Capuchino

    Spanish vocabulary assessment

    Find out how far your vocabulary reaches among the 10,000 most frequent Spanish words.

    Here’s how it works: rate words until the range is narrow enough for you. Most people rate about 25. Stop anytime; you can sharpen it later.

    For each word:

    1. You see the word on its own. Do you know it?
    2. If not, you see it in a sentence. Does the sentence make it clear?

    You don’t need to be able to say the word in Spanish yourself. That’s a separate skill.

    How it works

    What we measure

    How far down the list of the 10,000 most frequent Spanish words your solid understanding reaches. More precisely: the point where you still know about 8 in 10 words. Words are ranked by how often they appear in film and TV subtitles.

    How the estimate works

    • Each answer updates an estimate of your personal knowledge curve: the chance you know a word falls as words get rarer.
    • The next word is the one whose answer tells us most, so the test stays short.
    • You see a range, not a single number. The range covers your real level about 80% of the time, and narrows as you answer.
    • We suggest stopping when more questions would only narrow it a little.

    What your result means

    A range like “5,100 to 7,200” means you solidly understand roughly the 5,100 to 7,200 most frequent words. Below that point most words are known, but not all: expect a few percent of gaps, and your gap list is where you find them.

    What it doesn’t mean

    • It’s not a DELE score.
    • The level label (≈ B1, ≈ B2) is a rough, deliberately conservative guide based on studies of learners of English, French and Spanish.
    • It measures understanding, not whether you can produce the word yourself.
    • If you speak English, words that look like English (hospital, idea) are easy wins, so the estimate reads a little high.

    How Capuchino uses it

    Your result sets where your gap list starts: words beyond your range count as “probably unknown”, and words before it as “probably known”. You can correct any word. Topic lists and exports for Anki use the same statuses.

    Based on

    • Word frequencies: doozan/spanish_data (Wiktionary; OpenSubtitles via FrequencyWords).
    • Example sentences: Tatoeba.
    • Method: adaptive testing with a Bayesian estimate, choosing each next word for maximum information.
    • Level guide: Milton (2010), Milton & Alexiou (2009), Hacking, Rubio & Tschirner (2018).
    • More on our specs page.
    • The full methodology, with every number the model uses: Methodology.