Doctoral thesis: Task Difficulty in Semi-direct Speaking Tests

Abstract

This dissertation discusses the effect of intra-task variations in association with performance conditions on task difficulty in a semi-direct speaking test, namely, the GEPT-I. The variables identified for investigation are linguistic demand of task input in relation to code complexity; amount of time allowed for performance in relation to communicative demand; type of pre-task planning in relation to communicative demand; and test-takers’ familiarity with the non-verbal propositional content of the task input in relation to cognitive complexity.

The dissertation takes a multi-dimensional approach in measuring the effects of the variables in a comparison of data collected from the performances in a controlled task and in an experimental task. The dimensions of the comparison are based on three different sources of data, including task scores, test-takers’ responses to the post-task questionnaires eliciting views of difficulty, and interlanguage measures in the areas of accuracy, fluency, complexity, and lexical density. In addition, learners’ proficiency in English is treated as a moderator variable in order to investigate the extent to which learners’ English proficiency interacts with the effects.

A study to establish the parallel nature of the tasks to be used in the study was first carried out to demonstrate quantitatively and qualitatively that the controlled tasks and the experimental tasks were equivalent before the manipulations of the variables were made. 239 Taiwanese learners participated in the main studies and their performance was analysed. The results indicate that the variation in performance conditions altered the degree of difficulty as measured in part or all of the dimensions of the comparisons. The results also confirm that the effect was altered in some cases due to learners’ proficiency in English; in general the more proficient learners were found to react to the variable more strongly than the less proficient learners.

Significant implications for language testing, learning and research in general are suggested. In particular, recommendations in relation to reliability (i.e., for parallel tasks/forms and inter-/intra-raters) and content validity are made to exam boards.

留言

這個網誌中的熱門文章

何謂測驗信度、效度？ What is Test Reliability and Validity?

原文刊載於中華民國 91 年 3 月 23 日《中央日報．英語教與學》國內近來掀起了一股英語能力檢測的熱潮，各種英語測驗紛紛出籠，如全民英檢、托福、愛普、多益、劍橋認證等。這些測驗各自標榜特色，有的強調測驗簡單易考、有的則是強調含聽、讀、說、寫四項測驗的全方位英語能力評量，讓大家真是眼花撩亂。而對有興趣要報考英語測驗的人來說，更是難以做選擇。其實選擇英語測驗就像選購商品一樣，除了功能、價格等因素外，最重要的就是品質了。賣毛衣的商家為了說明商品的品質良好，會說所賣的毛衣絕對是純羊毛做的，而多數人大概也知道如何判斷羊毛衣的真假。相對於賣毛衣的商家，辦理英語測驗的機構會以測驗具有「信度」與「效度」來說明測驗的品質，專業的測驗機構甚至會提出些數據加以補充說明。但是很多人卻連「信度」與「效度」這兩個名詞的意義是甚麼都弄不清楚，更別提判斷其所言的真假了。事實上，「信度」與「效度」是測驗理論的術語，一般人較感陌生。作者從事「全民英檢」的研發工作，認為測驗單位應有責任提供測驗的使用者（ stakeholders ）有關「信度」與「效度」的資訊，協助大家了解，以便在選擇採用英語測驗時做出正確的判斷。壹、「信度」（ reliability ）信度是指測驗分數可靠的程度，也就是這一測驗受信賴的程度。而一測驗為甚麼會受到信賴，關鍵在於結果的一致。同一位考生在能力沒有變化的前提下，在不同時間或不同的測試狀況下重複受測，其所得的分數應該是一致的，否則就產生測驗的誤差。一個測驗有誤差是難免的，但是當誤差過大時就影響了測驗的公平性了。測驗理論上有一個基本假定：實得分數等於真實分數加上誤差（ X=T+E ），但是真實分數是一個未知數。例如甲生考了某英語閱讀測驗的 A 卷得到 80 分，一天後考了 B 卷得了 82 分。雖然有 2 分的差距，但這是可被接受的誤差值，顯示該測驗結果的一致性頗高。又例如乙生考了某寫作測驗，閱卷老師 A 給 60 分，閱卷老師 B 給 62 分，這個結果顯示評分標準相當一致，而測驗的信度自然就高。總之，實得分數與真實分數愈接近即表示誤差愈小，測驗的分數就愈能代表考生的能力，如此，測驗的可信度也就愈高。貳、「效度」（ validity ）測驗效度即指測驗分數的正確性，簡單的說，就是指一測驗是否評量到它所要評量的...

閱讀完整內容

「全民英檢」初級寫作能力測驗

原文刊載於中華民國 92 年 1 月 3 日《中央日報．英語教與學》「全民英檢」五級的測驗當中，以初級（相當國中畢的英語文程度）的報考人數最多，佔 40 萬全部報考人數的一半。初級的通過率也高於其他級數，以其初試（含聽、讀、寫三項）而言，即有四成二的考生達通過標準（聽、讀兩項均達 80 分及寫作達 70 分）。但是，事實上考生在聽、讀部分表現不錯，近六成的考生其實在這兩項上已達通過標準。換言之，其中有近兩成考生是因為寫作成績未盡理想才未能通過初試，實在是令人感到惋惜。新（ 92 ）年度第一次初級測驗即將於明日（ l 月 4 日）舉行，己有五萬多人報考。在測驗前夕，為協助報考者做好準備，特別藉這篇文章說明初級寫作的測驗方式，並就多數考生感到較困難的「單句寫作」部分的題型、評分標準等提出較詳細的說明。壹、測驗方式「全民英檢」初級之寫作能力指標訂為：「能寫簡單的句子及段落，如寫明信片、便條、賀卡及填表格等。對一般日常生活相關的事物，能以簡短的文字敘述或說明」。命題即根據上述能力指標設計，測驗時間共 40 分鐘，全部採直接寫作，以引導寫作（ guided writing ）方式為原則。各部分答題說明皆以中文書寫，試卷與答案卷分開，題目全部印在試卷上。該測驗共分兩部份（見下表）：第一部分單句寫作（包含句子改寫、句子合併、重組三種題型）和第二部分段落寫作。初級寫作測驗題型與題數分配分項測驗題型題數第一部分單句寫作 A. 句子改寫 5 題（ 1~5 題） B. 句子合併 5 題（ 6~10 題） C. 重組 5 題（ 11~15 題）第二部分段落寫作看圖寫作 1 篇（約 50 字的段落）貳、單句寫作題型說明一、句子改寫：試卷上有一英文單句或簡短對話及一個未完成的句子，考生須依題目之提示將原句改寫成指定型式，並將整句謄寫於答案卷中。這個部分通常測試的重點包括改變時態、直接問句改間接問句、肯定句改否定句或問句等。下面的例題所測試的就是評量考生是否會將肯定句改寫為問句，並且會正確的使用過去式助動詞 did 。例題： Mary went to the market. When___________...

閱讀完整內容

Rethinking the use of CEFR in Taiwan: What’s next? Jessica Wu The CEFR has been ‘borrowed’ by the Ministry of Education in Taiwan since 2005, aiming to establish a common yardstick against which learners’ English proficiency can be measured. Having been mandated in Taiwan for a decade, the CEFR has had several positive impacts as well as difficulties and problems in its application in the local education context (Chang, 2012; Wu, 2012). Therefore, it is necessary to rethink the use of the CEFR in Taiwan. The paper suggests that the CEFR has been overtly used in testing and assessment in Taiwan, but its significance in promoting transparency and coherence in language teaching and empowering autonomous learning has been overlooked. The paper argues that the current asymmetric application of the CEFR in assessment should be improved to achieve better balance between learning and assessment. While reflecting on our current use of the CEFR, a review of practices in simila...

閱讀完整內容

語言測驗．評量．教學

搜尋此網誌