The PAT is one of the two scored sources behind a school's Roost Rating. For a parent, the short version is this: a Provincial Achievement Test tells you how a school's Grade 6 or Grade 9 students did on Alberta's curriculum that year. That is useful, but it reflects the families a school enrolls at least as much as the teaching inside it. We use it as a baseline signal about your catchment school, averaged over three years and shown alongside its context, never as a verdict.
This page explains what the test is, how we read it, and where we hold it at arm's length.
What a PAT is (and isn't)
Provincial Achievement Tests are standardized exams Alberta administers in Grade 6 and Grade 9. (A Grade 3 PAT existed until it was cancelled in 2014.) Grade 6 students write up to four core subjects (English Language Arts, Math, Science, and Social Studies) with French-language versions for Francophone and immersion streams.
Each result is reported against two standards:
| Standard | Rough anchor | What it means |
|---|---|---|
| Acceptable Standard | ~50% | The student met the grade's curriculum expectations |
| Standard of Excellence | ~80% | The student showed in-depth, above-grade understanding |
Those percentages are conceptual anchors, not literal pass marks: Alberta Education re-equates the cut scores every year so that a given standard means the same thing even when one year's test is harder than another's. (Alberta's interpretation guidelines spell this out.)
This matters for which number we use. Alberta publishes both the percent of students at the Acceptable Standard (equated, so it is comparable across years) and a raw course average (not equated, so it is not).
We use both, deliberately, in a 70/30 split: 70% the percent at Acceptable Standard, 30% the average mark. The equated share carries most of the weight because it is the one that compares honestly across years. The average mark is there for stability rather than accuracy: it moves less from one year to the next, and a single year of the Acceptable share can swing about 11 points on a small cohort. Weighting it at 30% steadies the number without letting a non-comparable statistic drive it.
Why a three-year mean
A single year of PAT data is noisy. A small Grade 6 cohort, one disruptive test day, a wave of absences, or a difficult test form can swing a school's number several points in either direction. Alberta's own guidance is to read the trend across multiple years, not a single snapshot.
So both statistics are read as a three-year mean, averaged across the available Grade 6 subjects and weighted by the number of students who actually wrote each exam. This dampens single-year noise and matches how Alberta's own guidance says to read the data. Where a subject-year is missing, we average over the years we have rather than counting the gap as a zero.
Recent years carry real gaps you should know about:
- 2019–20 and 2020–21 PATs were cancelled (COVID). No reportable data.
- 2023–24 Grade 6 Math and English Language Arts were not administered (curriculum field-testing); Science and Social Studies ran.
- 2024–25 was the first operational year of the new Grade 6 Math curriculum, and the province-wide Acceptable rate fell to about 53%, down from 72% in 2019. That is a curriculum shift, not necessarily a school-quality shift.
We treat a recent "three-year mean" as what it is. Two to three years of partial data with a known difficulty break means reading the trend with care rather than as breaking news.
Participation and cohort-size caveats
A PAT percentage only describes the students who actually wrote.
- The 10% rule. Alberta's guidance is that if more than 10% of a school's students didn't write, "these results may not be representative." Calgary Grade 6 participation usually runs ~88–95% in normal years, but we flag schools that fall below the line.
- Small cohorts are masked. Results for fewer than 6 writers are reported privately to the school and not published. A small school with a tiny Grade 6 simply won't have a public number.
- The denominator can move. Exemptions are tightly controlled (superintendent-level, individual basis), but absences and discretionary excusals can shrink the writing group, which is one more reason a single year's figure deserves caution.
When a school has no usable PAT data, we show it as PND ("pending") and lean on the other sources. We never silently convert "no data" into a low score.
The structural gap: schools with no PAT, ever
Because PATs are written in Grades 6 and 9, schools whose grades end at Grade 4 or Grade 5 never have PAT writers: the students leave before they'd ever sit one. This isn't a bug; it's structural. In our Calgary dataset, a substantial share of CBE elementaries have no PAT data at all: they are predominantly K–4 or K–5 feeder schools whose students leave before Grade 6, with a handful masked by the small-cohort rule or too newly opened to have a Grade 6 cohort yet.
The same cutoff is why Fraser rankings omit these schools entirely. We don't. For a K-4 catchment school the exams simply do not apply. We still show its survey results, but we do not turn a survey alone into a letter, so most of these schools read as pending rather than graded. That is a deliberate choice, not missing data.
Where the test meets income
Here is the part single-number rankings tend to skip. School PAT results correlate strongly with neighbourhood income: a school's scores reflect the families it enrolls at least as much as the teaching inside it. It is one of the most consistent findings in the research on standardized testing: peer-reviewed analyses of Canadian results have attributed a large share (on the order of a third to 40%) of the measured differences between schools to household income alone (Raptis, Canadian Journal of Education, 2012).
That doesn't make the score meaningless: most of the variation still reflects genuine school effect, family factors beyond income, and ordinary noise. But it means a high PAT in an affluent catchment reflects who lives there as much as what happens in the classroom. Two further limits compound it:
- PAT measures a level, not growth. It captures where students are in Grade 6, not how far the school moved them. Alberta publishes no school-level value-added measure, so "school effect" and "intake effect" can't be cleanly separated.
- It's a narrow construct. Timed tests on core subjects can't see citizenship, collaboration, creativity, or the untested parts of the curriculum, which are the things the survey source (AEAM) is there to surface.
This is the heart of why a test result is read in context here rather than taken at face value. (See our critique of single-number rankings for the fuller argument, and a note that the Fraser proxy for student background uses only household income, not parental education or family structure.)
High scores don't always mean a happy school
Test results and how a school feels to its families can point in opposite directions.
Illustrative: in our own analysis we found a Calgary middle school whose Grade 6 students were well above the provincial average on the PAT, yet whose parent, student, and teacher survey measures came back below province on engagement, education quality, and access to supports. The students were clearing the test bar; the institution around them was struggling.
This is exactly why the Roost Rating weighs the survey source heavily and why we publish each source with its own date. A strong test number is a reason to ask better questions on a school tour, not a reason to skip the visit.
One caution about specific numbers: comparing a school's PAT across published sources, we've seen the same school's figure differ depending on the year, the subject mix, and whether the source used "percent Acceptable" or a raw course average. When sources disagree, we reconcile against Alberta Education's equated percent at Acceptable Standard and show our data date, rather than repeat a contested figure as if it were settled.
How to read a PAT number on FamilyRoost
How much of the grade it decides: 37.5%, at a school that also has survey results. At a school with no survey results, the exams decide the whole grade, and the school's profile says so in as many words.
- Prefer the three-year trend over any single year, especially across the disrupted recent years.
- Check participation. A number built on a small or partial cohort is a softer signal.
- For a K-4 catchment school, expect no exam results at all and read the survey instead.
- If the exams are the only source your school has, ask it what the exams do not show: a test-only grade tracks neighbourhood income more closely than one with survey results behind it.
This page explains methodology only. A school's actual PAT figures come from the live dataset and show their own data-freshness date on each school profile.