Solutions

Assessment

Testing that tells a teacher what to reteach on Monday, not just what a student scored in May.

Direct answer
Edves assessment provides computer-based testing with items tagged to state standard codes and cognitive level, supporting formative, interim and benchmark cycles with mastery tracking and subgroup analytics.

It includes 15,000+ items across state assessment frameworks and lets schools add their own into the same tagged structure, so classroom assessment and state test preparation report from one bank rather than two disconnected ones.

The problem with most assessment data

A typical U.S. school has plenty of assessment data and very little that changes a lesson. The state summative arrives after the students have moved on. The benchmark reports a percentile. The unit test reports a topic score. None of it answers the only question a teacher has on Tuesday morning: which specific skill did this section not secure, and what do I do about it tomorrow?

The difference is granularity. A topic score says the class found fractions hard. An objective-level score says the class can order fractions but cannot compare unlike denominators. Only the second is actionable, and only the second aggregates into anything useful at the district level.

What it covers

Computer-based testing

Scheduled and on-demand assessment delivered on web or mobile, with randomized item selection, time controls, accommodation support including extended time and text-to-speech, and offline resilience for buildings where connectivity is uneven. Objective items score automatically; extended response is rubric-assisted with the teacher moderating.

Standards-tagged item banks

15,000+ items mapped to standard codes and tagged by cognitive demand, aligned to Common Core, NGSS and state standards including TEKS, B.E.S.T. and Georgia Standards of Excellence. Schools add their own items into the same structure, so a teacher-written common assessment and a state-aligned benchmark report against the same codes.

Assessment cycles

CyclePurposeFrequency
FormativeAdjust the next lessonDaily to weekly, low stakes
Common assessmentCompare sections and calibrate teaching across a grade teamPer unit
Interim / benchmarkPredict state assessment performance and target interventionTwo to four times a year
SummativeCertify attainmentEnd of course or year

Most schools over-invest in the third row and under-invest in the first. The formative cycle is where instructional change actually happens.

Mastery tracking and MTSS

Performance is tracked against standard codes rather than topics, which is what makes the data usable in a Multi-Tiered System of Supports. An MTSS team looking at a Tier 2 candidate can see the specific standards a student has not secured across the year, not a composite score that hides the pattern.

Subgroup analytics

Attainment and growth disaggregated by the subgroups ESSA requires — race and ethnicity, economic disadvantage, English learner status, disability status — available continuously rather than in a June report. A gap that opens in October is worth seeing in October.

Predictive flags

Students trending away from a target performance level are flagged. These are built as a prompt to look, not a verdict. A model trained on historical outcomes will reproduce historical inequity unless a human interrogates it, and in a system with documented disproportionality that is a live risk rather than a theoretical one. Flags surface the evidence behind them so they can be argued with.

State assessment alignment

Item alignment runs against the assessment your students actually sit — STAAR in Texas, CAASPP in California, FAST in Florida, Georgia Milestones in Georgia, Regents in New York, Smarter Balanced across a dozen states, and every other state program. Alignment includes item format and cognitive demand, not just topic coverage, because a student who can answer a multiple-choice question about a standard may not be able to answer the constructed-response version of it. See all 50 states.

Assessment when students have AI

Any assessment completed outside supervision no longer reliably evidences the student. Three practical responses, in order of durability:

  • Assess process, not only product. Drafts, checkpoints and revisions carry evidence a final artifact does not.
  • Require defense. A short oral or written justification of the choices made is difficult to fake and quick to administer.
  • Shift weight toward frequent supervised formative assessment. Ten short in-class checks give a more reliable picture than one take-home artifact of uncertain authorship.

Detection tools are not on this list. Their false positive rates make them unsafe as the basis of an academic integrity finding. The postsecondary version of this thinking is the Innovation Challenge Studio.

How assessment connects

Results feed directly into lesson planning: standards a section has not secured become the standards the next sequence addresses. Attainment reaches families through family engagement as it is published, rather than waiting for a report card.

Common questions

Frequently asked

Which state assessments does Edves align to?

All of them, including STAAR (Texas), CAASPP (California), FAST (Florida), Georgia Milestones, Regents Examinations (New York), Smarter Balanced across multiple states, MCAS (Massachusetts), ILEARN (Indiana), M-STEP (Michigan), TCAP (Tennessee) and SOL (Virginia). Alignment covers item format and cognitive demand, not only topic coverage.

What is mastery tracking?

Mastery tracking measures performance against specific standard codes rather than broad topics, so a teacher learns which precise skill a section has not secured rather than only that a unit went badly. It is also what makes assessment data usable in an MTSS process.

Can teachers write their own assessment items?

Yes. School-written items go into the same standards-tagged structure as the provided bank, so a teacher-written common assessment and a state-aligned benchmark report against the same codes.

Does Edves support testing accommodations?

Yes, including extended time, text-to-speech and other accommodations configured against a student's IEP or 504 plan on the roster.

Are predictive flags reliable?

Predictive flags are designed as a prompt to look, not a verdict, and they surface the evidence behind them so they can be challenged. Any model trained on historical outcomes risks reproducing historical inequity, so human review of flagged students is built into the workflow rather than optional.

How should schools assess when students have access to AI?

The durable responses are assessing process rather than only the final product, requiring students to defend their choices, and shifting weight from take-home tasks toward frequent supervised formative assessment. Detection tools are not reliable enough to base an integrity finding on.

See it run on your own standards.

A 20-minute walkthrough using your state’s standards, your evaluation framework and your reporting requirements.

Schedule a demo

Keep reading