It's common for a teacher or administrator to pull up a student's data and see two different Amira Reading Mastery (ARM) scores, one from a Benchmark assessment and one from Amira Tutor, and wonder which one is "right." Both are correct. They're just measuring different things, over different time windows, using different amounts of data.
Here's the short version:
- Assessment ARM = a snapshot score from one complete assessment (the Screener and ORF tasks administered during that testing window).
- Tutor ARM = an oral reading fluency (ORF) score calculated from each tutoring passage the student reads.
- Estimated Mastery ARM = a composite score that factors in every assessment and every tutoring activity the student has completed across the school year, updated continuously as new data comes in.
Which one to use depends on what you're trying to do, see "So which score should I use?" below.
Why the scores don't line up
1. They're built from different amounts of evidence
A Benchmark ARM score comes from a single, complete administration of the Screener and its ORF (Oral Reading Fluency) task. It's a controlled, point-in-time measurement, the same kind of snapshot you'd get from any standardized assessment.
Amira Tutor's ARM score, on the other hand, isn't meant to be a formal assessment. It's calculated each time from the oral reading fluency (ORF) data in that session's tutoring passage. That makes it highly responsive to how a student read on that specific passage, but also means it can shift more than a Benchmark score, and it isn't drawing on the same breadth of skills.
2. They're measuring different moments in time
Because Tutor ARM is recalculated from each tutoring passage, it reflects how a student read in that specific session, which is exactly what makes it useful for day-to-day instructional decisions. A Benchmark score, by contrast, reflects performance during one specific testing window. If a student had a rough day during Benchmark testing, or has grown a lot since their last Benchmark, the two numbers can diverge quite a bit even though neither one is "wrong."
3. Different tasks contribute to each score
Benchmark ARM and Tutor ARM aren't built from the same ingredients. Benchmark ARM blends several literacy skills, phonological awareness, decoding, vocabulary, comprehension, spelling, together with oral reading fluency (ORF). Tutor ARM, by contrast, is driven by ORF alone.
ORF and the other literacy skills are highly correlated for most students, so most of the time Benchmark ARM and Tutor ARM will land close together. But they're not the same thing, and some students genuinely perform differently across skills. A student who reads fluently but struggles with comprehension or spelling will tend to score higher on Tutor ARM (ORF-driven) than on Benchmark ARM (which pulls those weaker skills into the blend), and a student with the opposite profile will show the reverse pattern. That's a real signal, not noise: if you see it, the Student Summary Report's skill-level breakdown can show you which specific skill is driving the gap.
So which score should I use?
If you're trying to answer "where is this student today?", Estimated Mastery is the best single score to use. It's the only score built from both Benchmark's multi-skill evidence and Tutor's ORF evidence, pooled together across the school year, so it reflects a broader, more current picture than either source on its own.
A couple of things to keep in mind alongside that:
- If a student just took a Benchmark, or you're reporting for state/high-stakes purposes, the Benchmark/Screener ARM is worth using directly, it's a fixed, complete measurement from that testing window, which is exactly what those use cases call for.
- Because Tutor's contribution to Estimated Mastery is ORF-driven, Estimated Mastery leans somewhat more on fluency than the Benchmark ARM does. That makes it a strong choice for tracking fluency growth specifically, but worth remembering if you're using it as a stand-in for a student's overall skill profile.
| Score | What it's based on | Best used for |
|---|---|---|
| Benchmark/Screener ARM | Full Screener battery (phonological awareness, decoding, vocabulary, comprehension, spelling) + ORF | Point-in-time comparisons across a testing window (BOY/MOY/EOY), norm comparisons |
| Tutor ARM | Oral reading fluency (ORF/WCPM) on that session's tutoring passage, calculated fresh each session, not averaged over time | Monitoring fluency trends within Tutor |
| Estimated Mastery ARM | Every assessment and tutoring activity across the school year | The best single answer to "where is this student today?" |
If a student's ARM looks lower on assessments than in Tutor
If you notice a student's ARM is consistently lower on formal assessments than what you're seeing come through in Tutor, that's worth digging into. The Student Summary Report has granular, skill-level data that can help you see why, for example, whether the gap is concentrated in a specific construct (decoding, fluency, comprehension, etc.) rather than showing up evenly across the board. That skill-level breakdown is often more useful for planning instruction than comparing two composite scores against each other.