The Section That Gets Read First
When the visiting team sits down with your self-study report, they do not start at page one and read linearly. They go to the data summary first.
The data summary is where your program makes its quantitative case: how many courses address each graduate attribute, what percentage of evidence is current, how many continuous improvement cycles are closed, what the assessment results show over time. It is the evidence backbone that the narrative sections rest on.
If the data summary is weak, the narrative has nothing to stand on. If the data summary is strong, the narrative has a solid foundation to build upon.
This section determines whether the visiting team enters the rest of the report confident or skeptical. Getting it right is not optional.
What the Data Summary Is (and Is Not)
The data summary is not a collection of charts pasted from spreadsheets. It is a curated presentation of the most relevant quantitative evidence, organized to support the claims in your self-study narrative. Every number should have a purpose. Every table should answer a question the visiting team will ask.
Think of it as the executive summary of your evidence. It gives the visiting team a quantitative overview before they dig into individual courses, faculty profiles, or assessment instruments. It tells them: here is what we claim, here is the data that supports it, here is where we have room to improve.
Our guide to what to include in your self-study report covers the full document structure. This post focuses on the data summary specifically.
Five Essential Data Tables Every Summary Needs
CEAB does not prescribe a specific format for the data summary. However, experienced coordinators know that five tables appear in virtually every strong self-study. They cover the core areas the visiting team evaluates.
Table 1: Graduate Attribute Coverage Matrix
This is the centerpiece. It shows which graduate attributes are addressed in which courses, at what level, and with what evidence.
Minimum columns: Graduate Attribute | Courses Addressing It | Courses with Current Evidence | Progression (I/D/M) | Evidence Count
Example row: GA 1 — Engineering Knowledge | 18 courses | 16 current | I→D→M across years 1–4 | 42 pieces of evidence
What the visiting team looks for: Complete coverage across all twelve attributes. Multi-course coverage (not single-source dependency). Progression from introductory to mastery level. Evidence counts that are proportional to the importance of the attribute.
Our guide to mapping graduate attributes to courses walks through building this matrix step by step.
Table 2: Assessment Results Summary
This table summarizes assessment outcomes across the program. It shows what percentage of students meet the expected standard for each graduate attribute, based on direct and indirect assessment data.
Minimum columns: Graduate Attribute | Assessment Method | % Meeting Standard | Trend (3 years) | Action Taken
Example row: GA 5 — Communicate Effectively | Technical report rubrics | 78% | 72% → 75% → 78% | Added peer review component in ENGR 201
What the visiting team looks for: Actual numbers, not vague statements like "most students meet the standard." A three-year trend showing direction of travel. Actions taken when results fall below the expected threshold. Consistency between what the narrative claims and what the data shows.
Table 3: Continuous Improvement Cycles
This table documents the closed CI cycles: what was identified, what action was taken, and what improvement resulted.
Minimum columns: Finding | Date Identified | Action Taken | Date Completed | Evidence of Improvement
Example row: GA 3 assessment scores below target | Sept 2023 | Redesigned lab rubrics in ENGR 301 | June 2024 | Scores improved from 65% to 82% |
What the visiting team looks for: A reasonable number of closed cycles (three to five is typical for a four-year program). Each cycle should show a clear finding → action → result chain. Cycles that are still "in progress" should have a realistic timeline. Zero closed cycles is a red flag.
Our guide to continuous improvement cycles covers how to build a CI process that actually closes the loop.
Table 4: Evidence Currency Status
This table shows how current your evidence is. The visiting team will ask to see evidence from recent academic years. If most of your evidence is from three years ago, the data summary should acknowledge that and explain your replacement plan.
Minimum columns: Graduate Attribute | Evidence from Current Year | Evidence from Past 2 Years | Evidence Older than 2 Years | % Current
Example row: GA 7 — Function on Multi-Disciplinary Teams | 8 pieces | 12 pieces | 3 pieces | 86% current
What the visiting team looks for: A high percentage of current evidence (80%+ is comfortable). Any attributes with low currency should have an active replacement plan. Evidence that is older than two years should be flagged for replacement in the next academic year.
Our posts on mid-cycle evidence management and the annual accreditation review cover how to keep evidence current year after year.
Table 5: Faculty Qualifications and Teaching Load
This table summarizes faculty credentials, research activity, and teaching assignments. It supports the program's claim that faculty have the qualifications and capacity to deliver the curriculum effectively.
Minimum columns: Faculty Name | Highest Degree | Research Area | Courses Taught | Industry Experience | Professional Engagement
What the visiting team looks for: Faculty with terminal degrees in relevant fields. Research activity that connects to the curriculum. A manageable teaching load that allows for mentorship and curriculum development. Industry experience that enriches teaching.
Common Mistakes in Data Summaries
Even experienced coordinators make these errors. They are easy for the visiting team to spot, and they undermine the credibility of the entire report.
Mistake 1: Inconsistent numbers. The data summary says GA 5 is assessed in 8 courses. The narrative section says GA 5 is assessed in 6 courses. The coverage matrix shows 7 courses. The visiting team will notice. They will assume the program does not have a handle on its own data.
Fix: Create a single source of truth for all numbers. Cross-check every table against every other table and against the narrative before finalizing the report.
Mistake 2: Numbers without context. "78% of students meet the standard for GA 5." Is that good? Bad? The visiting team does not know what standard you are measuring against, or what percentage your program considers acceptable.
Fix: Always state the target percentage alongside the actual percentage. "78% of students meet the standard for GA 5 (target: 75%). The program considers 75% the minimum acceptable threshold, based on CEAB Criterion 4 expectations."
Mistake 3: Only presenting positive data. Every program has areas where results fall below expectations. A data summary that shows 100% success across every attribute reads as unrealistic. The visiting team expects to see some gaps — what matters is how the program responds to them.
Fix: Include the gaps. Show the below-target results alongside the action plan. This demonstrates honesty, self-awareness, and a culture of continuous improvement.
Mistake 4: Data that does not match the evidence. The data summary claims strong coverage for GA 3. The visiting team asks to see the evidence for GA 3 in ENGR 301. The evidence folder contains a syllabus from 2021, a rubric from 2022, and no student work samples.
Fix: Before writing the data summary, verify that the evidence exists for every number you include. A claim without supporting evidence is worse than no claim at all.
Mistake 5: Stale trends. The trend data shows three years of results — 2019, 2020, 2021. The self-study is being written in 2025. The data is four years old.
Fix: Use the most recent three years of data available. If the program has been collecting assessment data for less than three years, say so. Explain what data is available and what data is still being collected.
How to Organize the Data Summary
Here is a practical structure that works well for most four-year engineering programs:
Section A: Program Overview — Brief description of the program, number of courses, number of faculty, enrollment by year. This gives the visiting team context for the numbers that follow.
Section B: Graduate Attribute Coverage — Table 1 (coverage matrix). Shows breadth and depth of coverage across all twelve attributes.
Section C: Assessment Results — Table 2 (assessment summary). Shows whether students are meeting the expected standards, with three-year trends.
Section D: Continuous Improvement — Table 3 (CI cycles). Shows the program's capacity to identify problems and implement solutions.
Section E: Evidence Currency — Table 4 (currency status). Shows how current the evidence base is.
Section F: Faculty — Table 5 (faculty qualifications). Shows that the program has qualified faculty to deliver the curriculum.
Section G: Limitations and Improvement Plan — Honest acknowledgment of gaps, with specific actions and timelines. This section is often overlooked, but it is the one that convinces the visiting team that the program is mature enough to identify its own weaknesses.
Our guide to proactive risk management covers how to identify gaps before the visiting team does — the same discipline applies to writing an honest limitations section.
A Data Summary Template
Here is a fill-in-the-blank structure you can adapt for your program:
Program: [Program Name] | Academic Year: [Year] | Report Version: [Number]
A. Program Overview
Total courses: [Number] | Faculty FTE: [Number] | Enrolment (current year): [Number] | Graduation rate (3-year avg): [Percentage]
B. Graduate Attribute Coverage
[Table 1 — see format above]
Summary: All twelve graduate attributes are addressed across [Number] courses. Each attribute has evidence from at least [Number] courses spanning [Number] years of study.
C. Assessment Results
[Table 2 — see format above]
Summary: [Number] of twelve attributes exceed the target threshold. [Number] attributes are at or near the target. [Number] attributes are below target and have active improvement plans.
D. Continuous Improvement
[Table 3 — see format above]
Summary: [Number] CI cycles have been closed in the past [Number] years. [Number] cycles are in progress. The average time from identification to closure is [Number] months.
E. Evidence Currency
[Table 4 — see format above]
Summary: [Percentage]% of evidence is from the current or previous academic year. [Number] attributes have currency below 80% and are flagged for replacement.
F. Faculty
[Table 5 — see format above]
Summary: [Number] of [Number] faculty hold terminal degrees in relevant fields. [Number] faculty have industry experience. Average teaching load is [Number] courses per term.
G. Limitations and Improvement Plan
The following areas require attention:
[Gap 1]: [Description]. Action: [Specific action]. Owner: [Name]. Target: [Date].
[Gap 2]: [Description]. Action: [Specific action]. Owner: [Name]. Target: [Date].
When the Spreadsheet Is Not Enough
The data summary requires pulling numbers from multiple sources: the coverage matrix, assessment databases, CI trackers, faculty files, evidence inventories. In a spreadsheet-based workflow, this means opening six different files, copying numbers, pasting them into a report document, and cross-checking for consistency.
It is error-prone. The numbers in the coverage matrix spreadsheet rarely match the numbers in the assessment database because they were updated at different times. The CI tracker is maintained by a different person. Faculty files are in a shared drive folder that was last updated two years ago.
A connected evidence map eliminates these inconsistencies. Every number comes from the same source. The coverage matrix, assessment results, CI cycles, and evidence currency are all views of the same underlying data. When you update one piece of evidence, every table that references it updates automatically.
The data summary becomes a one-click export rather than a manual compilation. Numbers are consistent. Trends are calculated automatically. Gaps are flagged in real time, not discovered at self-study time.
Our posts on moving from spreadsheets to an evidence map and the hidden cost of spreadsheet accreditation cover the full migration in detail.
A Practical Starting Point
If you are preparing a self-study report and the data summary feels like a blank page, start with Table 1: the graduate attribute coverage matrix. It is the foundation that every other table builds on. If you know which courses address which attributes, you can derive the assessment results, the evidence currency status, and the CI cycle needs.
Spend one afternoon building Table 1. Use the template above. Involve one faculty member who knows the curriculum well. Cross-check every number against the actual evidence files.
Once Table 1 is solid, the rest of the data summary follows logically. The assessment results table fills in the performance data. The CI cycles table shows how the program responds to gaps. The evidence currency table shows the health of the evidence base.
A complete data summary does not take weeks. It takes a focused afternoon and a disciplined approach to the numbers. The visiting team will thank you for it.
Summary
The data summary is the quantitative backbone of your self-study report. It turns scattered evidence into clear, credible numbers that support your narrative claims. Five tables cover the essentials: coverage, assessment results, CI cycles, evidence currency, and faculty qualifications.
Avoid the common mistakes: inconsistent numbers, data without context, presenting only positive results, claims unsupported by evidence, and stale trend data. Acknowledge gaps honestly. Show the visiting team that your program can identify its own weaknesses and take action to address them.
The template in this post gives you a starting point. Adapt it to your program, fill it in with real data, and cross-check every number before you submit the report. A strong data summary makes the rest of the self-study easier to write — and easier for the visiting team to evaluate.