eduKateSG · Justice · The Gold Standard of Education
Evaluate real opportunities, not resources alone
Choose the explanation you need, or follow all twelve chapters in order.
Full chapter index · Independent practice · Justice subject guide
The capability approach asks what people are genuinely able to be and do. Functionings are achieved states and activities—being nourished, travelling safely, learning, participating. Capabilities are the real opportunities to achieve combinations of functionings. Resources matter, but people convert the same resource into opportunities differently.
This guide makes that conversion visible without pretending capability sets are easy to observe. It uses a practical mobility example, separates freedom from achieved outcome, and shows how personal, social and environmental conditions alter justice claims.
Continue from The Gold Standard of Justice within The Gold Standard of Education: learning that transfers into life and the How X Works learning library.
Full chapter index
See the mechanism · Chapters 1–2
Build the worked model · Chapters 3–5
Test what changes · Chapters 6–8
Check evidence and limits · Chapters 9–10
Practise and continue · Chapters 11–12
CHAPTER 1 OF 12 · See the mechanism
1. Separate resources, capabilities and functionings
A bicycle is a resource. Being able to reach school safely and on time is a capability. Actually cycling to school is a functioning. The three can diverge: a person may own a bicycle but lack a safe route, or may have the capability to cycle and freely choose to walk.
This distinction matters because equal resources do not ensure equal substantive freedom, while equal functioning can conceal coercion. State which evaluative space a policy uses. Income, primary goods, satisfaction, achieved outcomes and capabilities answer different questions.
The distinction also changes evaluation timing. Resource delivery can be checked immediately, a capability may require testing real access across several situations, and a functioning may occur later. A programme should not declare justice achieved at the procurement stage when the promised opportunity still depends on inaccessible routes, information or institutional permission.
Evidence check: for Separate resources, capabilities and functionings, preserve the observation before the interpretation. Record resources, capabilities and functionings separately. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
A capability set contains feasible combinations of functionings, not one isolated option. Safe travel, paid work and caregiving may compete for the same hours, so opportunities interact. A transport policy that enables one trip but destroys another option may not expand the set overall.
Capability is freedom with practical substance. A nominal legal permission is insufficient when cost, inaccessible design, danger or discrimination makes the option unusable. Map the actual prerequisites and constraints rather than counting services on paper.
Evidence check: for Think in capability sets, preserve the observation before the interpretation. Record the feasible set rather than one nominal option. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
Personal conversion factors include health, age, metabolism, skills and disability. Social factors include norms, discrimination, care expectations, public policy and power. Environmental factors include geography, climate, transport design and pollution. These categories organise inquiry; they are not rigid boxes.
Conversion factors can interact. A ramp helps only if routes connect, information is usable and social treatment is safe. Collect evidence at the point where a resource becomes a feasible action, not only at the budget or delivery point.
Evidence check: for Identify conversion factors, preserve the observation before the interpretation. Record personal, social and environmental conversion. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
Suppose three students each receive a $100 transport credit. Student A has frequent accessible buses and turns most of the credit into reliable school and library trips. Student B uses a wheelchair, but several stops lack step-free access. Student C lives farther away and has daily caregiving duties that restrict departure times. Equal credits produce unequal opportunity sets.
A capability-focused evaluation records affordable reachable destinations, safe time windows, journey reliability and assistance requirements. It does not assume B or C must use a particular mode. The policy question becomes which additional changes—routes, scheduling, accessibility or care support—make valued options genuinely feasible.
| Student | Resource | Conversion condition | Capability consequence |
|---|---|---|---|
| A | $100 credit | Frequent accessible route | Reliable school and library trips |
| B | $100 credit | Stops lack step-free access | Many nominal trips infeasible |
| C | $100 credit | Distance and caregiving times | Usable departures sharply limited |
Evidence check: for Work an equal-resource mobility example, preserve the observation before the interpretation. Record every student’s complete-journey constraint. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
CHAPTER 5 OF 12 · Build the worked model
5. Distinguish freedom from achieved outcome
Two people may both remain at home. One chooses quiet study from several safe travel options; the other has no accessible transport. The functioning looks alike, but their capabilities differ. Conversely, a compulsory programme may produce a valued functioning while reducing agency.
Measure both opportunity and achievement when possible. Ask whether non-use reflects choice, adaptation to exclusion, risk, information or a constraint. Do not infer a capability directly from one observed behaviour.
Evidence check: for Distinguish freedom from achieved outcome, preserve the observation before the interpretation. Record choice, non-use and achieved outcome. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
People living with persistent deprivation may lower expectations or report satisfaction despite restricted options. This is one reason subjective satisfaction alone can miss injustice. Yet outside evaluators can also misread values and impose their own priorities.
Use respectful deliberation, disaggregated evidence and opportunities to revise the account. Treat testimony as essential evidence without making it the only measure. The response to adaptation is better inquiry and public reasoning, not dismissal of people’s stated views.
Evidence check: for Address adaptive preferences carefully, preserve the observation before the interpretation. Record adaptation evidence without dismissing testimony. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
Well-being concerns a person’s own life and states; agency includes goals a person has reason to pursue, including commitments to others. A caregiver may knowingly sacrifice one personal functioning to support a family member. That choice should not be reduced automatically to a welfare deficit.
A just evaluation can report both: the person’s real alternatives and the goals they choose. Policy should avoid relying on unpaid sacrifice as if it were costless while still respecting agency. This dual view is especially important in education, health and care.
Evidence check: for Relate well-being and agency, preserve the observation before the interpretation. Record well-being, agency and unpaid sacrifice. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
Sen’s capability approach is intentionally open about a single fixed list, emphasising reasoned selection in context. Nussbaum develops a more specified account of central capabilities for political purposes. They are related but not interchangeable.
Declare who selected dimensions, whose voices were absent, and what purpose the list serves. A school-access study may focus on learning, bodily safety, affiliation and practical mobility; a national index may need a broader set. Selection is a normative step, not a neutral preprocessing task.
Test the list against concrete cases. If a dimension cannot distinguish a student with several safe routes from one with only a dangerous route, it may be too abstract for the decision. If it treats one culturally preferred way of living as compulsory, it may be too prescriptive. Revise through public reasons that others can question, not merely expert convenience.
Thresholds need the same openness. A minimum journey time, safety level or service reliability can support policy, but the number should connect to what people must be able to do. Examine those just above and below it, because a hard administrative boundary can create new unfairness even when the underlying capability changes smoothly.
Evidence check: for Select dimensions through public reasoning, preserve the observation before the interpretation. Record dimension selection and whose voice counts. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
CHAPTER 9 OF 12 · Check evidence and limits
9. Measure without pretending capabilities are directly visible
Functionings and resources are often easier to observe than the full opportunity set. Use multiple indicators: availability, affordability, accessibility, legal entitlement, travel time, reliability, perceived safety and actual take-up. Qualitative evidence can reveal options and constraints that administrative data misses.
A practical design can combine an opportunity dashboard with short journey or decision narratives. The dashboard supports consistent comparison; narratives reveal why the same score may arise from different constraints. Where direct observation is impossible, label the proxy and test it with affected people. For example, distance to a bus stop is not a valid accessibility proxy if crossings, gradients or harassment make the route unusable.
Time is part of capability. An option available only after a deadline, during unsafe hours or with unpredictable waiting may not be a real option for the decision at hand. State the relevant window and the reliability threshold rather than counting theoretical availability.
Avoid collapsing dimensions into one score unless the weighting and compensation rules are justified. Report uncertainty and subgroup variation. An average expansion can coexist with a severe loss for a small group.
| Layer | Possible indicator | Question answered | Limit |
|---|---|---|---|
| Resource | Credit or service count | What was supplied? | Not real usability |
| Conversion | Access, time, safety | Can it become an option? | Context-specific |
| Capability | Feasible valued alternatives | What could the person choose? | Often partly latent |
| Functioning | Observed journey or attendance | What occurred? | May conceal constraint or coercion |
Evidence check: for Measure without pretending capabilities are directly visible, preserve the observation before the interpretation. Record observable indicators and latent opportunities. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
CHAPTER 10 OF 12 · Check evidence and limits
10. State limits and common misconceptions
The capability approach does not say resources are irrelevant; it explains why resources are means rather than the complete end. It does not require forcing people into valued functionings. It also does not automatically resolve conflicts between capabilities, define thresholds or choose institutions.
Capability measurement can be data-intensive, and proposed lists can become paternalistic. Some opportunities are hard to distinguish from preferences, and responsibility questions remain contested. State these limits rather than turning a flexible evaluative framework into a mechanical scorecard.
Aggregation adds another limit. A single index can hide non-compensable losses when strength in one dimension offsets absence in another. Present dimension-level results and examine the worst constrained groups before any total. If weights are used, show who chose them and whether conclusions change across defensible alternatives.
Evidence check: for State limits and common misconceptions, preserve the observation before the interpretation. Record paternalism, data and aggregation limits. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
Change the mobility example: the city makes every bus step-free, but Student B still cannot reach the stop because pavements are broken, while Student C’s last return service leaves before caregiving ends. Vehicle accessibility improved a resource characteristic without equalising the capability.
Redesign the evaluation around complete journeys and time windows. Add pavement repair, demand-responsive links or later service, then ask which combinations students can genuinely choose. The changed context demonstrates that conversion factors operate across a system, not at one service point.
Evidence check: for Practise a changed-context policy, preserve the observation before the interpretation. Record the accessible-bus but broken-pavement context. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.
For a practical audit, write the valued functioning, relevant resource, personal/social/environmental conversion factors, feasible alternatives, observed achievement, voice and choice evidence, subgroup distribution, unresolved conflict and next policy test. Frequently asked questions: Is capability the same as happiness? No. Is a functioning always better when achieved? Not if coerced or wrongly specified. Can income proxy every opportunity? No. Must every study use the same list? No, but selection needs reasons.
Continue with distributive justice, equality, priority and sufficiency, which asks how advantages are distributed once the evaluative currency is declared. For foundations, read the Stanford Encyclopedia of Philosophy entry on the capability approach and the United Nations Development Programme discussion of capabilities and human development. Use them to frame concepts and disagreements, not as a substitute for affected people’s reasoning.
Evidence check: for Build the evidence route and continue, preserve the observation before the interpretation. Record the narrowest capability claim supported. Then state the inference, one credible alternative, and the additional observation that would distinguish them. This compact audit keeps a fluent explanation tied to evidence that another reader can inspect.
Changed-context transfer: repeat this chapter’s decision after changing one consequential condition while holding the others fixed. Calculate or predict what should change, name the threshold or mechanism, and compare with the original case. If the conclusion survives, state the range tested; if it reverses, explain why. Transfer is demonstrated by the changed case, not asserted from the first example.
Practical decision record: write one sentence for what is known, one for what remains uncertain, and one for the next safe action. Include the evidence owner, the latest useful decision time, and the condition that would stop or reverse the action. This turns a technical result into a reviewable choice without pretending uncertainty has disappeared.