A reading group can begin as a support and quietly become an identity.
The teacher forms a group because several pupils need work on decoding, fluency, vocabulary or comprehension. The same children sit together next week. Then next month. The materials become easier, the expectations narrow, and everybody starts to know which table is the “low group.” What was meant to be a temporary instructional response has become a fixed social category.
Flexible reading groups are designed to prevent that drift. The group exists because of a current, specific reading need. The instructional content is chosen for that need. Progress is monitored. When the need changes, the group changes.
This sounds obvious, but it requires more than moving names between tables. In July 2026, IES published a tip sheet on planning instructional time to prevent reading difficulties, highlighting three connected jobs: using assessment data to inform instruction, allocating time for differentiated teaching, and staffing flexible reading groups. Its K–3 differentiated-reading toolkit goes further, helping teachers collect and interpret data, establish decision rules, form teacher-led small groups, plan content and scaffolding, monitor progress and flexibly regroup.
The deeper question is not “Should children be grouped by level?” It is: how can teachers temporarily concentrate instruction around a reading bottleneck without turning today’s evidence into tomorrow’s label?
The 50-second answer
Flexible reading groups are temporary, evidence-based small groups formed around a specific instructional need rather than a permanent ability category. A group may focus on phoneme–grapheme mapping, multisyllabic decoding, oral reading fluency, vocabulary, gist, comprehension monitoring or another defined component of reading.
The mechanism is concentration. In a small teacher-led group, instruction can target a narrow skill, offer more response opportunities, give immediate corrective feedback and adjust text difficulty or scaffolding. The group should then be reviewed using progress evidence. If a learner no longer needs that support, they move. If a different bottleneck appears, the grouping changes.
IES guidance is explicit about using assessment data, decision rules, teacher-led small-group differentiated instruction and flexible regrouping. The caveat is equally important: grouping itself does not teach. A badly targeted group can simply produce a smaller version of weak instruction. The unit of quality is the match among learner need, instructional content, intensity, feedback and timely reassessment.
1. Start with the reading component, not the child label
Reading is not one skill. A learner may decode accurately but slowly, read fluently but lack vocabulary, or understand oral language while struggling with print. Grouping should therefore begin with the component currently limiting performance.
A concrete example: three pupils who all scored low on comprehension are separated after diagnosis: one needs decoding repair, one needs vocabulary, and one needs inference practice.
The caveat: A single total score can hide very different causes. Groups formed from broad scores alone may target the wrong thing.
2. Assessment data need interpretation
A score becomes useful only when the teacher understands what task produced it and what errors mean. Data should answer an instructional question, not merely sort children.
A concrete example: a nonsense-word decoding task reveals unstable grapheme–phoneme mapping while passage reading shows the same pupil compensating through context.
The caveat: Assessment can over-test. Use the minimum data needed to make a better teaching decision.
3. Decision rules reduce arbitrary grouping
Clear rules help teachers decide when a learner enters, stays in or exits a group. This makes movement more consistent and less dependent on impressions.
A concrete example: pupils join a fluency group when accuracy is secure but rate and phrasing remain below a defined classroom benchmark, then exit after repeated evidence improves.
The caveat: Rules should guide judgement, not replace it. Exceptional cases may need closer review.
4. Groups should be homogeneous only for the target skill
A narrow skill group can be useful because the teacher can deliver one focused sequence. That does not mean the pupils are globally similar readers.
A concrete example: four pupils practise multisyllabic decoding together even though their broader vocabulary and writing levels differ.
The caveat: Teachers should not carry the same grouping into every reading activity.
5. Teacher-led time is the scarce resource
Small-group differentiation works partly because teacher attention is concentrated. Scheduling therefore matters as much as grouping logic.
A concrete example: the timetable protects a daily rotation where the teacher works directly with one group while other pupils complete well-taught independent tasks.
The caveat: If independent stations constantly require teacher rescue, the supposed small-group advantage disappears.
6. Independent work must be genuinely independent
While the teacher teaches one group, the rest of the class needs tasks they can complete productively without repeated clarification.
A concrete example: students reread familiar text, complete cumulative word work or use established vocabulary routines.
The caveat: Busywork can preserve silence while adding little learning. Independence and educational value must both be present.
7. Group size changes response density
Smaller groups allow more oral reading, more turns, faster feedback and better visibility of errors.
A concrete example: in a four-pupil decoding group, each learner reads and receives correction repeatedly rather than waiting through a whole-class queue.
The caveat: Very small groups are costly. The instructional gain should justify the staffing and time.
8. Text difficulty should fit the teaching job
A group targeting decoding may need controlled text; a comprehension group may need rich language and ideas. One universal notion of ‘right level’ is too crude.
A concrete example: a decoding group uses text that makes the target spelling pattern frequent enough to practise, while a vocabulary group reads a knowledge-rich passage with teacher support.
The caveat: Keeping learners permanently in simplified texts can starve language and knowledge.
9. Foundational-skill groups need explicit instruction
The WWC recommends intensive, systematic small-group instruction in foundational skills for pupils below benchmark. The teacher should explain, model, prompt and provide sufficient practice.
A concrete example: a group rehearses a vowel pattern, blends words, reads them in connected text and receives immediate correction.
The caveat: Discovery-only approaches may leave struggling readers guessing at a code they need taught clearly.
10. Fluency groups need more than speed
Purposeful fluency work should combine accuracy, phrasing and repeated successful reading rather than racing words per minute.
A concrete example: students echo a model, reread a short passage and discuss how punctuation changes phrasing.
The caveat: Speed pressure can reinforce errors or create anxiety. Fluency is efficient meaningful reading, not a contest.
11. Vocabulary groups should connect words to knowledge
Word teaching is stronger when definitions, examples, morphology and text context reinforce one another.
A concrete example: a group studies predict, prediction, predictable inside a Science text rather than memorising isolated definitions.
The caveat: A vocabulary group should not become a weekly list detached from actual reading.
12. Comprehension groups need a precise process target
‘Comprehension’ is too broad. The teacher should know whether pupils need gist, questioning, inference, monitoring, summarising or knowledge support.
A concrete example: students practise stating the gist after each short section and checking whether later information changes it.
The caveat: Strategy instruction cannot compensate for missing decoding or background knowledge. Diagnose upstream constraints.
13. Progress monitoring should match the target
The measure used to decide whether a group is working should sample the skill being taught.
A concrete example: a decoding group is monitored with unfamiliar words containing the taught patterns rather than a generic reading-attitude survey.
The caveat: Using the same broad assessment for every group can miss meaningful change.
14. Regrouping should be expected, not exceptional
Flexible grouping only becomes flexible when movement is normal. Students should understand that groups are temporary responses to current learning.
A concrete example: the teacher routinely changes groups after a short review cycle and explains that different readers need different practice at different times.
The caveat: Frequent changes without enough instructional time can create instability. Regroup when evidence warrants it.
15. Group names should avoid status signals
Colours, animals or levels can quickly acquire social meaning if they map permanently to attainment.
A concrete example: groups are named by the current job—’word-building’, ‘phrasing’, ‘gist’—or simply by neutral rotation labels.
The caveat: Children can still infer relative difficulty. The deeper protection is real movement and equal intellectual respect.
16. The highest group also needs instruction
Flexible grouping is not only remediation. Advanced readers may need vocabulary depth, complex syntax, disciplinary reading or stretch text.
A concrete example: a secure decoding group works on comparing arguments across challenging sources.
The caveat: If small-group systems reserve teacher expertise only for struggling readers, strong readers can become under-taught.
17. Groups should not reduce access to whole-class knowledge
Targeted work should sit inside a curriculum where all pupils still encounter important ideas, rich texts and shared discussion.
A concrete example: a struggling decoder receives extra code instruction but participates in the same knowledge-rich read-aloud as classmates.
The caveat: Removing pupils repeatedly from core content can create a knowledge gap that later harms comprehension.
18. Attendance matters
Small-group instruction is cumulative. Pupils who miss sessions may appear unresponsive when the real problem is inconsistent exposure.
A concrete example: the teacher records actual instructional minutes received, not only planned sessions.
The caveat: Attendance data should not be used to blame families; it helps interpret whether the intervention had a fair chance.
19. Tutor or TA groups need training and structure
Small groups led by teaching assistants can help when adults know the target routine and receive support.
A concrete example: a TA uses a structured decoding sequence with clear correction procedures and regular teacher review.
The caveat: Simply assigning extra adults without training can increase variation rather than quality.
20. Error correction should be immediate and specific
Small groups create the opportunity to stop a wrong pattern before it becomes repeated practice.
A concrete example: when a pupil misreads a vowel pattern, the teacher prompts the relevant sound mapping, asks for a second attempt, then returns later to check retention.
The caveat: Overcorrecting every hesitation can disrupt meaning and confidence. Correction should match the instructional target.
21. Pupils need enough successful practice
A group that spends most of its time listening to teacher explanation wastes the response-density advantage.
A concrete example: each pupil produces multiple oral, written or pointing responses during the session.
The caveat: High response count is not enough if items are trivial or feedback is weak.
22. Grouping can expose a curriculum problem
If many pupils need the same support repeatedly, the issue may lie in whole-class instruction or curriculum sequence rather than individual deficit.
A concrete example: half the year group enters a multisyllabic decoding intervention, prompting leaders to examine where morphology and advanced phonics are taught.
The caveat: Targeted groups should not become a permanent workaround for weak core instruction.
23. Flexible grouping is different from tracking
Tracking places learners into relatively stable pathways based on broader attainment. Flexible grouping changes according to immediate instructional need.
A concrete example: a pupil can be in a decoding group this month and later join a vocabulary group without a change in overall class placement.
The caveat: Schools should not use the language of flexibility to disguise effectively permanent low tracks.
24. Student voice can reveal hidden friction
Pupils can often say whether group text is too easy, instructions are unclear or practice feels repetitive.
A concrete example: a learner explains that they can decode the words but do not know what the passage vocabulary means.
The caveat: Preference is useful evidence, not the final decision. Instruction still follows demonstrated need.
25. Families need a clear explanation of the group’s job
Parents may worry when a child joins a reading group. Specific language is more helpful than broad labels.
A concrete example: ‘We are giving extra practice on reading longer words accurately for six weeks, then we will review’ is clearer than ‘your child is in the lower group.’
The caveat: Schools should avoid promising guaranteed gains from a short intervention.
26. Exit criteria protect against permanent support
Groups should have a planned review point and a definition of sufficient progress.
A concrete example: after four weeks, pupils who meet the decoding criterion move back to ordinary differentiation while others receive a new diagnostic review.
The caveat: Exiting too early can produce relapse. Use more than one successful performance where possible.
27. Non-response requires diagnosis, not endless repetition
When a learner makes little progress, simply continuing the same group for another term may be inappropriate.
A concrete example: the team checks attendance, hearing, language, instructional fit and whether the target skill was correctly identified.
The caveat: Escalation should be evidence-led and follow local support procedures; teachers should not diagnose clinical conditions alone.
28. Flexible groups can support motivation through visible progress
A temporary, precise target makes improvement easier to notice than a global identity such as ‘bad reader.’
A concrete example: students graph how many unfamiliar target words they can now decode accurately and celebrate exit from the group.
The caveat: Progress displays should not publicly compare children or create shame.
29. The classroom management system has to support movement
Rotations, materials and expectations need enough routine that regrouping does not consume the lesson.
A concrete example: students know where each station begins, how to request help and what to do when finished.
The caveat: Complicated rotation systems can create more cognitive and behavioural friction than the differentiation is worth.
30. Technology can assist but should not define groups
Digital assessment or practice can provide useful data, but teacher observation and direct reading remain important.
A concrete example: a platform flags slow word reading, and the teacher listens to the pupil before deciding the instructional target.
The caveat: Automated classifications can be wrong or opaque. Do not outsource grouping decisions blindly.
31. Data should be recent enough to describe the learner now
Reading changes with instruction. Old benchmark data can preserve obsolete groups.
A concrete example: a pupil’s group is reconsidered after several weeks of successful class reading rather than waiting for a term-end test.
The caveat: Testing too frequently can consume instruction. Use efficient progress checks.
32. Group instruction should reconnect to authentic reading
Skills need to return to meaningful text so learners see the purpose and transfer the component.
A concrete example: after practising a decoding pattern, pupils read a short passage where the pattern supports real comprehension.
The caveat: Isolated drills can improve item performance without becoming functional reading.
33. The system should get simpler as readers strengthen
As pupils become more independent, targeted group time can decrease and ordinary classroom reading can carry more of the work.
A concrete example: a pupil exits intensive word-reading support and receives only occasional monitoring.
The caveat: Support that never fades may be compensating for a deeper issue or creating dependence.
34. The real unit is need × instruction × time
Group labels are administratively convenient, but learning depends on whether the right content reaches the right learner often enough, with feedback.
A concrete example: two schools both run ‘four-person reading groups’; one targets precise skills daily, the other uses generic worksheets weekly, producing very different learning conditions.
The caveat: Do not treat group size or schedule as the intervention by themselves.
Practical route for teachers
Build groups from a diagnostic question, not a ranking list. Name the skill, choose a short teaching sequence, protect teacher-led time, and decide the progress measure before the cycle begins. Make movement normal and explain the job to pupils and families in specific language.
Audit the rest of the room as carefully as the teacher table. Flexible grouping fails when independent tasks constantly interrupt instruction. Review group membership on a planned cycle and investigate non-response rather than simply extending the same support.
Practical route for students
If you are placed in a reading group, treat the group as practice for one current skill, not as a statement about what kind of reader you are. Ask what the group is working on and what evidence will show you are ready to move.
When the target skill improves, use it in real reading. A decoding pattern, vocabulary routine or gist strategy becomes valuable when it helps you understand text outside the group.
Practical route for parents and families
Ask the school three questions: What specific reading skill is this group targeting? How often will the group meet? When and how will progress be reviewed? Clear answers usually indicate that grouping is instructional rather than permanent labelling.
Support at home should complement the target without turning family reading into intervention. Shared reading, conversation and access to interesting books still matter even when a child is receiving focused skills support.
Common failure modes
- Forming permanent groups from one broad reading score.
- Calling groups flexible while the same pupils remain together all year.
- Using easier texts for struggling readers so consistently that vocabulary and knowledge shrink.
- Running teacher groups while independent stations constantly require rescue.
- Monitoring progress with measures unrelated to the taught skill.
- Using teaching assistants without training or a structured routine.
- Repeating the same intervention when progress is minimal instead of rediagnosing.
- Turning neutral group names into obvious status hierarchies.
- Removing pupils from rich whole-class content for too much of the week.
- Treating grouping itself as the intervention rather than the instruction inside it.
Frequently asked questions
Are flexible reading groups supported by evidence?
IES resources recommend differentiated reading instruction based on current assessment, teacher-led small groups, progress monitoring and flexible regrouping. Specific effects depend on the quality and target of instruction.
How often should groups change?
There is no universal timetable. Review often enough that old data do not become permanent labels, while allowing enough instructional time for learning to occur.
Should children be grouped by reading level?
Broad level can be a starting signal, but instruction is stronger when groups are formed around specific needs such as decoding, fluency, vocabulary or comprehension processes.
Are small groups always better than whole-class teaching?
No. Small groups are valuable when concentrated attention and targeted practice solve a specific need. Important shared knowledge and core instruction still belong in whole-class teaching.
What if a child is not progressing?
Check instructional fit, attendance, target skill, practice quality and other possible barriers, then follow school support procedures. Do not simply keep the learner in the same group indefinitely.
The final idea
Flexible reading groups are not a seating plan. They are a temporary concentration of teaching around a current bottleneck.
That distinction changes everything. The child is not “a low reader.” The child is a learner who, today, needs more practice with a particular part of reading. The group exists to deliver that practice efficiently, collect better evidence and then dissolve or change when the evidence changes.
The best system therefore moves. It protects access to rich curriculum, avoids permanent status labels, and treats progress monitoring as a way out of support rather than a reason to preserve it.
The rule is: group the need, not the identity—and let current reading evidence decide when the group has finished its job.
Sources and further reading
- IES — Planning Instructional Time to Prevent Reading Difficulties in Elementary Grades
- IES — Toolkit for Differentiating Reading Instruction for Grades K–3
- What Works Clearinghouse — Assisting Students Struggling with Reading in the Primary Grades
- What Works Clearinghouse — Providing Reading Interventions for Students in Grades 4–9
Continue exploring on eduKateSG
- How X Works Hub
- How Reading Intervention Works | Repair the First Broken Link
- Did You Know Learning to Read Rebuilds the Brain? | How Print Becomes a Reading Network
- How Instructional Adaptation Works | Change the Route Without Changing the Learning Goal
A four-week flexible-group cycle
Week 1: diagnose and form
Use the smallest useful assessment set. Listen to actual reading. Identify the component that limits performance and form temporary groups around that component. Tell pupils the group’s job in plain language.
Avoid placing every low-scoring pupil together automatically. Two children with the same overall score may need completely different instruction.
Week 2: teach for response density
Protect teacher-led time. Use explicit modelling where needed, then maximise learner responses. Keep text and tasks tightly matched to the group target. Record recurring errors rather than every minor hesitation.
At the end of the week, ask whether the selected target still appears causal. If the group is practising the wrong thing, change early rather than waiting for a formal review.
Week 3: vary and connect
Move beyond the original examples. Put the same skill inside connected text, new word forms or changed comprehension contexts. This is where brittle learning becomes visible.
Continue whole-class access to rich texts and knowledge. Targeted groups should not become a separate low-language curriculum.
Week 4: review and regroup
Use a brief progress measure that matches the target. Look at more than one sample if possible. Pupils who are secure leave or change groups. Pupils making limited progress receive a new diagnostic review rather than simply more of the same.
This review is the moment that makes grouping genuinely flexible.
Three grouping mistakes that look efficient
The permanent “easy book” table
The classroom runs smoothly because the same pupils always receive easier material. Management is simple, but language exposure and expectations shrink. Over time, the group may become self-fulfilling.
A better system separates access support from intellectual diet. Provide the decoding help needed while preserving rich read-alouds, vocabulary and knowledge.
The test-score sort
Students are ranked by one benchmark and divided into thirds. The data are neat, but the instructional meaning is weak. A composite score cannot tell the teacher which component to teach.
Use the benchmark to flag who needs closer diagnosis, then group by the bottleneck that instruction can actually change.
The carousel of activities
Students rotate through beautifully organised stations, yet the teacher group spends half its time on worksheets and the independent groups practise low-value tasks. The system looks differentiated while response quality remains poor.
Judge the rotation by what learners are doing cognitively, not by how smoothly they move.
A family-facing explanation template
“We are giving your child extra practice on [specific skill] for [planned period] because current reading shows [observable evidence]. The group meets [frequency] and uses [brief description of approach]. We will review using [measure] on [timeframe]. If the skill improves, the group will change or end.”
This language protects dignity because it describes a temporary instructional response rather than a fixed identity. It also makes the school accountable for review.
The strongest signal that grouping is working
The best evidence is not that the child performs smoothly at the teacher table. It is that the taught skill appears in ordinary reading with less support. Words decode more accurately in new text. Phrasing improves outside the practised passage. Vocabulary supports comprehension in another subject. Gist can be identified without the group prompt.
That is when concentrated instruction has become reading capability.
A simple data board for teachers
For each temporary group, keep only five fields: target skill, starting evidence, teaching routine, review date, current decision. This is enough to stop groups drifting into permanence while avoiding a large administrative system.
The current decision should use verbs: continue, fade, regroup, intensify, rediagnose. A noun such as “low group” freezes the learner; a decision verb keeps the system dynamic.
Teachers can review the board together during grade-level meetings. The useful conversation is not “Which children are weak?” but “Which needs are recurring, which groups are moving, and what does that tell us about core instruction?”
Why flexible grouping can improve equity only if movement is real
Targeted small-group instruction can widen access by giving struggling readers more practice and feedback. But if the same pupils remain in simplified groups indefinitely, the system can produce the opposite effect: less exposure to complex language, fewer opportunities for advanced discussion and lower expectations.
Equity therefore depends on two protections. First, targeted instruction should repair access to the shared curriculum rather than replace it. Second, the system must create real opportunities to leave support when current evidence improves.
A flexible group is equitable not because everyone receives the same work, but because support is designed to increase future access rather than define a permanent ceiling.
A classroom scheduling model
Flexible reading groups need a schedule that protects direct instruction without fragmenting the rest of literacy. One workable pattern is a short whole-class knowledge or language launch, followed by a protected small-group block, then a return to shared text or writing. The exact timetable will vary, but the architecture matters: targeted skill work should connect back to meaningful literacy.
During the small-group block, independent tasks should be cumulative and familiar. Students can reread known texts, practise spelling patterns already taught, complete vocabulary retrieval or write from a shared text. The tasks should not introduce important new material while the teacher is unavailable.
The teacher group should have a visible target for the session and a planned final check. Five minutes of explanation followed by twenty minutes of worksheet completion squanders the advantage of small groups. The teacher should hear pupils read, think and respond repeatedly.
At the end of the block, a short whole-class reconnection can prevent the room from feeling like several separate curricula. One pupil may have worked on decoding, another on phrasing, but everyone returns to the shared purpose of understanding and using text.
A safeguard against the Matthew effect
Reading gaps can widen when stronger readers receive more complex language and knowledge while struggling readers spend increasing amounts of time on simplified material. Flexible groups should therefore be designed to repair foundational access quickly while preserving exposure to rich oral language, read-alouds and curriculum knowledge.
A pupil who needs decoding support can still discuss a sophisticated story heard aloud. A child receiving fluency practice can still learn ambitious Science vocabulary. Targeted instruction should add precision, not subtract intellectual diet.
This is one of the strongest reasons to keep groups temporary. The aim is not to build a parallel low-level curriculum. It is to help the learner participate more fully in the common one.
The minimum viable flexible-group system
A school does not need an elaborate platform to begin. It needs recent reading evidence, a small set of instructional routines, protected time, a review date and permission for groups to change.
The system should be simple enough that teachers can explain it in one sentence: “We group temporarily by the reading skill that needs the most practice, teach that skill intensively, and regroup when new evidence says the need has changed.”
If the system requires so much administration that teachers lose direct reading time, the management architecture has defeated the instructional purpose.
Why movement should be visible but not dramatic
Regrouping should feel ordinary. A child should not be publicly “promoted” from a low group or “demoted” after a weak assessment. Quiet movement based on current need protects dignity and helps students understand that learning states change.
The best culture is one in which different groups are expected to exist for different purposes and no single group represents a permanent hierarchy.
One final grouping rule
The group should change when the evidence changes. If it does not, the system is no longer flexible.