< Terug naar vorige pagina

Publicatie

It's all about words. Three empirical studies into the role of lexical knowledge and use in French listening and speaking tasks

Boek - Dissertatie

The Common European Framework of Reference for Languages (CEFR) is probably the most influential document for educational language policy in Europe. The language-neutral CEFR describes language tasks and linguistic competences for any European language and links them to six levels of language proficiency (from A1 to C2) with the explicit aim to make language courses and proficiency levels more comparable throughout the EU. However, in spite of its widespread use and popularity, the language-specific implementations of these six levels are theoretically and empirically insufficiently validated as they tend to be confined to teacher and expert judgments and intuitions and not to empirical data. In addition, claims regarding the relationship between the language tasks on the one hand and the language-specific linguistic competences on the other hand have rarely been investigated either. Finally, the CEFR seems to suffer from other limitations, such as the lack of an underlying theory of language proficiency. The question thus arises whether at present it is possible to reliably and validly assign a CEFR language proficiency level to a foreign language learner given these limitations. More precisely, this PhD-research aims to investigate the relationship between vocabulary knowledge and listening on the one hand and between vocabulary knowledge and vocabulary use and speaking on the other. Since vocabulary knowledge is a key predictor for predicting someone's language proficiency, the lexicon will be the central focus of the three empirical studies. The aim of the first study is to investigate how strong the relationship is between vocabulary knowledge and listening comprehension for French and English at the intermediate level (B1) (N = 199 English, N = 330 French). Further, this study investigated how much vocabulary is necessary for 'adequate' listening comprehension at the B1-level. This analysis is a replication study of Staehr (2009). The materials that are used are a receptive vocabulary knowledge test and a standardized listening comprehension test (PET for English, DELF for French). The results show that there is a strong correlation between learners' vocabulary knowledge and listening comprehension for English as well as for French. Moreover, the results show that around 1,000 words are necessary in English in order to come to adequate listening comprehension, whereas 2,000 words seem necessary in French. The second study investigates the relationship between receptive and productive vocabulary knowledge and learners' vocabulary use in speaking (N = 51). Therefore, a learner corpus was composed based upon speaking tasks linked to the CEFR and performed by L2 learners of French. Their language use was measured through (1) the number of words, the number of different words (types) and the number of lemmas, (2) the lexical profile of the words that are used, and (3) the lexical diversity of the used words. The results show that language learners who have a larger vocabulary knowledge, both receptive and productive, use more (different) words. Further, this study showed that a productive vocabulary knowledge test seems to be a better predictor of vocabulary use in spoken language than a receptive vocabulary knowledge test. The third study focuses on vocabulary use and speaking and investigates which lexical parameters could predict learners' and L1 speakers' vocabulary use in two dialogic speaking tasks (N = 27 L1, N = 51 L2). Indeed, the same speaking tasks as in the learner corpus were also performed by L1 speakers. Experts rated all the speaking performances. The lexical parameters that revealed to be important are the number of words (tokens, types and lemmas), the frequency profile of the words and their lexical diversity (D, HD-D). Further, this study also looked at the differences in vocabulary use between L1 and L2 learners, as well as the relationship between learners' scores on a receptive and a productive vocabulary knowledge test and the scores attributed to their speaking performance by experts. The results indicate that the lexical diversity of the spoken output has the greatest predictive value, followed by the number of types and the number of low-frequency lemmas used by a speaker. The results also show that both native speakers and L2 learners only use the 1,000 most frequent words in these situations. L1 speakers, however, use slightly more low-frequency words than L2 learners and their lexical output is more varied from a lexical viewpoint. The findings of this PhD project aimed at contributing to the development of theory by investigating both receptive and productive vocabulary knowledge, by focusing on dialogic tasks instead of monologic tasks and by targeting language learners at the end of secondary school, being at a (low-)intermediate level in a less studied language, viz., French. The findings in this PhD thesis have demonstrated that a strong relationship exists between vocabulary knowledge and both listening and vocabulary use in speaking. Moreover, the results have underlined the value of using a complex methodology with more than one vocabulary knowledge test, different lexical measures and human ratings.
Jaar van publicatie:2019
Toegankelijkheid:Open