A comparison of three different vocabulary size tests for testing lexical competence in an Efl context


Tezin Türü: Yüksek Lisans

Tezin Yürütüldüğü Kurum: Bursa Uludağ Üniversitesi, EĞİTİM BİLİMLERİ ENSTİTÜSÜ, Türkiye

Tezin Onay Tarihi: 2019

Tezin Dili: İngilizce

Öğrenci: Sezen Aksu Balamur

Danışman: LEVENT UZUN

Özet:

In this study, three receptive vocabulary size tests in similar designs to the Yes/No Test (Y/N Test) (Meara, 1992), Vocabulary Size Test (VST) (Nation & Beglar, 2007a, 2007b) and Vocabulary Levels Test (VLT) (Nation, 2001) and covering the first 5K words in bilingual format were used to measure English receptive vocabulary knowledge of university preparatory class students. These tests, though being in different formats, are assumed to be measuring receptive knowledge of vocabulary at the same level and treated as equivalent. However, there does not exist a study which uses these tests all together and handles the matter whether they measure receptive word knowledge at the same level, so this study aimed to contribute to the field by filling this gap. Towards this aim, the study questioned whether the three tests, as assumed, can estimate overall receptive vocabulary size offering similar vii statistical figures a) in different proficiency levels, b) across test sections and c) in different frequency bands for different proficiency levels groups. Beside this, how well the tests correlate with each other and which one correlates best with the participants’ university preparatory class exit scores and yields consistent results were investigated. The data were collected from elementary, pre-intermediate and intermediate level 581 students studying at two different universities in Turkey and statistically analysed through the SPSS 22.0 packet program. The tests were administered in two sessions. The findings put forth that in general, the Y/N Test provided the lowest scores in almost all frequency and proficiency based comparisons. The VLT and VST, though, presented diverse scores for different proficiency groups and at different frequency bands. According to the statistical analyses, all mean scores were in line with the participants’ level of proficiency, which meant higher level students did better in all frequency levels, and the scores declined accordingly as participants proceeded to lower frequency bands. In addition, the highest correlation was found between the VST and VLT, and lastly, the most correlated test with the students’ university preparatory class exit scores was the VLT. All these results are important in terms of creating awareness in EFL teachers about the fact that there are different tests which they can substitute for institutional exams to measure students’ vocabulary knowledge, and that they can use these tests in different contexts according to students’ individual differences.