Math intervention software for middle school Tier 2 and Tier 3 groups

Your intervention block is 30 minutes and six students who each got it wrong for a different reason. Coeducate tutors each student with hints, names the misconception, and hands you tomorrow's groups.

See pricing

Student work is never used to train models. No card required to run the demo.

Mastery Check

of 3 runs left

Subject

Grade band

Student work

Nothing here is stored

Use the sample or your own practice text. Never paste a real student's name.

Tutor mode

Session evidence card

This is what a teacher opens after every session. That was your third run. Nothing you typed was stored.

The short answer

What math intervention software should do in grades 6 to 8

Coeducate is math intervention software for middle school Tier 2 and Tier 3 groups. Each student works problems from your intervention material with an AI tutor that asks the next question instead of giving the answer. Every session ends with a dated, per-skill mastery record that names the specific misconception, such as adding denominators or scaling a ratio by adding, with the student's own working quoted as the evidence.

It does not replace your screener or your intervention curriculum. It removes the part of the block that eats the time: working out, paper by paper, why each student got it wrong.

At a glance

Grades: Math, grades 3 to 12
Group size: Any, the record is per student
Session length: Fits a 20 to 40 minute block
Diagnosis: Named misconception per skill
Teacher data entry: None
Export: CSV on every tier
Training on student work: Never

Last updated September 2026

01 Why grades 6 to 8

Why middle school math intervention stalls on fractions and ratios

On the 2024 NAEP math assessment, 39 percent of eighth graders scored below the Basic level, and the losses for the lowest-performing tenth of students grew about 70 percent larger between 2022 and 2024, as The Hechinger Report summarized.

Most of those students are not missing arithmetic. They are missing rational numbers: fractions, decimals, ratio and percent. The federal IES practice guide on math intervention recommends that materials in grades 4 through 8 focus intensely on rational numbers, and rational numbers are exactly where one wrong answer hides the largest number of different reasons.

Adds across fractions

Writes 1/3 + 1/4 = 2/7. Treats the numerator and denominator as two separate whole numbers.

Longer decimal is bigger

Says 0.125 is greater than 0.5 because 125 is greater than 5.

Ratio as a difference

Scales 3 to 5 up to 6 to 8 by adding 3 to both terms instead of multiplying.

Sign rules by memory

Answers -3 minus -7 with -10, applying the multiplication rule to subtraction.

Four students, one item, four different reteaches

A score tells the interventionist that all four missed it. It does not say that one needs a fraction model, one needs place value, one needs a ratio table and one needs a number line. Reading the working to find out is the job the software takes over.

02 Inside the block

What a 30-minute Tier 2 math intervention session looks like

Minutes 0 to 3: start from yesterday's groups

Cards from the last session arrive sorted by misconception, so six students sit in two groups of two or three by what they got wrong, not by their overall score.

Minutes 3 to 20: students work, the tutor coaches

Each student works items from your material. When a step breaks, the tutor asks one question aimed at that step. In Socratic mode it never gives the corrected value.

Minutes 20 to 25: you read the skill view

The class view groups the new cards by skill. Two or three red skills across the group is the normal picture, not twelve separate problems to untangle.

Minutes 25 to 30: a five-minute reteach

Each card ends with a short reteach for the named misconception. You run it with the students who share it, and the next card shows whether it moved.

The IES guide gives its strongest evidence rating to two recommendations: explicit, systematic instruction, and word problems taught through their common underlying structures. The tutor moves are written to be explicit about the step that failed, and the cards are what keep the reteach systematic across a six to eight week cycle instead of starting from zero each Monday.

03 Example output

One grade 7 answer, one named misconception

The item

A recipe uses 2/3 cup of oil for every 4 cups of flour. How much oil do you need for 6 cups of flour?

The student wrote

"Flour went up by 2, so oil goes up by 2. 2/3 + 2 = 2 2/3 cups."

A score marks it wrong and moves on. The card shows the student set up the quantities correctly and added a whole number to a fraction without error, but reasons additively about a proportional relationship. That calls for a ratio table, not another round of fraction arithmetic.

Running the same diagnosis in whole-class practice rather than a pull-out group? That is the AI math tutor that names the misconception.

Session evidence card, grade 7

Fixed example

Diagnosis

Scales a ratio by adding, not multiplying

The student keeps the difference constant (flour 4 to 6, so oil 2/3 to 2 2/3) instead of keeping the ratio constant. Six cups is 1.5 times four cups, so the correct amount is 1 cup.

Next tutoring move / hint only

Four cups of flour needs 2/3 cup of oil. How much oil would 8 cups need? Now look at 6 cups: which of those two amounts should the answer sit between?

Skill mastery read

Setting up a ratio from a context

Secure

78%

Matched 2/3 cup to 4 cups and placed 6 cups against the unknown.

Proportional versus additive reasoning

Not yet

18%

Wrote that flour went up by 2, so oil goes up by 2.

Adding a whole number to a fraction

Secure

81%

Computed 2/3 + 2 = 2 2/3 without error.

Checking a result for reasonableness

Developing

38%

Did not notice 2 2/3 cups is four times the oil for 1.5 times the flour.

Teacher note

Sets up ratio problems correctly but scales by adding a constant, so any recipe, map scale or unit rate item with a factor other than 2 fails.

Reteach in 5 minutes

Double first, then find the middle

Build a ratio table for 4, 8 and 12 cups of flour.
Ask where 6 cups sits in the table and what that means for the oil.
Compare the table answer with the add-2 answer and ask which one keeps the recipe the same.

04 Research alignment

How the software lines up with the IES math intervention recommendations

The eight recommendations below come from the Institute of Education Sciences practice guide Assisting Students Struggling with Mathematics: RtI for Elementary and Middle Schools (Gersten et al., NCEE 2009-4060). The right-hand column matters as much as the one next to it.

IES recommendation	Evidence level	What Coeducate does	What it does not do
Screen all students and provide interventions to those at risk	Moderate	Nothing at the screening stage. It starts once a student is in a group.	It is not a universal screener. Keep yours.
Focus on rational numbers in grades 4 through 8	Minimal	The grade 6 to 8 skill vocabulary is built around fractions, decimals, ratio and percent.	It does not supply a scope and sequence.
Make instruction explicit and systematic	Strong	Tutor moves target the exact step that broke, and each reteach is tied to one named skill.	It does not replace the teacher-led part of the block.
Teach word problems through common underlying structures	Strong	Separates a structure failure (additive or multiplicative) from an arithmetic slip.	It does not generate a word problem curriculum.
Use visual representations	Moderate	Reteach steps call for a table, number line or diagram where the misconception needs one.	There are no interactive manipulatives.
Spend about 10 minutes per session on fluent retrieval of basic facts	Moderate	Not covered.	Use a dedicated fluency routine for those ten minutes.
Monitor the progress of students in intervention	Minimal	A dated per-skill record from every session, exportable as CSV.	It is not a normed curriculum-based measure.
Include motivational strategies in Tier 2 and Tier 3	Minimal	Hints keep the student doing the thinking, so progress is visibly theirs.	No points, badges or games.

Screen all students and provide interventions to those at risk

Evidence level: Moderate

Coeducate: Nothing at the screening stage. It starts once a student is in a group.

Not covered: It is not a universal screener. Keep yours.

Focus on rational numbers in grades 4 through 8

Evidence level: Minimal

Coeducate: The grade 6 to 8 skill vocabulary is built around fractions, decimals, ratio and percent.

Not covered: It does not supply a scope and sequence.

Make instruction explicit and systematic

Evidence level: Strong

Coeducate: Tutor moves target the exact step that broke, and each reteach is tied to one named skill.

Not covered: It does not replace the teacher-led part of the block.

Teach word problems through common underlying structures

Evidence level: Strong

Coeducate: Separates a structure failure (additive or multiplicative) from an arithmetic slip.

Not covered: It does not generate a word problem curriculum.

Use visual representations

Evidence level: Moderate

Coeducate: Reteach steps call for a table, number line or diagram where the misconception needs one.

Not covered: There are no interactive manipulatives.

Spend about 10 minutes per session on fluent retrieval of basic facts

Evidence level: Moderate

Coeducate: Not covered.

Not covered: Use a dedicated fluency routine for those ten minutes.

Monitor the progress of students in intervention

Evidence level: Minimal

Coeducate: A dated per-skill record from every session, exportable as CSV.

Not covered: It is not a normed curriculum-based measure.

Include motivational strategies in Tier 2 and Tier 3

Evidence level: Minimal

Coeducate: Hints keep the student doing the thinking, so progress is visibly theirs.

Not covered: No points, badges or games.

05 Comparison

Online math intervention programs compared: where software fits

Middle schools usually buy one of four things for Tier 2 math. They solve different problems, and plenty of schools need two of them. This is a category comparison, not a ranking of vendors.

Question	Structured intervention curriculum	Adaptive practice software	Tutoring provider	Coeducate
What you get	Lessons, materials and a sequence, often blended with digital work	Leveled practice with automatic scoring	Trained tutors, in person or virtual	An AI tutor over your material plus a per-skill mastery record
Explains why an answer was wrong	Through the teacher delivering it	Usually marks right or wrong and adjusts the level	Yes, if the tutor writes it down	Yes, a named misconception per skill
Who runs the block	Teacher or interventionist	The software, with a teacher monitoring	The provider's tutors	Your interventionist or a trained paraprofessional
Uses the material you already teach	No, you adopt theirs	No, their item bank	Varies by provider	Yes
Independent efficacy evidence	Varies, many publish studies	Varies by product	Varies by provider	None yet
How it is bought	Per-student or site license, often multi-year	Per-student license	Per-student or per-hour contract	$89 to $1,290 per month, PO on District

What you get

Structured intervention curriculum: Lessons, materials and a sequence, often blended with digital work
Adaptive practice software: Leveled practice with automatic scoring
Tutoring provider: Trained tutors, in person or virtual
Coeducate: An AI tutor over your material plus a per-skill mastery record

Explains why an answer was wrong

Structured intervention curriculum: Through the teacher delivering it
Adaptive practice software: Usually marks right or wrong and adjusts the level
Tutoring provider: Yes, if the tutor writes it down
Coeducate: Yes, a named misconception per skill

Who runs the block

Structured intervention curriculum: Teacher or interventionist
Adaptive practice software: The software, with a teacher monitoring
Tutoring provider: The provider's tutors
Coeducate: Your interventionist or a trained paraprofessional

Uses the material you already teach

Structured intervention curriculum: No, you adopt theirs
Adaptive practice software: No, their item bank
Tutoring provider: Varies by provider
Coeducate: Yes

Independent efficacy evidence

Structured intervention curriculum: Varies, many publish studies
Adaptive practice software: Varies by product
Tutoring provider: Varies by provider
Coeducate: None yet

How it is bought

Structured intervention curriculum: Per-student or site license, often multi-year
Adaptive practice software: Per-student license
Tutoring provider: Per-student or per-hour contract
Coeducate: $89 to $1,290 per month, PO on District

HMH Math 180, a blended program for grades 5 to 12, is one well-known example of the first column. If your school has no intervention material at all, buy a curriculum first. Coeducate reads the work students produce from material you already have, which is why it sits next to a curriculum rather than competing with one. The same records also feed the student progress tracking software view your MTSS team reads at the end of a cycle.

06 Being straight about it

When math intervention software is the wrong purchase

Your state or grant requires an ESSA evidence tier. Coeducate has no efficacy study and no evidence tier rating. If funding depends on one, spend that money on a program that has it.
The gap is basic fact fluency. If students mostly need fast recall of multiplication facts, a daily fluency routine does more than a diagnosis of reasoning they are not yet ready to use.
Students cannot type their working. The tutor reads written work. For students who cannot yet produce it independently, a human tutor in the room is the better spend.

07 Who uses it

Built for the people who run the intervention block

Math interventionist

Three groups a day, six students each. The card replaces re-marking the same papers after school to find the pattern, and the regrouping for tomorrow is already done.

Paraprofessional-led groups

The REL Northeast and Islands fact sheet notes that when groups of two to three are too costly, schools can train staff other than a certified teacher to deliver them. The card puts the diagnosis and the next question in writing for that person.

Math coach or MTSS lead

Decides at the end of a six to eight week cycle whether to continue, intensify or exit. Dated per-skill records with the student's reasoning attached make that meeting shorter.

Administrator who signs the PO

Needs the cost and a privacy posture that survives review. The student privacy and COPPA compliance page lists what we hold and, just as plainly, what we do not.

08 FAQ

Questions schools ask before buying math intervention software

Comparing costs across AI tutors as well? We worked through what Khanmigo costs a school for a 600-student budget.

What is the best math intervention program for middle school?

The best program is the one that matches why your students are behind. If they lack a sequence of lessons, a structured intervention curriculum fits. If they have lessons but keep failing fractions, ratios and percent for a different reason each time, software that names the misconception per student saves the most intervention time. Many middle schools run one of each.

How long should a Tier 2 math intervention session be?

Tier 2 math groups typically meet for 20 to 40 minutes, three to five times per week, according to a December 2025 fact sheet from REL Northeast and Islands. Coeducate has no minimum session length. A card is written from a single piece of work, so it fits whatever block your schedule already has.

What group size works best for Tier 2 math intervention?

Groups of two to three students showed the largest gains in the meta-analysis that REL Northeast and Islands summarized: 27 percentile points over similar students with no Tier 2 support, and 11 points over one-on-one instruction. Coeducate cards are per student, so the software works at any group size and makes regrouping by misconception a two-minute job.

Is Coeducate a research-based math intervention program?

No, and we will not claim it is. There is no independent efficacy study of Coeducate and no ESSA evidence tier rating. The session design follows the IES practice guide recommendations on explicit instruction, rational numbers and progress monitoring, and the alignment table on this page shows the recommendations it does not cover.

Can math intervention software replace our intervention curriculum?

Not this one. Coeducate does not supply lessons or a scope and sequence. It tutors students through the material you already use and records what each one understands. If your intervention block has no material yet, choose a curriculum first and add Coeducate for the diagnosis and the documentation.

How much does math intervention software cost?

Most intervention products are sold per student, and prices vary widely by vendor and contract length. Coeducate is priced per program: Classroom is $89 per month for one educator and up to 60 students, School is $349 per month for up to 600 students, and District is $1,290 per month with purchase order billing. Full detail is on the pricing page.

Does it work for Tier 3 and special education students?

It works for any student who can type their working, including Tier 3 groups, where the same cards arrive more often and on fewer skills. It does not make eligibility decisions and does not replace the assessments an IEP team runs. Teams can attach the dated records as work-sample evidence.

Is student work used to train AI models?

No. Student work is not used to train models, not sold and not used for advertising, and records are deletable on request. Coeducate is designed around FERPA and COPPA obligations, but it holds no SOC 2 report or signed state data privacy agreement today, and the student privacy page says so plainly.

09 Pricing

Priced per program, not per seat

Annual billing is about 17% lower. There is no free tier, and the privacy posture is identical on every plan.

Classroom

$89/mo

One teacher, up to 60 students, CSV mastery export.

School

$349/mo

Up to 10 educators and 600 students, roles, SSO, scheduled reports.

District

$1,290/mo

Up to 75 educators and 5,000 students, standards mapping, SIS export, PO billing.

Full pricing and Enterprise

More AI tools for teachers

Product pages

Run one intervention item through the Mastery Check

Paste a student's working on a fraction, ratio or percent problem and read the card it returns. Nothing you type is stored.

Student work is never used to train models. No card required to try the demo.

Try the demo