No abstract

No abstract

on a general language proficiency test 5.3 Tasks used in the experiment 77 5.3.1 The concrete picture description task (task I) 5.3.2 The oral interview (task IV) 79 5.3.3 The story retell task (task III) 80 5.3.4 The abstract picture description task (task II) 82 5.4 Retrospective tasks 83 5.5 The experimentation 84 5.6 Processing of the data 86 The Identification of Compensatory Strategies 88 6.1 The criteria and the identification procedure adopted in the 88 Nijmegen project 6.2 Identification on the basis of problem indicators in the data 6.2.1 The use of problem indicators 6.2.2 Instructions 6.2.3 The correspondence between the two judges 6.3 Identification on the basis of retrospective comments 94 6.3.1 The reliability of retrospective data 94 6.3.2 The influence of "researcher bias" 6.3.3 The usefulness of retrospective data 6.4 A quantitative comparison of two methods of identification 6.5 Conclusion The Classification of Compensatory Strategies 7.1 The coding system 7.2 The applicability of the coding system 7.3 Conclusion The Use of Compensatory Strategies in Tasks Ι, ΠΙ and IV 114 8.1 A quantitative analysis of the data 8.1.1 The number of compensatory strategies used 8.1.2 Analysis of compensatory strategy types: frequencies 8.1.3 Analysis of compensatory strategy types in terms of ANOVA 8.1.4 A visualization of the results: correspondence analysis 8.2 A qualitative analysis of the data 8.2.1 The use of compensatory strategies in the picture description task (task I) 8.2.2 The use of compensatory strategies in the story retell task (task III) 8.2.3 The use of compensatory strategies in the interview (task IV) 8.2.4 Individual differences 8.3 Discussion 8.3.1 Proficiency-level effects 8.3.2 Task effects 8.3.3 Differences between super-and subordinate compensatory strategies 8.3.4 Individual differences 8.4 Conclusion 9 A Comparison of Referential Strategies in LI and L2 9.1 Method 9.2 Hypotheses 9.3 Some procedural information 9.4 Results 9.4.1 A general overview of strategy use in Dutch and English 160 9.4.2 Identical strategies versus shifts 161 9.4.3 Time and number of words 9.5 Discussion and conclusion 10 The Effectiveness of Compensatory Strategies 10.1 Research to date 10.2 Some pilot studies 10.3 The York study of effectiveness 10.3.1 Method 10.3.2 Subjects 10.3.3 Results of the guessing task and the pseudo-cloze task 10.3.4 Two experimental tasks compared 179 10.4 Conclusion 11 Conclusion 11.1 Evaluation of the experimental design 11.2 Some theoretical considerations 11.3 General discussion of the results 11.4 Implications of the Nijmegen project 190 11.4.1 Second language use 11.4.2 Foreign language pedagogy 190 11.

The study described in this paper was set up to investigate the effect of foreign language learners' proficiency level on compensatory strategies used by these learners to solve lexical problems. At the same time, the effect of task‐related factors on compensatory strategies was examined. The study involved three groups of Dutch learners of English at three different proficiency levels. The subjects were tested on three different tasks: a picture naming/description task, a story retell task, and an oral interview with a native speaker of English. It appeared that “proficiency level” is inversely related to the number of compensatory strategies used by the subjects: the most advanced subjects used fewer compensatory strategies than did the least proficient ones. Contrary to our expectations, however, the type of compensatory strategy chosen by the subjects was not to any large extent related to their proficiency level. Rather, the data indicate that task‐related factors play a large role in this respect. Whereas the subjects predominantly used analytic strategies in the picture naming/description task, they frequently resorted to holistic strategies and transfer strategies in the story retell task and the oral interview. To explain these differences it is suggested that in selecting compensatory strategies the subjects observed general conversational principles.

It has been common practice to classify communication strategies (CmS) by means of taxonomies which are largely product-oriented. In such taxonomies different types of achievement strategies (also known as compensatory strategies (CpS)) are distinguished on the basis of the resources (source language, target language, gestures) which are used to encode the strategy, and the linguistic structure in which the strategy is couched. In this paper it will be argued that these taxonomies are inadequate for a number of practical and theoretical reasons. As an alternative, a process-oriented approach towards the classification and study of CpS will be proposed. This approach distinguishes between two basic strategy types only, conceptual and linguistic. It will be demonstrated that the choice between these two strategies is largely constrained by the nature of the experimental task and, to a much smaller extent, by the subjects' foreign language proficiency level. It is expected that a systematic study of these constraints in terms of the process-oriented taxonomy described here will increase our ability to explain and predict CpS use.