stupid test questions and answers — i took 3 of them

online tests will ask you which fruit you are and conclude things about your iq from the answer. i took one. i picked pomegranate because the seeds felt accurate to my mental state. the test said i was stupid. i said the test was lazy. we disagreed and went on with our lives. notes follow.

workstation, thursday, 11:34am. the spreadsheet i’m meant to be reviewing has been minimized for nineteen minutes. nobody upstairs has noticed yet. the office printer two desks over is making a sound i’m choosing to interpret as approval.

so. stupid test questions and answers. a phrase that arrives in my search bar with the regularity of a notification i did not consent to. people google it. i googled it. then i took three of these tests in sequence, with the focus i normally reserve for tax forms i pretend to fill out. stefan, my colleague who treats every test like a wine pairing, would have approved of the methodology if not the conclusions. the broader case for the word lives in my pillar on stupid; this post is narrower. what those tests measure, what they pretend to measure, and what the pretending hides.

stupid test questions and answers are multiple-choice prompts that pretend to grade your intelligence using puzzles a mid-attentive child could solve. the questions measure pattern recognition under no time pressure. the answers measure your willingness to play along. the result measures, on balance, nothing. a colored badge arrives anyway, with a sentence about your potential.

A QUIZ. IS NOT. A VERDICT.

that goes on the record before the post argues with itself. the badge at the end of an online quiz has the legal authority of a fortune cookie and roughly the same ingredient list. read it. file it. forget it.

stupid test questions and answers, what they really measure

the marketing claim is intelligence. the actual claim, on inspection, is patience. fifteen minutes of clicking radio buttons next to options that all sound, deliberately, like the right one. the test is not measuring whether you can think. it is measuring whether you will sit still long enough to pretend you are thinking.

i ran three on a single morning. one asked me which animal i would be in a forest. one asked me which shape did not belong in a row of four. one asked me to choose between two paintings i had never seen. all three concluded, with a small celebratory animation, that i was below average. the average, here, is whatever number makes the next ad land cleanly. a marketing variable. the test is a funnel.

and the funnel knows what it is doing. you fill out a quiz that confirms what the quiz already decided, then click through to a course promising to fix it. forty-nine dollars. the fix does not exist. the funnel does.

the questions i refused to answer

three questions in particular broke the contract.

1. “how often do you feel useless?” — not a question on a logic quiz. a wellness pamphlet wearing a logic quiz’s coat. four options: never, sometimes, often, always. there was no “depends on whether the microwave is currently working”. the granularity was missing. i refused.
2. “if you had to choose between being smart and being kind, which would you pick?” — pretends to be philosophy. is in fact a personality reveal trap. either answer makes you look bad. there is no third radio button labelled “this question is the problem”. i refused.
3. “do you often feel that other people misunderstand you?” — on a grid pretending to be cognitive. it is not cognitive. it is a horoscope with a grading scale. anyone alive answers yes. i refused.

three refusals in fifteen minutes. the test still gave me a score, lower than it would have been had i answered. the test punishes silence. a feature, not an oversight.

the answers i wrote in the margins

since the radio buttons would not accept my actual answers, i typed them into a notes app.

question: which fruit best represents your inner state? answer: a pomegranate, because the seeds are organized but each one is its own small confused project, and also because i once dropped one in 2021 and have, in a small way, never recovered.

question: which of the following best describes your decision-making style? answer: the one not listed. mine is closer to a coin flip held over a hot stove for too long. the coin lands. the stove also has feelings. neither option acknowledges that decisions are made in environments, not vacuums.

question: what is your biggest weakness? answer: i killed seven microwaves. not a weakness in the way the test means. on closer inspection, a research budget i did not have permission to spend. the seventh is currently making a sound a microwave should not make.

the third yoga mat. still under the couch from 2023, this morning. the test did not ask. any quiz that does not, at some point, ask “how many yoga mats live in your apartment unused” is not measuring you. it is measuring the version of you the quiz can fit in a graph.

what the test got right by accident

here is the part i did not expect. the cheapest-looking test, the one with the worst typography and the most obvious affiliate links, accidentally asked a useful question.

the question was: “how often do you change your opinion when given new information?” by the working definition of stupid as a visible repeated bad call by someone informed, that is exactly the right question. opinion-flexibility under new information is, in fact, one of the two or three things i would put on a real cognitive scorecard if i were running a more honest funnel.

the test, however, weighted it the same as “which fruit are you”. that is the giveaway. the right question buried in the wrong test. pair it with a real scoring system and you would have a quiz worth taking. and, possibly, a small working theory of how people separate honest mistakes from quiet lies they tell themselves — the liar inquiry runs adjacent, and the overlap between an unflexible opinion and a quiet personal lie is not a coincidence. you do not have the right scoring system, though. you have a funnel. it does not want you fixed. it wants you returning.

why the format itself is the issue

multiple-choice with four options is not a format for measuring intelligence. it is a format for sorting people into a database. four options exist because four fits comfortably on a phone screen above an ad. that is the engineering constraint. the entire honest answer. there is no cognitive science behind the four. there is a screen size and a pre-roll.

by that constraint, the questions take a particular shape. they cannot be questions where the answer is “it depends” or “i would need to be in the room”. they have to be ranked, scorable, returnable.

i hold a take on adjacent matters and i’m filing it here because the post is mine. ice cream is breakfast. it contains milk. i hold it with the same confidence the test holds its score, and with strictly more justification. milk is, by every accepted breakfast standard, a breakfast input. ice cream is mostly milk. the temperature is decorative. if a four-option quiz asked me to defend it, i would have nowhere to put the argument. the format problem in miniature.

the only honest movie about an iq test is the 2006 film “idiocracy”, which sets the bar at “did the protagonist water the crops with sports drink” and grades civilization on the answer. that is, in fact, the right level of question. it is also a comedy. those two facts are related.

the work of figuring out what stupid actually does in a sentence happens in places like a closer reading of the stupid is as stupid does line, where the meaning has to be argued instead of clicked through. a different shape of work. the shape this format cannot hold.

verdict, the test failed me, not the other way around

the seventh microwave just made the sound again. three weeks from retirement, conservatively. i will not be quizzed on it. that is, this morning, the only score that counts.

↑