eduKate Singapore

VIEW THIS AS

Auto mode follows the Route Engine until you choose a viewpoint.

YOU ARE HERE

ROUTE CHECK

CONNECTED TO

WHAT NEXT

Use the canonical route for this room, or HELP if you are unsure.

Next useful routeCanonical hubHELP

How Intelligence Works | Endowment Effect — Ownership, Asking Prices and Better Trade Decisions

—

by

eduKate Asia

Did you know? People sometimes demand a higher price to give up an object they own than they would offer to acquire an equivalent object they do not own. Economists call this a willingness-to-accept versus willingness-to-pay gap, and ownership-related patterns are often described as the endowment effect. It is a useful lens for understanding valuation, negotiation and reluctance to trade, but it is not a universal law: the experimental evidence has important methodological disputes.

The helpful practice is to identify a fair comparison, separate the object’s practical value from the feeling of already owning it, measure both buying and selling decisions under clear rules, and consider actual alternatives and transaction costs. If you are deciding whether to keep an old subscription, accept a job transfer or sell a used item, ask what you would choose if the current allocation had never been made. Then check the answer against real constraints rather than assuming every attachment is irrational.

This guide explains the mechanism, original experiments and subsequent objections, works through a fully checked hypothetical trading market, diagnoses misleading interpretations and applies the reasoning to schools, workplaces and everyday decisions. Invented examples are explicitly labelled. The goal is better independent judgement, not a claim that everyone has the same preference or that ownership is always a mistake.

Contents

An object can be unchanged while its reference point changes

Imagine two identical mugs. One is on a shop shelf; the other has sat on your desk for a month. If you consider paying nine dollars for the shop mug but would not sell the desk mug for less than fifteen dollars, the physical object may be much the same while the choice is framed differently. Acquiring something can feel like a gain; surrendering something already possessed can feel like a loss. The ownership reference point changes the comparison.

That observation does not automatically prove a bias. The desk mug may have a memory attached, may be convenient to use, or may have replacement costs the shop mug does not. A rational owner might also value the time required to find a buyer, the risk of nonpayment or the inconvenience of replacing a familiar tool. Those are real differences that an honest analysis must include before attributing a gap to psychology.

An endowment-effect explanation becomes more compelling when ownership is randomly assigned, objects are essentially interchangeable, transaction rules are clear, and measured valuations shift with allocation. The method asks whether possession itself changes revealed or reported value after other explanations have been controlled. Experimental procedures matter because saying ‘what would you pay?’ is not always understood in the same way as making a consequential purchase.

Ownership can be temporary or symbolic. Someone may be given a default allocation, reserve a seat, choose a draft project title or receive a preliminary offer and soon begin treating that arrangement as the reference point. Yet a mere preference for one’s current option can also come from switching costs, information, status-quo convenience or a genuine match. The analyst must not treat all reluctance to move as the same psychological mechanism.

Researchers and readers should therefore distinguish an observed valuation gap from one particular theory of why it exists. The pattern, the loss-aversion interpretation and the practical decision rule are related but separate questions.

Willingness to pay and willingness to accept

Willingness to pay, usually abbreviated WTP, is the maximum a person would offer for a gain under a specified transaction. Willingness to accept, WTA, is the minimum payment they would require to give up a good or tolerate a loss under a specified transaction. A WTA greater than WTP for the same kind of object is sometimes called a valuation gap. In an ordinary market, different people can of course have different values without any inconsistency.

Even within one person, the two amounts need not match. If a person has limited cash, their ability to pay may cap WTP more sharply than their minimum selling price. If an object cannot be replaced, surrendering it is different from acquiring one more unit. Tax, search effort, shipping, repair responsibility and uncertainty can also make the sell and buy situations economically different. Simplifying them away without checking creates a false psychological diagnosis.

The 1990 Journal of Political Economy paper by Kahneman, Knetsch and Thaler reports market experiments with consumption goods such as mugs and contrasts them with induced-value tokens. Random endowments and a comparison with predicted trading volume are central to its logic. The paper is primary research into specific laboratory arrangements, not evidence that every product acquires a fixed markup when owned.

An induced-value token has a clearly assigned payoff in an experiment, which can help test whether the market mechanism itself is understood. A consumption good such as a mug allows personal tastes and reference points to enter. Comparing these settings can reveal whether unusual trading behaviour arises only from the structure of the experiment or is associated with the kind of object and ownership.

An article about ownership should always say which comparison is being made. Are we comparing one person’s maximum buy bid with the same person’s minimum sell ask? Different groups’ medians? Total trades after random allocation? A ratio can be calculated correctly and still answer the wrong scientific question.

The original experiments and their intended test

Kahneman, Knetsch and Thaler’s paper is titled Experimental Tests of the Endowment Effect and the Coase Theorem. The researchers used market procedures in which subjects could trade goods after an initial allocation. Under simplifying assumptions of well-defined preferences and sufficiently low trading frictions, the initial assignment of objects should not by itself suppress mutually beneficial exchange. The reported limited exchange for consumption goods therefore challenged a straightforward reference-independent model.

The experiments are historically influential because they linked a psychological account of ownership to observable market behaviour rather than relying only on people verbally agreeing that losses feel painful. Their evidence should still be read within the laboratory’s task rules, incentives, market experience, goods and sampled participants. A small set of experiments cannot specify how large a gap should be for every modern marketplace.

The broader theory often invokes loss aversion: giving something up is evaluated as a loss from an owned reference point, and losses can be weighted differently from otherwise comparable gains. That is a plausible mechanism and a central interpretation in behavioural economics. It should not be stated as a physical property of every human brain or as a guaranteed explanation of each observed market price.

The Coase-theorem connection also needs care. That theorem concerns idealised bargaining and efficient allocation under specific assumptions, not a claim that half of all real possessions must be traded. In experiments with random allocation and suitable underlying distributions, expected trade volume can be predicted. Outside those conditions, market thickness, bargaining cost, credit, information and preference heterogeneity strongly influence actual trades.

The right scientific question is whether observed non-trading exceeds what the explicitly stated model predicts. Merely seeing that someone keeps an object provides no benchmark. A careful explanation identifies the comparison, the assumptions and the alternative mechanisms before choosing a story.

Why experimental design remains contested

The effect has an important methodological counterpoint. In their 2005 American Economic Review study, Charles Plott and Kathryn Zeiler investigated whether misunderstandings of the valuation-elicitation procedure contributed to WTA–WTP gaps. With additional training and controls under their designs, they reported no such gap in their studied settings. Their conclusion questions whether all observed gaps can be confidently attributed to loss aversion.

That finding does not simply erase the earlier result, because the studies used different procedures and the literature includes replication and interpretation debates. Nor does the existence of earlier positive results prove that experimenter instructions are irrelevant. When two studies disagree, inspect treatments, participants, incentives, comprehension checks and measured outcomes instead of selecting a favourite citation.

People can misunderstand what a reported bid means. A participant who treats an instruction as a negotiation opening may give a strategic price rather than a truthful minimum or maximum. A participant who does not trust the random selection of an outcome may answer differently. A procedure that seems transparent to the experimenter may not feel transparent to the subject. Training and practice trials can change what valuations the measured response represents.

The practical consequence is methodological modesty. When an organisation claims an ownership premium, ask whether the buy and sell groups understood the choice mechanism, whether both groups faced equivalent transaction costs, and whether the outcome was an actual incentivised trade or a hypothetical questionnaire. A gap in a survey is evidence of a reported preference under that wording, not automatically a market transaction.

Good science does not require eliminating every alternative explanation before discussing a phenomenon, but it requires keeping those alternatives visible. The original paper and its critic belong together in a responsible reader’s evidence trail.

A numerical market: six owners, six potential buyers

Consider a completely fictional exercise with six randomly assigned owners of identical souvenir notebooks and six potential buyers. Each owner gives a minimum acceptable selling price in dollars: 14, 16, 12, 15, 17 and 13. The six buyers give maximum purchase prices: 10, 12, 14, 9, 15 and 11. These are invented numbers for arithmetic, not measurements from a real study.

Order the owner asks from smallest to largest: 12, 13, 14, 15, 16, 17. The median owner asking price is the average of the third and fourth values, (14+15)/2 = 14.5 dollars. Sort buyer bids: 9, 10, 11, 12, 14, 15. The median buyer willingness to pay is (11+12)/2 = 11.5 dollars. The median gap is 14.5−11.5 = 3 dollars.

Dividing the seller median by the buyer median gives 14.5/11.5≈1.26087, or a ratio of about 1.26. That does not mean an individual person necessarily values an object 26% more after receiving it. The statistic compares medians of separate fictional groups, which can differ for numerous reasons. It is not an estimate of a causal ownership effect without an appropriate design.

Can these people trade? If we artificially pair each listed buyer with the owner in the same position, only the third pair has a buyer whose bid, 14 dollars, meets or exceeds that owner’s ask, 12 dollars. That gives 1/6 possible within-pair trades, or about 16.67%. But a real market need not lock buyers and sellers into those pairings. The highest buyer may trade with the lowest seller, so the artificial pair rate is not the efficient market-clearing volume.

Now allow all buyers and sellers to meet. Compare the highest buyer bids 15 and 14 with the lowest seller asks 12 and 13. Two mutually acceptable pairings exist at prices within the relevant bid–ask intervals, assuming voluntary exchange, no additional costs and appropriate matching. A third-highest buyer bid of 12 falls below the next-lowest seller ask of 14, so no third such trade is available under this sorted matching.

The difference between one paired trade and two potential market trades demonstrates why the market mechanism matters. Median WTA, median WTP and actual exchange volume are distinct metrics. A study must define which is used, how participants can match and whether transactions happen rather than treating every comparison as interchangeable.

Another numerical check: an apparent premium and an alternative cost

Suppose a fictional person says an existing desk lamp is worth at least 40 dollars to sell, yet they would pay only 25 dollars to buy another. The raw WTA–WTP gap is 40−25 = 15 dollars. Expressed relative to the 25-dollar WTP, it is 15/25 = 60%. That number describes the stated asymmetry in this example, not the share caused by loss aversion.

Now discover that replacing the lamp requires a 12-dollar shipping fee and a three-dollar journey or setup cost, which the buy question did not include. The total replacement burden is 25 + 12 + 3 = 40 dollars. If the owner is unwilling to sell for less than replacement cost, the same 40-dollar asking price may be economically consistent without any ownership distortion.

Change the condition again: the lamp is already available for immediate local pickup with no fee, yet the stated gap persists under a clear incentivised protocol. Ownership-related preferences become a more plausible explanation, but other differences—sentimental history, repair risk, information and liquidity—may remain. The analyst still needs evidence to isolate a mechanism.

The important calculation is not merely the percentage premium. It is the correct comparison including relevant money, time, risk and replacement quality. Asking ‘what would you pay to regain this identical object after selling it?’ can sometimes reveal hidden assumptions, though it does not replace a properly controlled experiment.

Readers should resist false precision. A neat 60% gap can be perfectly calculated from numbers that are not comparable, while a modest observed trading difference may be important under a strong randomised design. Mathematical clarity and causal clarity are separate requirements.

Ownership is a reference point, but attachment has different causes

An object can be valuable because of use, identity or memory. A family photograph is not meaningfully interchangeable with a generic image, even if both have the same dimensions. An employee may prefer a familiar team because accumulated trust improves work. A student may keep handwritten notes because they contain a personalised trail of corrections. In such cases ownership can be associated with real benefits rather than an avoidable cognitive bias.

A reference-point mechanism instead predicts that assignment itself changes evaluation when comparable alternatives are otherwise held constant. The distinction can be tested by randomly giving one group an item while asking an equivalent group how much it would pay to obtain it. But even then the experiment must control instructions, expectations and whether subjects believe the exchange is real.

Endowment effects can overlap with other forms of decision friction. The status quo can require less effort. A person may worry that an unfamiliar replacement will disappoint. Someone may dislike having to justify a decision to others. These mechanisms can reinforce one another but do not have the same remedy. Reducing switching costs helps one; more accurate replacement information helps another; a reversible trial helps uncertainty.

The best practical question is not ‘am I biased because I want to keep this?’ It is ‘which benefits would remain if I obtained this item from scratch today, and which costs would I face if I changed?’ This reframing encourages examination without dismissing genuine attachments or personal preferences.

Treating emotional attachments as irrational by definition is itself a poor judgement. People are allowed to care about memories, continuity and convenience. The analysis becomes useful when it reveals assumptions the person did not intend to include or trade-offs they had not noticed.

Better selling and buying decisions

For a seller, begin with current alternatives. What would replacing the item cost in cash and time? Are there sentimental benefits that cannot be replaced? What transaction fees or fraud risks are realistic? What is the best available offer? Write down the minimum acceptable price after considering these elements, and separate it from an arbitrary high opening ask intended for negotiation.

For a buyer, ask whether the item delivers enough additional utility compared with keeping the cash or purchasing a substitute. A low bid may reflect limited liquidity rather than a weak preference. Keep that distinction visible if you compare seller and buyer prices. A marketplace that combines people with very different budgets will not produce a clean experiment on reference points.

Use neutral comparison questions. If the object were not already yours, would you acquire it today at its actual market price? If you owned cash of equivalent value instead, would you exchange it for the object? Does the answer change because of tax, effort or risk? These questions do not mandate trading; they expose which features determine the choice.

Take care with sunk costs. The amount originally paid for an object may influence feelings but is not necessarily the best guide to its present value. The separate escalation-of-commitment guide addresses continuing a project because resources have already been spent. Endowment effects concern valuation shaped by possession and reference points; the two can interact but are not identical.

If the decision is consequential or irreversible, seek independent valuations and compare real offers. A purely introspective exercise can help reveal a frame, but it cannot substitute for checking a legal agreement, a medical choice or a financial obligation.

Negotiation: gaps are information, not proof of bad faith

Negotiators often begin with different reference points. A seller starts from giving up a possession; a buyer starts from parting with cash to acquire it. Each may perceive the trade-off as a loss in a different dimension. Observing a bid–ask spread does not by itself establish manipulation, dishonesty or cognitive error. Normal market bargaining begins with differences in preferences and information.

A practical negotiator should ask what evidence supports the other party’s valuation and which transaction costs matter. A seller may require a high minimum because a replacement is scarce; a buyer may offer less because the item is easy to substitute. If both sides understand these facts, the efficient decision may be not to trade. ‘More trades’ is not automatically a moral or economic goal.

Where a negotiator suspects a framing problem, use comparison options rather than pressure. Show realistic substitutes, allow a cooling-off period, state which benefits are retained or lost, and avoid manipulating a person’s sense that a provisional offer is already owned. A transparent choice is more valuable than forcing agreement by exploiting psychological discomfort.

Organisations should also examine their own default allocations. Assigning equipment, benefits or project responsibilities can make later changes feel like confiscation. Explaining the policy, allowing appeals and demonstrating fair replacement options may reduce unnecessary conflict while preserving legitimate rights.

A negotiation outcome depends on authority and consent. Cognitive research is not permission to ignore somebody’s clear boundary because one suspects an endowment effect. The role of the explanation is to improve understanding and fair offers.

A classroom example: choosing between familiar and better practice

Consider a fictional learner who has used the same revision book for six months. A new set of practice materials more closely matches the learner’s current gaps, but the old book feels familiar and ‘already mine’. The student may be reluctant to switch because of ownership, but also because the old pages include useful annotations. These are different reasons and should be tested separately.

The tutor can compare one week of practice from each resource using equivalent question difficulty, independent corrections and later recall. If the old book remains better for a particular skill, keeping it is sensible. If the new resource more effectively addresses the verified gap, the child can retain helpful notes while adopting the stronger practice. The goal is learning, not winning an argument about attachment.

A numerical example makes the distinction clear. Suppose the learner answers 14 of 20 comparable independent questions correctly after using the old resource and 17 of 20 after using the new one, under equivalent conditions. The observed difference is 17/20−14/20 = 3/20 = 15 percentage points. This fictional short trial is not proof of a causal improvement; sample variation and question differences could matter.

A better decision would inspect which errors changed, whether the benefit persists on a delayed test, and whether the alternative costs more study time. The ownership feeling is one factor to be considered, not a reason to discard evidence. A student should not be told that affection for a familiar book is a weakness.

Teachers and parents can use the question ‘would we choose this resource today if neither book were already in the bag?’ as a discussion prompt. The useful result is a reasoned choice with an opportunity to revisit it, not a forced switch.

Organisational ownership and change programmes

Teams can grow attached to a dashboard, process or software platform partly because they built it. Their expertise, embedded integrations and social responsibility can create genuine value. But the effort already invested may also cause them to discount a safer or simpler alternative. A thoughtful change review separates practical switching costs from the symbolic value of ‘our system’.

Before replacing a process, define a common outcome standard. Compare reliability, cost, accessibility, security, auditability and how new employees learn the tool. Include migration risk and knowledge loss. An outside option should not be praised solely because it is new, and a familiar system should not be protected solely because the team designed it.

Consider a fictional process that costs 4,000 dollars annually to maintain and would cost 1,500 dollars annually after migration, but requires a one-time 5,000-dollar transition and 800-dollar training investment. Ignoring other changes, annual savings are 2,500 dollars and initial added costs are 5,800 dollars, producing a simple payback time of 5,800/2,500=2.32 years. Whether switching is worthwhile depends on expected life, discounting, service quality and risk, not feelings alone.

The team may still correctly reject migration if it requires unsafe downtime, loses essential functionality or has legal retention issues. Endowment-effect reasoning is a prompt to review the comparison, not a substitute for engineering and financial analysis. Document the assumptions and independently challenge both options.

This connects to ambiguity aversion, where people respond differently to known and unknown probabilities. Familiarity and possession can coincide, but the better repair depends on which uncertainty is doing the work.

Defaults, trials and reversible choices

Default settings are powerful partly because they spare effort. A person enrolled automatically in a useful programme may benefit from the default, and abandoning it could be harmful. In another setting, default ownership may make an unnecessary subscription feel like something that would be lost if cancelled. A behavioural explanation should not assume all defaults are manipulative or all opt-outs desirable.

Design choices to support informed judgement include plain comparisons, clear cancellation terms, reminders about recurring charges and non-coercive opportunities to review alternatives. A reversible trial can teach a person about a replacement before asking them to surrender the incumbent option. That reduces uncertainty while preserving freedom of choice.

An experiment comparing defaults should separate the influence of ownership from the influence of information and friction. If one condition requires ten clicks to switch and another requires one, differences in switching do not isolate an ownership mechanism. Equalise relevant effort and explain the transaction plainly before drawing a psychological conclusion.

People also change preferences over time. A mug or tool may become more useful after weeks of use; a subscription may become redundant when circumstances change. A decision rule that once made sense can legitimately change without any previous reasoning error. Review outcomes prospectively instead of using hindsight to blame every past choice.

The standard of good choice architecture is clarity, consent and an opportunity to reconsider—not engineering a transfer merely because a particular study observed trading inertia.

Errors in diagnosing the effect

Error one is declaring every high selling price evidence of irrational loss aversion. The repair is to check replacement cost, scarcity, sentiment and transaction risk. Error two is comparing medians from unequal groups as if they were matched individual changes. The repair is to name the experimental design and comparison unit.

Error three is using hypothetical survey answers as if they were completed trades. Actual market behaviour involves incentives, budgets and constraints. Error four is forgetting that a fixed pairing of buyers and sellers does not describe a competitive market. The remedy is to state matching and pricing rules before predicting trade volume.

Error five is assuming that a published laboratory pattern settles every practical context. Read the original research alongside methodological challenges. Error six is treating any deviation from a simple economic model as a personal failing. Models describe assumptions; people may have legitimate preferences outside a model’s scope.

Error seven is using an ownership explanation to pressure a vulnerable person into giving up property. Consent, rights and safety are not overridden by a psychological hypothesis. Error eight is ignoring the possibility that a person wants to keep an item because the alternatives are worse for them. The correct comparison is theirs, not the observer’s favourite option.

Error nine is reporting an exact ratio without a denominator or uncertainty. A 26% gap compared with buyer WTP is not the same as a 21% gap compared with seller WTA. Both use the same difference but answer different relative questions. Error ten is inventing causality from one unusual anecdote. Record the evidence, test alternatives and remain open to correction.

Transfer exercise: an inherited workspace and a new tool

A group inherits a shared online workspace and becomes attached to its folders and custom views. Another platform offers superior accessibility but requires a migration. The group says the current tool is ‘worth keeping at any price’ because it is theirs. Your job is not to classify them as biased; it is to construct an honest comparison and decide which part of attachment actually matters.

First calculate current costs and a transition case. Suppose maintaining the present workflow takes 12 staff-hours each month, while a tested alternative would take 7 hours after a two-month migration requiring 30 staff-hours. Monthly savings are 12−7=5 staff-hours. Ignoring new risks and assuming those estimates hold, the 30-hour migration cost has a simple labour-time payback of 30/5=6 months after the new workflow is operational.

Second compare capabilities: data ownership, accessibility, necessary integrations, collaboration and ability to reverse the change. Third ask what current benefits would remain if the team chose this tool from scratch today. Fourth permit a narrow pilot using representative work. These steps separate real switching costs, uncertainty and symbolic ownership.

Now change the context: a legally required archive cannot be transferred to the alternative. The migration must preserve the archive in an approved form, so the original 30-hour estimate is incomplete. The ownership attachment might coexist with a genuinely binding reason not to migrate. The correct answer is to change the project scope and cost model, not declare the decision irrational.

The transferable method is to identify the reference point, include genuine ownership benefits, compare equivalent alternatives and record the evidence. A bias label adds little if it prevents analysis of the actual task.

Limitations and unresolved research questions

The endowment effect is influential, but observed valuation gaps vary by good, design, experience and market procedure. Some studies report gaps under controlled allocation; others have challenged the interpretation or found no gap with different elicitation training. The safe scientific summary is a documented family of findings with meaningful boundary conditions, not a fixed human constant.

Loss aversion is one proposed explanatory mechanism, not a unique signature. Misunderstood instructions, strategic bidding, market power, income effects, sentimental value, non-fungibility and switching costs can all change buying or selling responses. Different experiments isolate different pieces of this puzzle.

Data from a particular consumer market cannot automatically predict educational choices, health decisions or public policy. Those domains contain obligations, unequal power, uncertain outcomes and rights that a mug exchange experiment does not replicate. Use ownership framing as one possible diagnostic, not as the principal explanation for every refusal to change.

Ratios from tiny hypothetical examples are descriptive and should not be presented as population effect sizes. The calculations in this guide are checked teaching exercises rather than reported results of external trials or live market observations.

Finally, rational choice does not require emotional indifference. People can value continuity and memory. Intelligence here means recognising the reasons behind a choice, examining the evidence and acting with appropriate respect for people and constraints.

Frequently asked questions

What is the endowment effect? It refers to ownership-related valuation patterns in which giving up a possessed good is sometimes valued differently from acquiring an equivalent good, often studied through WTA versus WTP.

Is a higher selling price automatically irrational? No. Replacement cost, transaction friction, memories, uncertainty and personal preferences can justify different buy and sell valuations.

What is willingness to accept? It is the minimum compensation someone requires to give up a specified good or tolerate a specified loss under stated transaction rules.

What is willingness to pay? It is the maximum a person would pay to obtain a specified good or benefit under stated budget and transaction conditions.

Did the original experiments prove everyone has this bias? No. They reported particular findings under laboratory procedures. Later research has disputed how robustly those observations demonstrate loss aversion rather than procedural misunderstanding.

Can an endowment effect cause people to keep a poor subscription? It may contribute, but cancellation costs, forgotten payments, uncertainty and genuine usefulness can also matter. Check the contract and real alternatives.

Does ownership attachment always reduce trade? No. Markets, goods and people differ, and some rational owners should not trade. A valid diagnosis requires an explicit model and comparison.

How can students learn the idea? Compare what they would pay for a new item with what it would cost to give up a similar owned one, then discuss replacement, sentiment and alternative explanations without assuming the gap proves bias.

How does this differ from sunk-cost reasoning? Sunk-cost errors involve past irreversible expenditure shaping future choices. The endowment effect concerns valuation relative to ownership or a reference point; they may overlap but are distinct.

Primary research and useful next reading

If you are a student, reproduce the six-person market arithmetic and explain why medians are not paired causal estimates. Then change a buyer’s budget or a seller’s replacement cost and observe how potential exchanges change. The lesson is to name assumptions, not memorise one famous mug experiment.

If you manage a team, compare inherited and alternative processes on a shared set of outcome measures, account for migration burden, and ask colleagues which aspects of familiarity provide real value. Invite independent challenge without dismissing attachments or treating an experiment as permission to override preferences.

For the research debate, read Kahneman, Knetsch and Thaler, Experimental Tests of the Endowment Effect and the Coase Theorem (1990) and Plott and Zeiler, The Willingness to Pay–Willingness to Accept Gap, the Endowment Effect, Subject Misconceptions, and Experimental Procedures (2005). Their contrasting conclusions make evidence boundaries especially important.

Related reader jobs include escalation of commitment and ambiguity aversion. For broader learning and group decisions visit How Intelligence Works and education that transfers into life.

This article makes no claim that a real experiment, marketplace or school trial was run here. It is a careful interpretation of cited original research plus labelled numerical teaching cases that readers can independently check.

Discover more from eduKate Singapore

Subscribe now to keep reading and get access to the full archive.

Continue reading