It would have been noteworthy in the original design if more than one group of participants were, on average, within their quartiles on the guessing. I also fin

It would have been noteworthy in the original design if more than one group of participants were, on average, within their quartiles on the guessing. I also find it noteworthy that the average guess of the lowest quartile is lower than the average guess of the second lowest quartile, and on up the quartiles. On one hand this shows some awareness of relative ability along a massively smooshed logarithmic scale. On the other hand I wonder if this laddering follows as the averages are split into quintiles and deciles.