The New Jersey Student Learning Assessments (NJSLA) are the state's annual standardized tests, administered in New Jersey public schools to measure how well students are meeting the New Jersey Student Learning Standards (NJSLS). The 2026 administration introduced some of the most significant structural changes in years. Here is how the NJSLA is now organized, and what each part measures.

1. The move to adaptive testing

Beginning with the Spring 2026 administration cycle, New Jersey transitioned the English Language Arts (ELA) and Mathematics portions of the NJSLA to a computer-adaptive format, now known as NJSLA-Adaptive (NJSLA-A).

  • How it works: Rather than every student sitting the same fixed test form, the testing software adjusts the difficulty of upcoming question groups based on how the student has performed so far. The adjustment isn't made after every single question; it happens once enough responses accumulate within a group (a testlet) to refine the system's estimate of the student's ability, at which point the next group of questions is set harder or easier accordingly.

For the exact, unit-by-unit structure of each test, along with the most current administration details, refer to the official NJDOE Assessments page.

2. Subjects and grades

  • English Language Arts (ELA): Tested annually in grades 3 through 9. It focuses heavily on close reading and analyzing text, and includes a standalone extended writing task.
  • Mathematics: Tested annually in grades 3 through 8, plus high-school end-of-course exams for Algebra I, Geometry, and Algebra II.
  • Science (NJSLA-S): Unlike ELA and Math, the science test remains a fixed-form assessment: every student answers the same questions. It is administered only to students in grades 5, 8, and 11.

3. NJSLA vs. NJGPA

The two are easy to confuse, but they serve different purposes:

  • NJSLA is for general progress tracking across grades 3–9.
  • NJGPA (New Jersey Graduation Proficiency Assessment) is the test high-school juniors (grade 11) must pass in ELA and Math to meet state graduation requirements. It has also adopted the new adaptive format, the NJGPA-Adaptive (NJGPA-A).

4. How to read the results

NJSLA results are reported as one of five performance levels. For the adaptive test, NJDOE’s NJSLA-Adaptive FAQ lists them as Did Not Yet Meet Expectations, Partially Meet Expectations, Approaching Expectations, Met Expectations, and Exceeded Expectations. The fixed-form NJSLA used the same five steps in the past tense — “Approached expectations,” “Met expectations.” A report spanning both years will therefore show two different wordings for what is the same rank, which is worth knowing before anyone tries to reconcile them.

The proficiency line is not in the middle of the scale. NJDOE’s School Performance Reports reference guide counts students as proficient if they have met or exceeded expectations — a performance level of 4 or 5. Level 3 is below the line, and a district that reads it as close enough is reading the scale wrong.

One caution specific to 2026: the adaptive transition introduced a new scoring scale. NJDOE’s standard-setting presentation states that prior fixed-form cut scores “no longer apply” and that new performance boundaries had to be established before results could be reported. Standard setting fixes where the cuts fall on the new scale; it does not express the new scale in terms of the old one. Those are different pieces of work, and only the first one has happened.

What the adaptive shift means for your data

Format changes like the move to adaptive testing are what make year-over-year analysis difficult. New units, renamed columns, and altered data schemas mean a file from 2026 rarely lines up cleanly with one from 2024 or 2025, and merging them by hand in a spreadsheet quickly becomes unmanageable. DataVot validates each official file against the correct testing blueprint and lets you map differently-named columns from each year to a shared field, so historical results stay comparable even as the state evolves the assessment — the column matching is a decision you make once, not something inferred silently. (See also: Beyond the Spreadsheet and Multi-Year NJSLA Comparisons Start With a Column-Naming Problem.)