KPG research
The Task Analysis Project
This project, delivering analyses of the tasks included in the KPG exams, aims at the collection of essential data for evaluation, on an ongoing basis, of the KPG specifications and the leveled descriptors therein. It draws on descriptions of the KPG tasks, for each language, exam period, and proficiency level, including metadata related to the task type (e.g. multiple choice, matching, etc.), the language skill tested (comprehension, language awareness, language production, mediation), and the (written or spoken) texts associated with the tasks (i.e. source texts to which candidates are asked to respond and target texts that candidates are asked to produce, where applicable). In the theory of language underlying the KPG exams, the notion of 'text' is understood as the material configuration of the communicative context in which language functions. KPG texts are thus described in terms of bundles of features specifying their genre and function. These features include the topic and domain to which a text pertains (e.g. entertainment, travel, sport, etc.); its discourse environment (e.g. newspaper, magazine, webpage, encyclopedia, dictionary, etc.) incorporating distinctions regarding the communicative roles of the author and addressee; the text type (e.g. article, announcement, report, advertisement, email, etc.), identified in terms of the structure and the communicative purpose(s) of the text; the text process (description, narration, explanation, argumentation, instruction) by means of which the communicative purpose is fulfilled. The values that individual genre features may take are listed in comprehensive taxonomies included in the KPG specifications. A treatment of genres based on bundles of features has significant advantages over their conception as atomic categories. The model that has been implemented enables the classification of texts in terms of well-defined criteria designating coherent categories. Such a classification is readily extensible to new categories while it supports the sort of analyses carried out by the Language Learner Profile Project, focusing on how linguistic choices in texts produced by KPG candidates correlate with specific aspects of their communicative contexts.