應用字彙能力測驗之驗證研究

Abstract

本研究旨在根據教育部公佈的基本1,200個字彙發展並驗證一個可用於測量國中生英語拼字知識的應用字彙能力測驗。本測驗採用多元計分方式來評鑑英語學習者的拼字能力，本研究更試圖去探究單字長度與單字難度之間的關係。受試者來自台北市都會區的一所國中共2,149人。資料分析使用部分得分模式(partial credit model)，研究結果顯示該應用字彙能力測驗所提供的測量結果適切，具有效度與信度(Rasch受試者信度 = .96)。雖然整體上該測驗對受試者的程度而言稍嫌簡單，不過試題仍然涵蓋了大部分受試者的能力水平。 本研究的發現指出單字長度與單字的困難度有中低度相關(r = .38)，不過，此相關度隨者單字長度縮短而減弱。經由分析高難度單字的拼字錯誤，結果顯示拼字難度可能與字音一致性及跨語言轉移現象密不可分。

The present study aims to develop and validate a productive vocabulary knowledge test (PVKT) for measuring junior high school English learners’ orthographic knowledge for productive use based on the Basic 1,200 Word List published by Ministry of Education. A partial credit scoring procedure was adopted to evaluate the spelling proficiency of the English learners. The study further attempts to explore the relationship between word length and word difficulty. A total of 2,149 students from a junior high school in the metropolitan area of Taipei, Taiwan, participated in the study. The partial credit model (PCM) was used to analyze the data. The results revealed that the PVKT was valid and reliable (Rasch person reliability = .96) in providing an appropriate measurement. Further, the test was generally slightly easier for the participants while the items still cover the major range of the participants’ ability level. The findings of this study suggest that word length is moderately correlated (r = .38) with word difficulty; however, the correlations weaken as word length shortens. Indications that spelling difficulty could also reside in issues of letter-sound correspondence and cross-lingual transfer were found in error analysis of the spelling of the most difficult words.