Fluency, Retell, and a Screener That Disagree: Placing a Sixth Grader
Mr. Okafor is a sixth-grade reading interventionist placing students for a 10-week Tier 2 group. One student, Jalen, reads a sixth-grade passage aloud at 128 words per minute with 97% accuracy, but on retell he names only 2 of 12 key ideas. On the computer-adaptive screener his score lands near the 22nd percentile, described by the report as a third-grade level, with a reported standard error of measurement of plus or minus 7 percentile points. A maze comprehension probe gives 12 of 25 correct, below the 25th percentile. Jalen's teachers say he reads fine but does not remember anything. His science notebook, however, contains accurate if brief summaries of lab procedures he watched demonstrated rather than read. The school has one Tier 2 slot left, a Tier 3 slot reserved for students two or more years below grade level, and a vocabulary-focused elective. Jalen's family wants him in whichever group will make him read faster.
1. Which data source should carry the most weight, and what does the pattern rule in and rule out?
Read the worked answer
Word recognition should be ruled out first, and it is: 128 wpm with 97% accuracy is at or near grade-level expectations for sixth grade, so decoding is not his bottleneck. That leaves comprehension - and both comprehension measures point the same way. Retell (2 of 12 ideas) and maze (12 of 25, below the 25th percentile) are different formats with different response demands, so their agreement is meaningful. Two independent low scores beat one discrepant high one.
The screener should carry the least weight. A percentile with a plus or minus 7 SEM band means the plausible range for Jalen runs roughly from the 15th to the 29th percentile - comfortable enough to rule out the 'two or more years below' inference the report's grade-level label invites. A point estimate converted to a grade equivalent is not evidence of a three-year gap.
The oral reading score and the science notebook also matter diagnostically. He comprehended procedures he watched, which argues that background knowledge and oral language are assets, not deficits, and that his difficulty may be specific to written text and to producing a retell. Before placing him, run a listening-comprehension check: read a comparable passage aloud, then ask for retell. If listening comprehension is also weak, the problem is broad language comprehension and the instructional target shifts. If listening comprehension is strong, the target is reading-specific comprehension and text-structure strategies.
2. Choose among Tier 2, Tier 3, and the vocabulary elective, and build a progress-monitoring plan with specific decision rules.
Read the worked answer
Tier 2 is the defensible placement. Tier 3 requires evidence of two or more years below grade level, and the only source suggesting that is a screener score whose confidence band does not support it - while his oral reading is at grade level, meaning Tier 3 would target decoding skills he demonstrably has. The vocabulary elective alone is too narrow: his retell and maze scores implicate comprehension monitoring and text structure, not just word knowledge, though vocabulary should be a component.
Instructional content for the 10 weeks: reciprocal teaching or another explicit comprehension-strategy routine, text-structure instruction with graphic organizers, discussion-based knowledge building, and short fluency practice only as maintenance - pushing rate is not indicated at 128 wpm.
Progress monitoring:
- Maze probe twice weekly; baseline 12 of 25.
- Goal: move to roughly 18-19 of 25, near the benchmark, by week 8, with a slope of about 0.6-0.8 correct per week.
- Add a monthly oral retell scored against a 12-idea key, since retell is the skill his teachers notice missing.
- Decision rule set in advance: if four consecutive data points fall below the goal line at week 6, change the intervention rather than wait out the 10 weeks; if he is at or above the goal line for six points, fade to monitoring.
Re-run the screener only at the end of the cycle, and report the confidence band alongside the score.
3. How do you answer the family's request for a group that will 'make him read faster'?
Read the worked answer
Answer with the data and with respect for what the request is protecting. The family is describing a real phenomenon - Jalen reads smoothly, so he looks fluent, and smooth reading is what most people mean by good reading. Do not correct the premise bluntly; redirect it.
Say plainly what the assessment showed: Jalen reads sixth-grade text quickly and accurately, which is genuinely good news and means he does not need a speed group. What the assessments show is that he is not holding on to what he reads - two of twelve ideas. So the group will target remembering and using what he reads, and speed will be checked but not pushed, because faster reading without comprehension would make the problem worse, not better.
Give one concrete, checkable commitment: at weeks 4 and 8 the family gets a one-page update with his maze scores and a short passage he retells, plus one thing to do at home - read a two-page article together and ask him to explain it back, which costs nothing and builds exactly the skill being taught. Invite them to observe a session. Note that his science notebook shows strong comprehension when he watches rather than reads, and that this is a strength the group can build on.
Log the conversation and the family's stated preference in the placement record, since a family request that was heard and answered with reasons is different from one that was overridden silently.