and B language are first

(7)

defined on top of the questionnaire using AIIC’s current definitions. The third is to be very careful with “The ‘Don’t Know’ option” (p.76); it means that sometimes respondents do not want to reply as required or “legitimately do not know” the answer (p. 76). The best way to approach the problem is to first provide a “standardized screening question” (p. 76) to all respondents and see if they are familiar with the subject in question. This study thus designs the screening question (Have you ever heard of the notion suggesting that interpreting into A is the standard practice?) as the first question in the questionnaire to make sure all respondents do have some knowledge about the argument to begin with. The three questions only vary slightly for the three intended groups and are indeed very similar in nature. The first question across the three groups is the same and it is the screening question as stated above.

The second question differs only in the way the respondent is addressed; for example, for the group of school directors, Question 2 is “Can (name of the school) put the interpreting direction as supported strictly into all related practices?” but for the presidents/directors of various organizations the same question is “Can (name of the organization) put the interpreting direction as supported strictly into all related practices?”. Question 2 requires all respondents to clearly state the reason(s) for their response. Again the wording in Question 3 is different except the difference is to a larger extent for the group of interpreters. For the school directors Question 3 is “Can (name of the school) put the interpreting direction as supported strictly into all related practices? (Please do not answer this question if the response you made in Question 2 was “We support interpreting in both directions” or “We do not take a side on this issue”) and for organization presidents/directors it is similar “Can (name of the organization) put the interpreting direction as supported strictly into all related practices? (Please do not answer this question if the response you made in Question 2

(8)

was “We support interpreting in both directions” or “We do not take a side on this issue”). For the interpreters participating in this survey, Question 3’s wording is “Do you insist on interpreting only in a certain language direction when interpreting assignments are referred to you?”, while basically Question 3 in all three groups aims to reveal if a strict practice of working into A is possible for the intended respondents.

A short answer question also came with Question 3 for all respondents and the only condition in which they were not required this attached question was if their indication in Question 2’s provided responses was either “We support interpreting in both directions” or “We do not take a side on this issue.” (See AppendixⅡfor the three versions of questionnaire)

In addition, this study insists on using only one language, English, for the questionnaire survey due to a concern raised by Fowler (1993). Respondents answering questionnaires in different languages may pose an “extreme challenge”

because “it is doubtful that adjectival rating scales are ever compatible across languages.” (p.77) Therefore, all the respondents answered the questionnaire in English as the questionnaire was formulated. Last but not least, Fowler mentioned

“Standardized expectations for type of response.” (p. 78) That is, it is essential to design good survey questions that refer to the same thing to all respondents who should be able to perceive an adequate response. It can be done through using close questions which provides the respondent a set of alternatives to choose as the answer to a question or through making use of open questions that are specific. This study uses the provided-response type of questions and the specific open questions for the survey as mentioned earlier. Two out of the three questions are attached with an open question (short answer question as mentioned earlier) which simply asks the

(9)

respondent to state the reason(s) for the response they made. The advantages of open questions are that they allow not only unanticipated answers from the respondents but also provide “more closely the real views of the respondent… ., who also like the opportunity to answer some questions in their own words.” In addition, when possible answers are more than numerous, the choice of replacing provided response with open questions is appropriate (Fowler 1993, p. 82), such as in our case when the respondent is asked to state the reason(s) for their response.

This study’s questionnaires are generated in the way which is also supported by Chang (2004) who stated three similar key points in the design of questionnaire content:

l Questions asked must be essential concerning the research question:

This means that the designated questions must be necessary and relevant to the research. Some question designing techniques include do not ask or paraphrase the same question twice and each question should pose a clear objective instead of “By the way” kind of inquiry.

l Question adequacy: To word the question adequately and reduce question complexity, etc.

l Questions must be to the respondent’s knowledge to reply: For example, does the respondent have the knowledge to answer the questions in the questionnaire?

Can they recall the required information in order to make reply, etc.?

(10)

4.7 Pretest

A pretest of the questionnaires was sent to the first respondent via email.

Slight changes in the questionnaire were made after it was sent to the first respondent who was not informed of these changes because they were of trivial significance, such as one typo and phrasing of a sentence as later discovered. Although the pretest did not include the questionnaires for the other two groups (only the group of interpreters), the functioning of the questions for the three groups of respondents was indeed very similar in nature except that wording was not exactly identical as explained earlier.

The first pretest respondent indicated that she was aware of the standard claim of working into A but she supported interpreti ng in both direction for reasons quoted below. The first pretest respondent was not required to answer Question 3 due to her response in Question 2.

“An interpreter must be able to interpret both into his/her A and B languages. The speaker almost always takes questions from and has some form of discussion with the audience.

It’d be senseless to require an interpreter to interpret only into his/her A or B language”

After examining the reply, slight revision of a question and phrasing were ma de; for example, requesting the respondent to provide the reason(s) for the response they made clearly in details. The researcher consulted the first respondent through telephone to know if there were any perceived problems from the questionnaire. The revised questionnaire was then e-mailed to a second pretest respondent. The second respondent replied the next day and telephone consulting with the second respondent was also conducted. The second respondent in the pretest had no knowledge of working into A as a standard whatsoever and indicated that she did not take a side on the issue for reasons as follows:

(11)

“Under the premise that only accurate enough interpretations are considered, it is my guess that if an interpreter is capable of interpreting accurately enough between a certain set of languages (namely the interpreter’s A language and B language) in whichever direction, his or her B language should be at least of a certain level, in which case he or she should be able to speak the B language adequately.

Therefore, if an interpreter chooses to interpret in one direction only, it may be more due to personal reasons or principles, rather than because of limitations of his or her language abilities.

Thus, in my opinion, choice of interpreting direction should be treated as personal preference, and nothing more.”

Both pretest respondents chose an answer in Question 2 that did not require them to answer Question 3 as indicated. To avoid the possibility that respondents might choose an answer because they knew by selecting this option the effort to answer the next question could be waived, such indication was erased from Question 2 and moved to Question 3. A short-answer question was deleted for similar reason.

The format of the questionnaire and wording were also modified as well.

4.8 Data Analysis

This study analyzes and compares the primary data contained in the collected questionnaires for results against the claim that working into A is a traditional standard in the interpreting field. All data are cross-compared to reveal if there is any general agreement or inconsistency in the respondents’ opinions as well as their actual practices as to how many support or refute the claim.

參考文獻

相關文件

This research is conducted with the method of action research, which is not only observes the changes of students’ creativity, but also studies the role of instructor, the

The second question in this paper is raised from the first question – the relationship between constructing Fo Guang Pure Land and the perspective of management beginning

‘Basic’ liberty entails the freedoms of conscience, association and expression as well as democratic rights; … Thus participants would be moved to affirm a two-part second

In this chapter, the results for each research question based on the data analysis were presented and discussed, including (a) the selection criteria on evaluating

The purpose of this study was to investigate if providing consumers with a general dietary guidance or impose a cash punishment policy on customers for not finishing their food on

不過母體血壓降低得太快可能導致胎盤 / 子宮血液灌 流量減少而危及胎兒生命。 投予 hydralazine

In order to measure students’ learning achievements, students in the two groups were tested with “The Chinese rhetoric question are (pre-test)” before teaching and “The

This study attempts to Question Answering, Intelligent Agents and Feedback technologies, the development of an online SQL learning system with automatic checking