The Paperwork Conversion: When OBE Is Adopted in Form Only
A regional state university has committed to outcome-based education to satisfy its accreditor. The dean circulates a memo: every course must have five course learning outcomes uploaded to the LMS by Friday. Two semesters later an external review team reads the files. Of 214 syllabi sampled, 168 list CLOs beginning with 'Understand' or 'Know'. The program educational objectives were drafted in a single afternoon by three associate deans; no alumni, employer, or student input exists. Assessment practice is unchanged: final exams worth 60-80% of the grade, no rubrics. Attainment reports show 90% or better for every outcome in every program. Faculty say privately that OBE is 'paperwork for the visit' and that with 18 contact hours a week they have no time to redesign anything. The dean wants a one-semester fix before the accreditation team arrives.
1. The dean asks, 'Are we compliant?' Give the judgement you would actually deliver, and name the specific OBE principles the current practice violates, justifying each from the evidence in the setup.
Read the worked answer
Judgement: compliant in form, not in substance. The artefacts exist; the design logic does not. A defensible answer separates the two.
Principles violated, with evidence:
- Clarity of focus. Outcomes must state what the learner can do, observably. 'Understand' and 'Know' are not observable and cannot be assessed, so 168 of 214 courses have no assessable outcome at all.
- Design-down, deliver-up. Outcomes are supposed to drive assessment and instruction. Here assessment is unchanged (60-80% exams, no rubrics), which is proof that the outcomes were written after the course and attached to it, not used to design it.
- Ownership by the community of interest. PEOs written by three administrators in one afternoon have no constituency. OBE outcomes are a promise made to stakeholders; a promise nobody outside the room knows about is not a promise.
- High expectations and credible evidence. 90%+ attainment on every outcome, with no instrument capable of evidencing any outcome, is not a finding. It is a measurement artefact. Attainment computed from exams that do not sample the outcome tells you nothing, and uniformity across all outcomes is itself a red flag.
What I would not say: that the university is 'not doing OBE' and should start over. The instinct to restart is what produces the next round of paperwork. The structural problem is that OBE has been routed through the document pipeline instead of through assessment design.
2. Design a realistic one-semester intervention, given 18 contact hours per week and no new budget. State explicitly what you would not do, and why.
Read the worked answer
Principle: leverage a few real cases, not blanket compliance. The failure mode of remediation is the same as the failure mode of the original rollout.
Do:
- Pick two courses with high enrolment and a high failure rate. Nothing spreads faster than a worked example in a course colleagues already worry about.
- Run a single 2-hour session whose output is an artefact, not a discussion. Faculty leave with (a) three rewritten CLOs using observable verbs and (b) one 4-level analytic rubric for the course's major assignment. With 18 contact hours a week, a session that does not produce something usable in next week's class will not be repeated.
- Recompute attainment for those two courses only, per CLO, from rubric-mapped items, and publish the result even if it is 55%. An honest baseline buys more credibility than a fake 90%.
- Run one 90-minute employer/alumni focus group to test the PEOs. Not a survey, not a retreat.
Do not:
- Do not rewrite all 214 syllabi. It consumes the entire semester and changes nothing in a classroom.
- Do not mandate a format conversion (e.g., all exams to performance tasks). It generates resistance and produces compliance theatre.
- Do not run a 3-day retreat. Attendance will be coerced and the cynicism will harden.
- Do not touch the PEOs further than the focus group. Rewriting them again without stakeholder evidence repeats the original error at a higher cost.
Timeline: weeks 1-3 pilot, 4-9 implementation, 10-13 analysis and one dissemination session.
3. The dean asks whether to submit the 90% attainment figures to the accreditor, since 'everyone does it'. Advise, with reasoning an accreditation-savvy reader would accept.
Read the worked answer
Advise against submitting them as attainment evidence. The reasoning is not primarily ethical, it is evidentiary.
- An attainment percentage has the form numerator / denominator, where the numerator is students meeting a criterion on an instrument that samples the outcome. Here there is no criterion and no sampling argument, so the number is not interpretable. A reviewer who asks 'what instrument, what criterion, what evidence source?' will get an answer that collapses the claim.
- Presenting invalid evidence is worse than presenting no evidence, because it converts a maturity gap into a credibility problem. Reviewers routinely reward an honest baseline plus a dated plan; they rarely reward implausible success.
What to submit instead:
- A transition statement naming which courses now have rubric-based, CLO-mapped data (the two pilot courses) and reporting those honestly.
- A remediation plan with owners, dates, and the next measurement point.
- A withdrawal of the programme-wide attainment table, with a one-line explanation that the prior method did not sample the outcomes.
This converts the visit from a defence into a demonstration of exactly the improvement capability the framework is asking about.