Skip to content
Lizely

Spelling Quiz

Find one misspelled entry, type its verified standard correction, and clear five source-checked rounds with accessible touch and keyboard controls.

Privacy: your files never leave your device. All processing happens locally in your browser.

How to use

  1. 1.Inspect the four displayed entries and select the only misspelled word by touch, pointer, or arrow keys followed by Enter.
  2. 2.After the misspelled entry is confirmed, type its standard spelling as one ASCII word; surrounding spaces and letter case are normalized.
  3. 3.Submit five verified corrections for exactly 200 points each; one distinct error remains playable, while a second requires Restart.

About Spelling Quiz

Spelling Quiz is a five-round English game with two deliberate steps. First, inspect four visible entries and identify the only misspelled one. Second, type the standard spelling from memory. A round awards 200 points only after both steps are correct, so completing all five rounds produces exactly 1,000.

The two-step design tests recognition and recall separately. Recognizing that a form looks unusual can be easier than producing its correct letter sequence. The game does not award points merely for selecting the odd-looking candidate. Once the misspelling is found, the candidate buttons lock and a blank text field asks for a one-word correction.

Every detection set contains four unique strings. Exactly one is the intended misspelling, and the other three are verified standard English words. The corrected form is deliberately absent from the four candidates, so the detection stage cannot reveal the answer beside the misspelled entry. Correct positions vary across all four slots rather than following a repeated pattern.

Princeton WordNet 3.1 index data is the primary machine-readable spelling reference. During implementation, every standard correction and every correctly spelled distractor was found as an exact indexed lemma. Each intended misspelling was absent. Index presence alone is not a universal proof about every possible English name or inflection, but it is a consistent validation boundary for this fixed common-word fixture.

Cambridge Dictionary supplies an independent cross-check for the five corrections. Each exact entry URL returned HTTP 200, and the downloaded page contained the expected standard spelling. Combining the WordNet index check with current dictionary entries reduces the chance of approving an invented correction or a historical variant that does not fit the quiz's intended modern standard.

The game widget does not publish the correction before submission. Gameplay state stores only the round, keyboard cursor, selected candidate index, score, normalized error signatures, and terminal flags. The standard answer is not copied into candidate aria labels, data attributes, hidden elements, status text, input placeholders, or CSS class names. The correction field starts empty and contains only what the player types.

Input normalization is narrow and explicit. Leading and trailing whitespace is removed, then ASCII letters are converted to lowercase for comparison. One uninterrupted sequence from a through z is accepted. Blank text, multiple words, apostrophes, hyphens, digits, accented letters, emoji, and punctuation are rejected atomically. This prevents a correct answer embedded inside extra text from passing.

Pointer and touch play use native candidate buttons with a minimum height greater than 44 pixels. Candidate words wrap inside their controls. The mobile layout stays in one bounded column and changes to two columns on wider screens. Minimum-width protection and overflow containment stop long entries from forcing the page to scroll horizontally.

Keyboard play covers both stages. Arrow keys move the detection cursor through the four entries and wrap at the ends. Enter selects the highlighted entry. After the unique misspelling is selected, focus can move to the native text input, where Enter submits the correction through the form. The global shortcut ignores Enter originating from a button or input, preventing duplicate submission.

Errors from both stages share one deterministic policy. Selecting a correctly spelled entry records a signature containing the round and candidate index. Submitting a valid but incorrect correction records the round and normalized word. Repeating the identical selection or the same correction with different case or surrounding spaces does not add another error. The first distinct error remains recoverable; the second different error deadlocks the run until Restart.

Invalid candidate indices, fractional indices, unknown movement commands, and invalid correction formats are atomic no-ops. They do not alter score, history, or progression. Once the correct detection is made, further candidate changes are disabled until correction. Completion and deadlock freeze all game actions. Restart restores round one, the first cursor position, no selection, zero score, a blank input, and an empty error history.

The shared GameShell stores the best completed score locally when browser storage is available. It also includes Restart and a double-Escape boss key. Press Escape twice within the timing window to replace the quiz with a spreadsheet-like cover, then twice again to return. Boss mode does not modify the chosen entry, typed correction, errors, or score.

All detection and correction logic runs locally in the browser. Typed text is not uploaded, no account is required, no paid service is called, and no package dependency is added. The quiz is fixed recreational spelling practice rather than an adaptive literacy assessment, educational diagnosis, language certification, or measure of intelligence.

Focused tests keep the reference fixture independent from production evaluation. Ten literal external golden cases record five accepted standard forms and five rejected misspellings. A distinct private round fixture pins candidate order, unique misspelling position, absent correction, and correct answer without deriving expectations from the product table. Tests also verify all three distractors against a fixed standard-word set, four-position coverage, normalization, exact 200-point progression, selection and correction deduplication, the shared second-error deadlock, invalid atomicity, terminal freeze, and Restart. The sources support spelling claims only; round order, UI, scoring, and error policy are product decisions.

Methodology & sources

Define five fixed four-candidate rounds with one misspelled index and one absent standard correction. Require all corrections and all correctly spelled distractors to occur as exact WordNet 3.1 indexed lemmas, require each misspelling to be absent, and independently fetch the five Cambridge correction entries. Keep ten external golden spellings plus a private literal round fixture. Validate candidate uniqueness, exactly one misspelling, correction absence from candidates, all four answer positions, trim plus ASCII lowercase normalization, two-stage 200-point transitions to 1,000, selection and normalized-correction signature deduplication, shared second-distinct-error deadlock, invalid and unknown atomicity, terminal freeze, Restart, and hidden correction state.

Frequently asked questions

What counts as a valid correction?
Enter one ASCII alphabetic word. The game trims surrounding spaces and compares lowercase letters, but rejects internal spaces, punctuation, digits, and non-ASCII characters.
How were the standard spellings checked?
All corrections and correctly spelled distractors were found in Princeton WordNet 3.1 index data. Each correction also has an independent live Cambridge Dictionary entry.
Can I see the correction before I submit it?
No. The game shows only the four detection candidates, then a blank correction field. Correctness and the standard answer are not placed in the game widget's accessible attributes or hidden text.

Related tools

Mini Games guides

View all