Delayed Gratification Test: A Marshmallow Test for Adults, Scored Across Four Areas

Twenty statements are rated on a five-point agreement scale from strongly disagree to strongly agree, scored 0 to 4. Five of them describe the pull of the instant option and are reverse scored, so a higher total always means more delay skill, never less. Each of the four areas can reach 20 points and the total runs from 0 to 80. Four bands divide that range: the fast reward usually wins at 0 to 27, delay is situational at 28 to 47, solid delay skill at 48 to 63, and strong delay skill at 64 to 80.

Those bands are educational divisions of our own scale, not clinical cutoffs. No published validation study stands behind them and they should not be compared with scores from research instruments. The four-way grouping is also ours: it follows the strategies the delay-of-gratification literature identified rather than a validated factor structure, so read the four bars as a map of where your score came from.

The reason the four bars matter more than the number is that they come apart in predictable ways. A high persistence bar with a low temptation-handling bar is the person who can grind through slow work but has nothing to do when a craving shows up. A high temptation-handling bar with a low planning bar is the person who handles the moment well and keeps getting ambushed because nothing was set up in advance. A low score on the instant option with everything else fine usually means one specific pull, often a phone or a card, rather than a general problem with waiting.

In the Stanford experiments run by Walter Mischel and colleagues from the late 1960s onward, a preschooler was left alone in a room with one treat and told that if they could wait until the researcher came back, they would get two. The measure was simply how many seconds or minutes the child waited before ringing the bell. It was a behavioral task, with a stopwatch, on children aged roughly three to five, and nothing about it can be reproduced with a questionnaire on an adult. Anyone selling you an online marshmallow test is selling you a self-report, and so are we; the difference is that we are saying it.

The part of that work that holds up best, and the part this test is built on, is the 1972 study of what the waiting children did. Mischel, Ebbesen and Zeiss found that making the reward more vivid shortened the wait, while distraction from it, either physically or in the child's own head, lengthened the wait considerably. Waiting was not a feat of endurance. It was attention management: look away, think about something else, make the thing less real in your mind. Later work added the strategies adults recognize, including committing in advance in ways that are hard to undo, and removing the cue rather than staring it down.

This test is for adults, and a score here says nothing about any child. If you are here because you are wondering about your own child's self-control, the research is worth reading and the task is not worth running at the kitchen table: delay time in a preschooler varies enormously with hunger, trust in the adult, how the instructions were phrased, and whether the child has learned that promised second treats actually arrive. There is no home version that means anything, and a child's few minutes of waiting is not a forecast of their life.

The marshmallow test became famous because of the follow-ups. Shoda, Mischel and Peake reported in 1990 that preschool delay time predicted adolescent academic and social outcomes, with correlations striking enough to make the task a cultural shorthand for self-control. The sample behind those headline numbers was small, drawn from a university nursery school population, and the subgroup analyses smaller still. That is not a scandal; it is what early exploratory work looks like. It is a reason to be careful about how far the finding was stretched.

In 2018 Watts, Duncan and Quan ran a conceptual replication on roughly 900 children from a national study, a much larger and more varied sample. They found the association between early delay and later achievement was about half the size previously reported, and that it shrank further, in most specifications to nothing of consequence, once family background, home environment and early cognitive ability were taken into account. Family income predicted both how long a child waited and how they did later, which means a large part of the original effect was picking up the circumstances the child was waiting in. A 2020 commentary by Falk, Kosse and Pinger argued that the two studies are less contradictory than the coverage suggested, and in 2024 Sperber, Vandell, Duncan and Watts followed the same cohort to age 26 and found that preschool delay time did not reliably predict adult functioning once the usual controls were applied.

The useful conclusion is not that self-control does not matter. It is that waiting ability is better read as a skill shaped by circumstances than as a fixed trait that decides your future. A child who has learned that promises are kept, who is not hungry, and whose household is stable finds waiting easier, and the same is true of adults. That is good news for anyone with a low score on this page, because skills and circumstances both change and destiny does not.

Start with your lowest bar, not your total. If temptation handling is lowest, you are relying on willpower in the moment, which is the one approach the research consistently finds weakest. Pick one specific pull and change the situation rather than your resolve: put the card somewhere inconvenient, delete the app from the phone and keep it on the laptop, leave the snack out of the house instead of in a cupboard, pay for the course up front so quitting costs money. These are not hacks; they are what the original research found the successful waiters doing.

If planning for your future self is lowest, the fix is usually making one decision once instead of many decisions daily. A standing transfer on payday, a recurring appointment, clothes and bag ready the night before, a weekly ten minutes of maintenance in the calendar. If the instant option bar is the low one and everything else is fine, treat it as one habit rather than a character problem, and look at what the pull is actually doing for you: most of the time it is interrupting boredom, tiredness or an uncomfortable feeling, and naming which one changes what you do next.

Delay of gratification sits next to other things worth looking at separately. Our self-control test Self-Control Test covers the broader trait, our procrastination test Procrastination Test looks at delay in the other direction, meaning putting off what you meant to do, and our grit test Grit Test covers persistence over years rather than minutes. If the pull toward intensity rather than reward is the recognizable part, our sensation seeking test Sensation Seeking Test fits better, and our time management test Time Management Test and motivation test Motivation Test: What Drives You cover the practical end.