I've come to the conclusion that group ability tests almost always underestimate IQ scores.
I'd be surprised if that were true, since the companies that make group ability tests have psychometricians who know how to calibrate the scores on the group ability tests so that they are comparable to individuallly administered IQ tests.
Bostonian, I would disagree on two grounds.
Group IQ tests are primarily devised for a different purpose. They are to screen numbers of individuals for ability to enter into a program. I believe the armed services were among the major users of such instruments.
An individual IQ test is a discrete measure and observance of one person's ability on one occasion. It's use is to provide information to guide that individual in a number of areas, including academic placement but also for neurological problems for example. The results should come with a comprehensive report noting, not just scores, but how the participant responded and what observations the tester made that can explain those responses if they are unusual. That information is as valuable as the numbers in guiding future educational decisions or if looking for discrepancies suggestive of either exceptionality which is not easily determined by the numbers.
The setting of administration can greatly influence the outcome of group tests in a negative way. The instructions may not be heard or clear to the participants. That should be picked up in an individual setting and prompt the administrator to ensure further instructions are clear and also have any impact noted in the report.
The noise, or even the simple presence, of other participants can distract an individual, especially if they are 2e, and that will not be accounted for in a group test.
Fine motor performance, difficulty managing a pencil to fill in the answer will not be detected by group administrators. Visual disturbances leading to the wrong answer bubble being selected, missing a bubble and skewing all further answers because they are out by one, fatiguability and therefore failing to finish.....these are all traps in group administered tests that can impact on the final score and mean it is not an accurate picture.
Quick thought experiment: What if the tests have the same average?
