River
Y CombinatorBacked by Y Combinator
FREE TEMPLATE

Learning Outcomes Examples and Template

Five documents and three sheets that place every outcome against your framework's descriptor rows and cap it at what your assessment can evidence.

Free download  ·  No account needed

Outcome Register

Three levels per outcome, not one

The verb's level, the ceiling of the evidence behind it, and the lower of the two. A gap is a number.

COVerbVerb levelAssessmentCeilingRealisedRuleVerdict
CO-1
CO-2
CO-3
CO-4
CO-5
CO-6

Framework coverage, both directions

RowOutcomes claiming itEvidenced at the ruleClauses namedAssessed, never stated
(1)
(2)
(3)
(4)

A verb list cannot know your descriptor rows, and a sentence formula cannot know whether your assessment reaches the level it helped you claim.

Every guide for this query gives you the same two things: the ABCD sentence formula, and a Bloom verb list. Both are generic by construction, so neither can settle the two questions that decide whether an outcome survives review. What level does your framework's descriptor row actually require, and can the assessment you already have produce evidence at that level? An outcome that fails the second question reads exactly like one that passes.

Kenniston Institute of Technology, ME 341 Heat Transfer, a 300 level course in an accredited engineering programme whose curriculum policy asks 300 level outcomes to reach Analyse. Eight outcomes as inherited. Three lead on understand, be familiar with and know, so they carry no level at all. Of the five that can be placed, three meet the rule. One sits a level short. One claims Evaluate with a multiple choice quiz behind it, over-claiming by two levels. Nothing in the course measured the three that carry no level at all.

The coverage join was sharper. Of ABET's seven student outcomes, four are claimed and three carry evidence at Analyse or above, with five of the eight outcomes on row (1). Rows (3), (5) and (7) are claimed by nothing, and the design project worth 25% of the grade assesses two of them every term. Read against the European framework's three dimensions the verdict changes again, which is why the course design pack starts from your own descriptors.

Eight outcomes, three at the rule, one over-claiming by two levels

What each verb can be placed at, what each assessment can actually evidence, and the three framework rows the course assesses without stating.

Outcome Register

Three of eight can be placed at the rule

Kenniston Institute of Technology, ME 341 Heat Transfer, 300 level. Kenniston's rule for 300 level: Analyse.

COVerbLvlEvidenceCeilRealRuleVerdict
CO-1understandnoneQuiz3none4Unplaceable
CO-2calculate3Homework334Short 1
CO-3analyse4Midterm444Met
CO-4be familiar withnoneQuiz3none4Unplaceable
CO-5design6Project664Met
CO-6attribute4Lab report444Met
CO-7knownoneFinal4none4Unplaceable
CO-8justify5Quiz334Over-claims 2

Two kinds of failure, two costs

RowKindRemedyCost
CO-1, CO-4, CO-7No level at allRewrite around an observable actWording
CO-2Under-claimsVerb up, evidence to the midtermWording
CO-8Over-claims by 2Evidence to the design projectAssessment

Softening CO-8's verb to match a quiz would close the gap on paper and make the course worse. That is why the two kinds are scored separately.

Assessment Evidence Ceiling

30% of the grade cannot reach the rule

A form of evidence caps what it can show whatever the question asks. Ceilings come from the task, not the topic.

IDAssessmentFormCeilGradeCarries rule
A-1Quizzes 1 to 4Selected response315%No
A-2Homework 1 to 8One-method problem315%No
A-3Midterm examConstructed response420%Yes
A-4Lab reports 1 to 3Own measured data415%Yes
A-5Exchanger design projectArtefact plus rationale625%Yes
A-6Final examConstructed response410%Yes
A-7End-of-course surveySelf-reportnone0%Never

What that leaves

ReadingShareConsequence
Ceiling below the rule30%Nothing mapped here can claim Analyse
Ceiling at or above the rule70%Where every outcome has to live
Able to reach 5 or 625%One assessment, so one home for Evaluate

A-7 gets a ceiling of none rather than a low number. A confidence rating is useful and evidences no outcome at any level.

Framework Coverage

Four rows claimed, three carrying evidence

ABET Criterion 3, outcomes (1) to (7). Five of the eight outcomes pile onto row (1).

RowWhat it asksClaimsEvidenceClausesVerdict
(1)Solve complex problems5CO-3Evidenced
(2)Apply engineering design1CO-51 of 8Half stated
(3)Communicate to audiences0noneUnclaimed
(4)Ethical responsibilities1none0 of 4No evidence
(5)Function on a team0noneUnclaimed
(6)Experiment and interpret1CO-6Evidenced
(7)Acquire new knowledge0noneUnclaimed

The reverse direction, and a second framework

FindingDetail
Assessed, never statedA-5 is a team project worth 25% of the grade and no outcome claims row (5)
Assessed, never statedA-5 requires a written client memo and no outcome claims row (3)
Same eight sentences, EQFKnowledge 5, Skills 2, Responsibility and autonomy 0 of 3 dimensions

Row (2) names eight factors a design must consider and CO-5 names one, cost. A tick matrix cannot tell one from eight.

What's in the pack

01

Outcome Register

One row per outcome with its verb level, the ceiling of the assessment behind it, the realised level, the gap against your rule, and whether the row under-claims or over-claims.

02

Assessment Evidence Ceiling

One row per assessment with the highest level that form of evidence can elicit and its share of the grade, so you can see what your outcomes are allowed to claim before writing them.

03

Framework Coverage

The join run both directions, with a column counting the consideration clauses inside each descriptor row that your wording actually names. One of eight still reads as a tick.

04

Outcome Statements

Ten worked outcomes at the level the framework asks, each carrying the row it answers to and the assessment that evidences it, with the reasoning behind every change from the inherited set.

05

How the Level Check Works

The three joins in order, why realised level is the lower of the verb and the ceiling, and why an unplaceable verb gets no level rather than a low one.

06

Observable Verb Bank

Verbs by level, each paired with the minimum evidence that can carry it, plus the unplaceable ones with what each usually meant so the rewrite is a decision rather than a guess.

07

Writing Guidance

Six sentence patterns that hold up, the conditions that move an outcome a level, and why a performance threshold belongs in the rubric rather than in the outcome.

08

Mapping Note

The document a reviewer opens three years from now: what was found, what changed, which rows are still open with an owner, and what would invalidate the mapping.

How to use it

  1. 1

    Open it in River, or download it

    Open the pack in River and let the agent run the three joins against your own framework and assessments, or download the eight blank Word and CSV files with no account.

  2. 2

    Send the descriptor rows, not a summary

    The framework text as published, because the consideration clauses inside a row are the thing an outcome half-claims. Add your institution's course-level rule if it has one.

  3. 3

    Send the assessment list with weights

    Every graded thing, its form and its share of the grade. This is the input people skip and the one that decides what any outcome can honestly claim, so the register cannot be filled without it.

  4. 4

    Read the register before rewriting anything

    Verb level, ceiling, realised level. Fix the under-claims in the wording, and send the over-claims to whoever owns the assessment, because softening a verb to match weak evidence is the wrong trade.

Frequently asked questions

Is this template free?

Yes. Download all eight files as Word documents and CSV sheets with no signup and no card. Edit with AI is the optional path, where the agent places your outcomes against your own descriptor rows and your own assessment ceilings. The rest sit in the template library.

What format are the downloaded files?

Word documents (.docx) for the five write-ups and CSV (.csv) for the three sheets, zipped together. They open in Word, Pages, Google Docs, Excel, Numbers and Sheets. In River the same content opens as live Docs and Sheets the agent can fill in.

Why not just use a Bloom verb list?

Use one, for the verbs. What a list cannot do is tell you the level your institution requires at this course number, or that the quiz behind your Evaluate outcome caps it at Apply. Those two joins are where the findings are, and both need documents a verb list never sees.

What is wrong with an outcome that says understand?

Nothing you can observe. No evidence separates a student who understands from one who does not, so the row gets no level rather than a low one. Three of the eight outcomes in the worked example were in this state, which means nothing in the course measured them.

Our framework is not ABET. Does the method still work?

Yes, and the framework is the input rather than an assumption. The same eight sentences read against the European framework cover knowledge and skills and say nothing about responsibility and autonomy. Send your own descriptor rows and the coverage join runs against those.

An outcome claims more than its assessment can show. Now what?

Say which of the two kinds it is, because they cost differently. Move the evidence to an assessment that can carry the level, or record it as an open curriculum item. Do not soften the verb: that closes the gap on paper and leaves the course weaker.

Where does the rest of the course work happen?

Outcomes sit upstream of nearly everything. The graded work and the alignment matrix come from the course design pack, the marking criteria from the rubric pack, and attainment against these statements feeds the outcome assessment pack. The programme-level view of how a whole degree collectively delivers these outcomes is the curriculum mapping pack.

Find out which of your outcomes nothing can evidence

Download the blank pack as Word and CSV files, or open it in River and have every outcome placed against your own descriptor rows and assessment ceilings.

Edit with AI