Skip to main content
. Author manuscript; available in PMC: 2019 Dec 20.
Published in final edited form as: Stat Med. 2018 Aug 21;37(29):4441–4456. doi: 10.1002/sim.7931

TABLE 2.

Weighted cross-tabulation of random class assignment and maternal age at the birth of the child using only male subjects, and weighted cross-tabulation of class assignment and of ever being convicted of a crime during ages 14 − 18 using only male subjects. The class assignments were obtained by using the estimated posterior probabilities to randomly assign each subject to one of the four latent classes. The random assignment procedure was repeated 1, 000 times, and the results were averaged. In the top table, we display in parentheses class proportions conditional on maternal age while in the bottom table we show conviction proportions conditional on class.

Class 1 Class 2 Class 3 Class 4
maternal age < 20 34.2 (0.178) 44.8 (0.233) 40.2 (0.209) 73.1 (0.380)
maternal age ≥ 20 968.7 (0.284) 941.7 (0.276) 581.7 (0.171) 917.6 (0.269)

 
Class 1 Class 2 Class 3 Class 4

ever convicted-yes 18.7 (0.031) 16.5 (0.025) 42.6 (0.105) 102.9 (0.144)
ever convicted-no 588.0 (0.969) 646.8 (0.975) 363.1 (0.895) 609.4 (0.856)