M-STEP / Smarter Balanced Concordance
State as of April 2026: two draft layouts reviewed, substantial committee feedback on communication, one unresolved placeholder, and a suggestion that would make the whole artifact unnecessary.
What it is
A table mapping Smarter Balanced interim scale scores to M-STEP scale scores, by subject (ELA, Math), grade (3–7), and performance level (1–4). Each cell gives lowest and highest scale score on each side. Built by OEAA to promote use of the interim assessments — see Smarter Balanced Interim Adoption.
The M-STEP assesses the same content as Smarter Balanced but its scale scores have been converted, so the two are not directly comparable without a crosswalk.
The committee’s reading
The committee treated this as a communication artifact and reviewed it accordingly. Its feedback clusters into three concerns.
Say what the number means. The worked example says a student “could score” a given M-STEP score. Asked what could means, the answer was “student would most likely score, or something along those lines.” A concordance table’s entire job is to convey a probabilistic mapping to people who are not psychometricians, and “or something along those lines” is not a definition. The table also carries its own disclaimer that it “is not a perfect prediction,” without quantifying the imperfection.
Put the conclusion first. Move key takeaway and performance level to the top of the concordance document — key takeaway and performance level information belong at the top of the document, not after the explanatory narrative.
Show the work. Publish the concordance formula, not just the table — districts and schools will want the formula used, not only the lookup table. Extended in Build a tool that accepts a theta value to a tool where a district could enter a theta value directly.
The suggestion that would obviate it
Ask Smarter Balanced whether interim reports can carry the M-STEP score. The committee asked how much the interim reports themselves can be modified and whether the corresponding M-STEP score could simply be inserted. The answer was that this is a question for Smarter Balanced.
If Smarter Balanced can embed the M-STEP concordant score in the interim report, the concordance table stops being a document educators must find, read, and apply correctly, and becomes a number that appears where they are already looking. That is a strictly better outcome than any layout revision and it remains unasked.
The intended workflow
Students take the interim assessment; scores come from Data Recognition Corporation (DRC); educators consult the concordance table. Baseline administration is three times per year though it varies, and a student score is returned each administration.
The committee asked what other systems do and was told they make predictions of student performance on the summative test. OEAA’s stated preference is for the present-tense framing — what a student would score today — over an end-of-year projection, on the grounds that it better supports acting on the result. That framing preference is defensible and is not currently stated in the document itself.
Unresolved
Both draft examples end with “Additional blueprint information can be found at XXXXXX.”
A route to quantifying “could”
Michigan holds interim and M-STEP scores on many of the same students, which is what makes a concordance possible at all — but the table maps band to band, discarding the joint distribution. copula-approach-to-subtest-indicators observes that a copula fitted to that pair yields the conditional distribution of the M-STEP score given an interim score, turning “could score” into a statement with a measured spread. Same machinery as the subscore problem, different pair, data already collected.
History
| Meeting | What moved |
|---|---|
| Session 1 — Using the Smarter Balanced Interim Assessment | Two layouts reviewed; feedback on ordering, formula transparency, and the meaning of “could”; Smarter Balanced embedding raised as the preferable path |