Births, from a voice that knows more than age

Improved A later registered study met the bar: Births with the level set once, on a survey nobody had opened.

A voice trained on women's own records predicted who would have a baby in the next three years far better than age alone (AUC 0.728 against 0.620) and better than a regression on the same facts (0.701), on a national survey nobody here had opened. Its predicted share came in at 0.169 against the observed 0.178, but the registered adjustment for the national birth rate pulled it to 0.155, below the observed range, because the voice had already adjusted for smaller families itself. Registered before the data was read; scored once.