A= Your IQ is meant to be 79
B= Your IQ is meant to be less than 79
C= Your IQ is intended to be 81
D= Your IQ is meant to be less than 79
E= Your IQ is meant to be less than 79
F= Your IQ is meant to be less than 79
G= Your IQ is meant to be less than 79
H= Your IQ is meant to be less than 79
So what we can say is you can definitely score 79 but below that it’s a vague “Your IQ is meant to be less than 79”. I would say though if properly calibrated it would be near to A ie only a few points away from it.
Tried this with Mensa Norway choosing all the first choice answers - " Your IQ lies outside the area that the test is able to measure " so Mensa Norway gives no indication of a minimal score.
I contrast this to tests where I have scored 65 or under . Obviously those had a lower minimal IQ to Mensa Denmark. This raises the question that if scores become increasingly inaccurate above a certain level is the same true at the lower end of the IQ scale ?
Yes, because of the normal distribution of IQ scores and the fact that most IQ tests are not designed to measure extremely high or extremely low scores (i.e. they are designed to capture almost all, but not all of the variation). The maximum and minimum measured total IQ scores are about equally far from 100, and the confidence intervals increase almost identically as scores increase or decrease from 100 (reflecting the almost-perfect normal distribution of the population scores and the focus on capturing as much variation as feasible). However, good IQ tests are easily able to measure scores lower than 70 with good accuracy. They are also supposed to correct for response styles like random guessing or consistently choosing one option.
What I would offer up is that if you are very much guessing, as I do with 80-90%+ of non-verbal type questions , then your intelligence level for that type of question is going to be much lower than if you are able to try and reason out an answer even if you may sometimes be wrong in your reasoning .
Yes, but since guessing is in effect the same as not knowing, a good test would try to account for it by removing the positive effect of guessing on the total score, so that the total score would be approximately equal to not having answered the questions you guessed on.
The first method with the short formula would be the preferred one for IQ tests (and I’m pretty sure they use something just like that one in the scoring programs). The second one, where answering wrong is strongly penalized, is more likely to give inaccurate scores since it could affect scores very differently depending on whether you guess or not, and on whether you understood the instructions well or not. IQ tests always contain questions that very few people are supposed to be able to answer (to avoid ceiling effects), and the test administrator is always supposed to encourage the test-taker to answer, in order to minimize the effect of different levels of motivation on the final test results (e.g. if someone says “I’m not sure…”, the test administrator should encourage the test-taker to try anyway, and then stop that part of the test after a criterion regarding the number of incorrect answers is reached). This does not go well with heavily penalized incorrect answers.
IQ tests can kiss my keyboard
I took an online romanian test my score was very very low and i answered all questions to get the result. I took only first 25 or so questions on this link/test (DK mensa) and i scored normal so i believe what Treebeard said about penalized answers/wrong answers…
