(Gardner, 1983) at issue.
Q.3 Is there a significant relation between the trainees’ personality type (extroversion-introversion) and their performance improvement in simultaneous interpreting?
This question, too, includes two further sub-questions: one relating to extroversion and the other introversion.
1.7 Research Hypotheses
To help answer the above-mentioned questions statistically, the following three null hypotheses were formulated:
H0.1 There is no significant relation between the trainer’s application of interpreter-training-specific techniques and the trainees’ performance improvement in simultaneous interpreting.
H0.2 There is no significant relation between the trainees’ multiple intelligences and their performance improvement in simultaneous interpreting.
As explained in the previous section on questions, this hypothesis too needs to be further divided into eight separate ones each pertaining to one of the eight different types of intelligence as theorized by Gardner (1983).
H0.3 There is no significant relation between the trainees’ personality type (extroversion-introversion) and their performance improvement in simultaneous interpreting.
Similarly, this hypothesis is to be divided into two sub-hypotheses each relating to one personality type.
1.8 Theoretical Framework
Chesterman and Williams (2002, p. 48) have observed that:
Any research makes use of a theoretical model of the object being studied, either explicitly or implicitly. So if we are studying translation, or the translating process, we need some preliminary model of this kind in order to orient ourselves, to give ourselves an initial framework within which we can begin to think.
They then move on to propose a threefold classification of the theoretical models traditionally used in Translation Studies: “comparative, process, and causal models” (Chesterman & Williams, 2002, p. 49). According to this classification, the present research falls under the last, which is the causal, model since we will be trying to establish some kind of cause-effect relationship.
On a larger scale, it could be said that the present work is conducted within the framework of ‘cognitivism’ since we are looking at interpreting as a cognitive struggle demanding certain mental capabilities (cf. Lambert, 1988; Daró, 1994; Fabbro & Gran, 1994; Klonowicz, 1994; Lambert, 1994; Gile, 1995; Déjean Le Féal, 1997; Al-Khanji et al., 2000; Hamers et al., 2002; Lee, 2002; Mizuno, 2005; Pöchhacker, 2005; Seeber & Zelger, 2007).
Among the most important approaches emerging in the 1980s and the 1990s were those adopted by Lambert (1988) and Gile (1995) both within
[…] the paradigm that described and analyzed interpreting as a cognitive process. Lambert focused on the necessity of developing the component skills in an interpreter education program, and Gile has dealt with the difficulties of processing capacity management. (Shaw et al., 2004, p. 73)
The theoretical discussions and conceptual elaborations in the present study are mostly within the framework of Gile’s (1995) ‘Effort Model’ of interpreting. And the pedagogical approach adopted falls within Lambert’s framework of interpreter education which is centered on development of the component skills.
In discussions pertaining to the second question of the study, i.e. the possible relation between one’s multiple intelligences and their simultaneous interpreting potentials, we will be discussing things within the framework of ‘Multiple Intelligences Theory’ as proposed by Howard Gardner (1983).
1.9 Limitations and Delimitations
To carry out experimental research per se, regardless of the topic, the treatment, the participants, the experiment time period, etc. means that one is faced with numerous obstacles to overcome from the very beginning of the research project, particularly when it is in a society where research has not yet come to receive the recognition it deserves; neither from the general public nor even from the scholarly people at times.
Prior to embarking on an experimental research project, the main issue one has to bear in mind is that of feasibility: scarcity of resources required, be it money, time, space, equipment, subjects willing to cooperate, etc. have more often than not prevented keen researchers from either launching an experimental research project in the first place or pursuing it along as initially designed, hence distorting validity, reliability, or generalizability of the findings. This can very well explain why at times reliability or fruitfulness of experimental research designs is questioned. Amongst the arguments against experimental research designs are also unnaturalness of the settings as well as the presence and intervention of the researcher, which all work against validity and reliability of research findings.
Monti et al. (2005) maintain that the body of knowledge on interpreting has been increasing steadily through observational and experimental studies, based on case-studies or limited samples of data, which have produced insightful results.
Touching on two main types of empirical investigations, Riccardi (2005) compares and contrasts case studies and experimental studies. By case studies, she means “those studies whose objective is the description or evaluation of consecutive or simultaneous interpretations when they occur in their ‘natural’ environment” (Riccardi, 2005, p. 759). In such studies, one starts with a thorough description of the event and then embarks on analyzing it in detail. While these studies ay help validate, confute or question existing theories and may point to new research ground, a major drawback is that they cannot be replicated nor can the results be generalized. Experimental studies, the term experimental referring to the artificial, laboratory conditions in which they are carried out, enjoy the advantage that the variables are chosen and kept under control by the researcher and specific hypotheses may be tested on a sample of subjects. Therefore replication and generalizability are possible. However, “the greatest drawback of experimental studies is their lack of ecological validity; the communicative setting is missing or at least many variables are not present” (Riccardi, 2005, p. 759).
Even so, the necessity of conducting experimental research is far from questionable. There are obviously certain research areas which fall outside the scope of other research designs (descriptive, conceptual, etc.). There are given questions that other research designs fall short of answering. Therefore, in spite of all the shortcomings and difficulties involved in conducting experimental research, there is no denying that they are inevitable if light is to be shed on certain issues. Research projects with a didactic approach, like the present study, typically call for an experimental design. What researchers should do is to take measures to counter the effects of the shortcomings and remove the barriers so that, to the extent possible, feasibility conditions are met and reliability is ensured.
The first, and normally the greatest, problem in most experimental studies is finding participants who meet the conditions and who are willing to take part in a research project. We had to find over one hundred subjects since we could anticipate that the fatality rate (subjects who would be left out in the course of the experiment) due to numerous reasons (see chapter 3) would be high. This was specifically important because the experiment was designed to last for one academic year (two semesters). We had to make sure that enough subjects (at least sixty) would remain in the study. As to the subjects to be placed in the experimental group, there was not a big problem since I was going to teach three classes doing ‘interpretation 2’ (and ‘interpretation 3’ in the subsequent semester) at Allameh Tabataba’i University. The treatment was the teaching and the SI posttest was to be taken as their final exam. The extra things they had to do were a G.E. test, an SI pretest, and two tests of MI and personality type. As they were mostly keen students, the majority were willing to participate.
As for the control group subjects, I spent a long time trying to find appropriate subjects. Finally a colleague of mine told me about a friend who was teaching three classes of ‘interpretation 2’ (and ‘interpretation 3’ in the subsequent semester) at Karaj Azad University. Arrangements were made and as she herself was a Ph.D. candidate, she was understanding and kind enough to cooperate wholeheartedly throughout the experiment. To make sure that the subjects took things seriously, it was decided that these steps would be considered as part of their course assignments.
The other challenging difficulty was administration of the SI tests (pretest and posttest). The tests had to be administered individually in a private room without any sort of distraction. This was such a time-consuming process and oftentimes finding a suitable room was very difficult.
Evaluation of the SI sessions posed another challenge. One rater’s judgment would not suffice; at least three raters were needed. It was a most challenging task to find three qualified raters (experienced interpreters and/or interpreter trainers) who would be willing to spend a great deal of time rating one hundred and forty different SI sessions each lasting for five minutes (a total of 700 minutes of SI equaling some 12 hours!). From among many professionals who were called upon to score the SI sessions, two raters, both practicing interpreters one working for Press TV and the other in Iran’s Ministry of Foreign Affairs, eventually agreed to cooperate. In spite of the fact that due to validity considerations, it is undesirable, the researcher himself had to act as one of the three judges since finding a third