This is a strange way of phrasing the result, since the participants were all from a group of university students their skill distribution in all those fields w

This is a strange way of phrasing the result, since the participants were all from a group of university students their skill distribution in all those fields was probably above average with respect to the overall population (especially since they were Cornell students). So I do not find it that surprising that they assessed themselves higher than warranted because in their experience they probably encountered many more people who had a lower skill level compared to them.