The ordered-letter review compares confirmed text against the roof, lantern, and notebook keys. It keeps the fictional first NOTEBOOK attempt visible, checks a separately entered retry, and holds uncertain hearing for human review.
Give each role a concrete responsibility
The pronouncer presents the current word and supplies the reviewed definition or sentence when requested. The learner may ask for clarification before attempting the letters. The recorder preserves what they hear. The judge compares the confirmed transcript with the current key and gives feedback.
Scripps describes pronouncer/judge roles and questions about definitions, sentences, and pronunciation in its local-bee guidance. Its page also emphasizes the order of spoken letters. Those limited ideas inform this packet; our word bank and retry policy are original. The page mixes program years, so this guide does not claim to reproduce current national competition rules. Scripps guidance for spelling bees.
Our practice policy keeps every participant in the activity after an error. A retry creates another attempt record; it does not replace the earlier one. This policy is declared here rather than attributed to Scripps.
Prepare the pronouncer cards
Use stable word IDs and bank version 1:
| Word ID | Word | Original definition | Original sentence |
|---|---|---|---|
| W1 | roof | The top covering of a building. | Rain tapped on the roof above our room. |
| W2 | lantern | A lamp in a protective housing that can be carried. | We placed a lantern beside the campsite table. |
| W3 | notebook | A book of pages used for writing notes. | I drew a map in my notebook. |
The definitions and sentences are authored practice copy. Have the teacher review their language and confirm the pronunciation route before use. Do not invent a phonetic transcription or rely on the spelling display as proof that a pronunciation was checked.
The learner-facing packet should provide access to the bank during preparation. During an oral attempt, use the teacher-declared presentation arrangement so the learner knows whether the written word is available. Do not imply a memorized oral performance if the key was visible.
Keep the judge keys exact and separate
| Word ID | Exact ordered key | Number of letters |
|---|---|---|
| W1 | R-O-O-F | 4 |
| W2 | L-A-N-T-E-R-N | 7 |
| W3 | N-O-T-E-B-O-O-K | 8 |
For this practice packet, comparison is case-insensitive for the English letter names. The order and repeated letters remain essential. R-O-F contains the same distinct letters as ROOF, but it omits an O and does not match the key.
Hyphens in the record separate spoken letter names. They are not additional spelling characters. Do not sort the letters, convert the sequence into a set, or silently insert an expected repeat.
Keep teacher keys separate in distribution. A collapsed or hidden panel in a client-side review page is not secure separation: someone with access to the underlying page may still see its content. Later prepare distinct learner and teacher copies appropriate to the activity.
Record before comparing
Use this blank attempt sheet:
| Round / attempt | Word ID / bank version | Clarification requested | Exact heard letters | Hearing status | Comparison / feedback | Next attempt reference |
|---|---|---|---|---|---|---|
The recorder should preserve the first transcript exactly, including an omission or repeated letter. If hearing is uncertain, mark the specific uncertainty and hold the comparison for human review. “I could not distinguish the last two letters” is different from “the learner spelled them incorrectly.”
Do not let a software field complete a partly heard answer from the key. An uncertain transcript is not a finished sequence. Keep any clarified transcript as a separate record so the judge can distinguish it from the first hearing.
Actual responses belong in this blank sheet only after someone observes them. The following example is fictional teacher material.
Review NOTEBOOK without rewriting the attempt
N-O-T-E-B-O-K
The target is N-O-T-E-B-O-O-K. The heard sequence has seven letters rather than eight. Its first six positions agree with the key; the recorded K arrives where the key expects the second consecutive O.
Feedback can say: “Your recorded sequence matches N-O-T-E-B-O, then goes to K. The key has another O before K. Try the whole sequence again.”
Fictional retry A2: N-O-T-E-B-O-O-K.
Try the worked example
Enter N-O-T-E-B-O-K for NOTEBOOK and check the transcript: the first mismatch is position seven, where the key expects O and the record has K. Enter a new retry as N-O-T-E-B-O-O-K. The match does not rewrite the first attempt printed above it. Use the uncertain-hearing control when the transcript is unresolved. Unknown hearing is not a spelling error.
This compares recorder-entered English letters, not speech or pronunciation. Actual oral judging and pronunciation review remain human tasks.
Recreate this example
Build one complete self-contained HTML page for this task: compare exact recorded letter sequences while retaining a fictional first attempt and uncertainty.
Use these exact original practice data and worked expectations: Bank v1: W1 roof R-O-O-F, four letters; W2 lantern L-A-N-T-E-R-N, seven; W3 notebook N-O-T-E-B-O-O-K, eight. A1 NOTEBOOK N-O-T-E-B-O-K, seven letters; first mismatch at position 7. A2 complete retry matches.
Implement the following useful local interactions: Use word buttons, a hyphen-separated transcript input, confirmed/uncertain buttons, and Check. Accept case-insensitive single English letters, preserve repeats/order, hold uncertain or blank transcripts, and do not fill unknown letters from the key. Keep A1 static and the current retry separate. Preserve the supplied source text and distinguish an uncomputed state, a successful local result, and an invalid or incomplete input. Update output only from the current input snapshot. An input edit must retire stale feedback until the reader runs the relevant action again. Do not silently invent missing values or replace a failing case with a success.
Design the page as a judge’s letter-record sheet with large ordered letters and a visibly retained first attempt. Put a compact task heading, actual sample data, the main result, and usable controls early enough to be useful on a narrow screen. Use readable text of at least 16px, generous paragraph spacing, strong contrast, native labels, accessible buttons, and visible keyboard focus. Stack long records at 360px and wrap code without page-level horizontal overflow.
Use type=button and explicit querySelector-based click handlers rather than relying on form submission. If Enter is useful, handle it only on the relevant input and ignore isComposing. Keep effects within this open page; no remote dependencies, network calls, persistence, fake downloads, or automated sending. This compares recorder-entered English letters, not speech or pronunciation. Actual oral judging and pronunciation review remain human tasks. Place worked explanations beside their data and identify teacher or synthetic examples clearly. Output only the complete HTML document, with inline styles and script and LF line endings.
Review the attempt records in NOS
Prepare separate learner and judge copies before an oral round; a hidden panel does not protect a key. In an optional teacher review copy in NOS, keep the first attempt beside the retry and revise eligible static feedback wording in visual editing mode.