Sorting Four Readers with Conflicting Screening Data
A fourth-grade team reviews fall screening results before allocating its single 12-week intervention slot (30 minutes, four days a week). Benchmark: 100 words correct per minute (wcpm), accuracy at or above 95%, retell at least 6 of 10 key ideas.
- Ada: 98% accuracy, 112 wcpm, retell 2/10, inferential questions 2/6.
- Ben: 91% accuracy, 68 wcpm, retell 8/10, inferential 5/6.
- Cara: 96% accuracy, 95 wcpm, retell 3/10, literal 6/6, inferential 1/6.
- Dev: 99% accuracy, 130 wcpm, retell 9/10, inferential 6/6.
Everyone not selected stays in core instruction. One teacher argues that Ada's rate "rules out" a fluency or comprehension problem, and another wants Cara in the slot because her inference score is lowest in the grade.
1. Rank the four students by urgency for the single intervention slot and justify each ranking using the pattern across measures, not any single score.
Read the worked answer
Ranking: Ben > Cara > Ada > Dev.
- Ben is the only student below benchmark on decoding accuracy (91%) and rate (68 wcpm), while comprehension stays strong (8/10 retell, 5/6 inferential). That profile is the classic signature of a word-reading bottleneck: he understands what he manages to read. A targeted 12-week slot is well matched to that need and likely to close the accuracy gap fastest.
- Cara is accurate and near benchmark on rate but collapses on inference (1/6) with intact literal comprehension (6/6). Her need is real, but it is a strategy need that a comprehension-focused group in core instruction can plausibly serve, especially since her accuracy and rate are not the constraint.
- Ada reads fast and accurately yet retells almost nothing. That is a genuine red flag, but it is diagnostically ambiguous (see next task) and should not be treated as a decoding problem.
- Dev exceeds every benchmark; the appropriate move is enrichment, not intervention.
Key reasoning: the ranking follows the mismatch between scores, because the slot should go to the student whose profile most specifically matches what 30 minutes of targeted intervention can fix.
2. Respond to the teacher who says Ada's 112 wcpm "rules out" a fluency problem. What follow-up assessment would you run before writing her goal?
Read the worked answer
The claim confuses rate with fluency. Rate is one component; a reader can decode quickly and still not construct meaning. Ada's 98% accuracy plus 112 wcpm shows her word reading is efficient; her 2/10 retell and 2/6 inferential scores show comprehension is not keeping pace. Something between the word and the meaning is breaking down.
Before writing a goal, separate the plausible explanations:
- Language comprehension - give her a listening comprehension task (same retell format, text read aloud to her). If listening retell is also 2/10, the problem is meaning-making with language, not print.
- Task/format artifact - retell depends on memory and oral expression. Try a prompted retell ("tell me about the problem the character faced") or a written response to see whether the knowledge is there but the production is not.
- Attention/engagement - note whether the pattern holds across two texts, including one on a topic she chose.
Only if listening comprehension is intact and only reading comprehension is weak does a text-level strategy goal make sense. Without this step the team risks writing a fluency goal for a student who is already fluent.
3. Name two specific pieces of evidence that would change your allocation decision, and explain the direction of the change.
Read the worked answer
A decision rule is only as good as the data behind it, so state in advance what would flip it:
- A decoding inventory for Ben. If his 91% accuracy turns out to come from a handful of proper nouns and multisyllabic words with no consistent phonics pattern - and a re-read of a familiar text puts him above 100 wcpm - then his need is practice and volume, not intervention. That would drop him below Cara.
- A listening-comprehension check for Cara. If Cara's listening comprehension is also weak on inference, her difficulty is not reading-specific and a reading-strategy group will not address the root cause; the team would need a language-focused referral instead, freeing the slot.
- A prompted retell for Ada. If Ada's retell jumps to 7/10 with prompts, her 2/10 reflected task demands rather than comprehension, and she moves to enrichment.
The general principle: the ranking is a hypothesis to be tested, and the team should name the disconfirming evidence in advance rather than defending the original sort.