Vocabulary Size Test英文單字量測驗English Vocabulary Size Test
估計你大概認得多少個英文單字。免費、不用登入,手機或電腦都可以做。Estimate roughly how many English words you know. Free, no login needed, and it works on a phone or a computer.
題目載入失敗,請重新整理頁面再試一次。The questions failed to load. Please refresh the page and try again.
測驗方法與理論依據How the test works and the research behind it
- 依照單字出現的頻率分級。英文裡越常出現的字,通常越早學會。我們把英文單字依照在大量真實英文(書籍、新聞、網路、字幕)中出現的頻率排序,每 1,000 個字分成一級,共 10 級(10,000 字)。這是 Nation 與 Beglar 的 Vocabulary Size Test 所用的方法:從每一級抽出一些字來測,就能推估整級認得多少(Nation & Beglar, 2007)。一個字以基本字形計算,例如 run、runs、ran 算同一個字。Words are grouped by how often they appear. The more often a word shows up in English, the earlier it is usually learned. We rank English words by how often they appear in large amounts of real English (books, news, websites, subtitles) and put every 1,000 words into one level, 10 levels in all (10,000 words). This is the method of Nation and Beglar’s Vocabulary Size Test: by testing a sample from each level, we can estimate how many words you know in the whole level (Nation & Beglar, 2007). Words are counted by their base form, so run, runs and ran count as one word.
- 測的是「看得懂意思」。每題給一個英文字和一個簡單的句子,句子只告訴你這個字怎麼用,不會洩漏意思;你要從四個中文選項選出正確意思。這測的是閱讀時認得字義的能力(receptive knowledge),還不是會不會自己寫出來或說出來。用中文選項,是因為研究發現,用學生母語出題,比全英文選項更能反映學生真正認得的字(全英文選項還要先看得懂選項本身)(Elgort, 2013; Nguyen & Nation, 2011)。It tests whether you understand the meaning. Each question gives an English word and a simple sentence. The sentence only shows how the word is used and does not give away its meaning. You choose the correct meaning from four Chinese options. This measures recognizing word meaning while reading (receptive knowledge), not whether you can write or say the word yourself. We use Chinese options because research found that questions in the learner’s first language reflect what students really know better than all-English options (with all-English options you first have to understand the options themselves) (Elgort, 2013; Nguyen & Nation, 2011).
- 「我不知道」算答錯。四選一的題目用猜的也有四分之一的機會猜對,會讓分數虛高。研究以華人學生測試發現,加上「我不知道」可以明顯減少亂猜(Zhang, 2013; Stoeckel et al., 2019)。所以結果是比較保守、可靠的估計。“I don’t know” counts as wrong. On a four-choice question, guessing has a one-in-four chance of being right, which inflates the score. Research with Chinese-speaking learners found that adding an “I don’t know” option clearly reduces random guessing (Zhang, 2013; Stoeckel et al., 2019). So the result is a more cautious and reliable estimate.
- 兩階段,找到你的程度就停。第一階段從最常用的字開始,每級考 5 題,往上一級一級測;連續兩級只答對 2 題以下就停止,這時已經找到你開始覺得難的地方(Meara & Jones, 1988 的分級設計)。第二階段在這個程度附近的三級,每級再多考 20 題,細測你認得的比例。比這三級簡單的級數算作全部認得(如果第一階段在那一級就有答錯,則按答對比例計算),更難的級數算作不認得。這樣題目集中在最有鑑別度的地方,不必每一級都考很多題。Two stages, and we stop once we find your level. Stage 1 starts with the most common words and asks 5 questions per level, moving up one level at a time. When you get 2 or fewer right on two levels in a row, it stops, because we have found where words start to feel hard for you (a level-by-level design like Meara & Jones, 1988). Stage 2 asks 20 more questions on each of the three levels around that point to measure the share you know. Levels easier than these three count as fully known (if you missed some in Stage 1, they are counted by your share of correct answers), and harder levels count as unknown. This puts the questions where they tell us the most, so we do not need many questions on every level.
- 為什麼要每級多考幾題?研究指出,每 1,000 字只考 5 到 10 題,每答對一題就代表 100 到 200 個字,誤差太大(Gyllstad, McLean & Stewart, 2021)。所以在你的程度附近,我們每級考 25 題,讓估計更精確。Why ask more questions per level? Research points out that with only 5 to 10 questions per 1,000 words, each correct answer stands for 100 to 200 words, which leaves too much error (Gyllstad, McLean & Stewart, 2021). So around your level we ask 25 questions per level to make the estimate more precise.
- 每次隨機抽字。每一級有一個題庫,每次測驗都從題庫隨機抽題,所以重測時題目會不一樣,結果也比較不會因為記住題目而失真。Words are drawn at random each time. Each level has its own question bank, and every test draws questions from it at random. When you retake the test the questions will differ, so the result is less likely to be distorted by remembering questions.
- 結果是一個範圍。只抽一部分字來測,一定有誤差。我們依照統計方法(每級答對比例的標準誤差)算出範圍,例如「約 2,400 – 2,800 個字」,代表你的單字量最可能落在這個區間。這比給一個看似精確的數字更誠實。The result is a range. Because we test only a sample of words, there is always some error. Using a statistical method (the standard error of the share correct at each level), we work out a range, such as “about 2,400 – 2,800 words”, in which your vocabulary most likely falls. This is more honest than giving one number that looks exact.
題目和選項由 MJ School 自行編寫,沒有使用其他測驗的試題。頻率排序依據 wordfreq 英文詞頻資料(Speer, 2022)。範圍的計算方式是我們依照上述研究自行設計的,並非直接來自單一論文。The questions and options were written by MJ School; no items from other tests are used. The frequency ranking is based on the wordfreq English word-frequency data (Speer, 2022). The way the range is calculated is our own design based on the research above, not taken directly from a single paper.
參考文獻References
- Nation, I. S. P., & Beglar, D. (2007). A vocabulary size test. The Language Teacher, 31(7), 9–13.
- Nguyen, L. T. C., & Nation, I. S. P. (2011). A bilingual vocabulary size test of English for Vietnamese learners. RELC Journal, 42(1), 86–99.
- Elgort, I. (2013). Effects of L1 definitions and cognate status of test items on the Vocabulary Size Test. Language Testing, 30(2), 253–272.
- Zhang, X. (2013). The I don't know option in the Vocabulary Size Test. TESOL Quarterly, 47(4), 790–811. record
- Stoeckel, T., and colleagues (2019). A comparison of bilingual Vocabulary Size Test formats, including an “I don’t know” option, with a meaning-recall test. record
- Meara, P., & Jones, G. (1988). Vocabulary size as a placement indicator. In P. Grunwell (Ed.), Applied Linguistics in Society (pp. 80–87). CILT. PDF
- Gyllstad, H., McLean, S., & Stewart, J. (2021). Using confidence intervals to determine adequate item sample sizes for vocabulary tests. Language Testing, 38(4), 558–579. article
- Speer, R. (2022). wordfreq (Version 3.0) [Computer software]. Zenodo.
Your result你的英文單字量Your English vocabulary size
各級單字認得多少How many words you know at each level
每一條代表 1,000 個字,從最常用的開始。條越高,那一級的字你認得越多。斜線的條是推估的(太簡單沒有細測,或太難沒有測到)。Each bar stands for 1,000 words, starting with the most common. The taller the bar, the more words you know at that level. Striped bars are estimates (the level was too easy to test closely, or too hard to reach).
看看答錯或不知道的字See the words you missed or did not know
這個結果怎麼算出來的,請看測驗方法與理論依據。To see how this result is calculated, read how the test works and the research behind it.
想讓單字量穩穩往上升?Want your vocabulary to keep growing?
MJ School 的學生每天用自己的練習系統背單字:每個字會在快忘記的時候回來複習,老師也每天陪你開口說英文。歡迎來問問看。MJ School students learn vocabulary every day with their own practice system: each word comes back for review just as you are about to forget it, and a teacher speaks English with you every day. Get in touch to find out more.
