It's always about the lens on these studies. So, I look at the data and am most interested in the bimodal verbal/math aspect. They toss it as an aside and don't even run it as a controlling variable. "However, early school entry was also related to mediocre math performance in 1922"
Anecdotally, we often hear of the mathematically gifted kid not quite being "ready for school" as early as the more verbally oriented. Looking at part of table 3:
Outcome Variable Early Entry (N = 394) On-Time Entry (N = 540) Late Entry (N = 89)
...
1922 verbal performance 3.35 (1.44) 3.30 (1.36) 3.09 (1.28)
1922 math performance 2.54 (1.06) 2.63 (1.01) 2.84 (1.01)
...
What a pretty crossover.
