Anyone interested in this topic should read the primary research papers.
The pattern of differences in IQ test answers between generations is not congruent with the pattern of differences between individual who do better and worse on average within generations -- at least according to the few studies which have looked at this question. For this reason, we can infer that the generational changes in IQ test scores are not changes in the same underlying factor that causes differences in performance within a generation. If we call the factor that causes within-generation differences in IQ test scores intelligence, then what differs between generations isn't intelligence. In other words, the Flynn effect is something other than what it might appear at first glance. It's still something of a mystery.
Multi-group confirmatory factor analysis can be used to demonstrate this, or you can just look at the correlation matrices.
The pattern of differences in IQ test answers between generations is not congruent with the pattern of differences between individual who do better and worse on average within generations -- at least according to the few studies which have looked at this question. For this reason, we can infer that the generational changes in IQ test scores are not changes in the same underlying factor that causes differences in performance within a generation. If we call the factor that causes within-generation differences in IQ test scores intelligence, then what differs between generations isn't intelligence. In other words, the Flynn effect is something other than what it might appear at first glance. It's still something of a mystery.
Multi-group confirmatory factor analysis can be used to demonstrate this, or you can just look at the correlation matrices.