Results from LPA models applied to a subset of SNPs pruned at r2 = 0.1 were very close to the results from the full set of SNPs. Table 4 summarizes the results of 2- to 4-class models using the pruned subset of SNPs. Class profiles and proportions from 2- to 4-class models from pruned set of SNPs were close to corresponding classes from the full set of SNPs and were labeled accordingly in the Table 4. The results of k-means clustering applied to both the full and pruned subset of SNPs summarized in Tables 5 and 6. Although k-means clustering provides less guidance on the number of classes than in LPAs (Magidson and Vermunt 2002), we used the aligned box criterion (ABC) (Hall et al. 2014) to guide determining the number of clusters. ABC compares within-cluster sum of squares to those from simulated null distributions from varying number of clusters, and the largest difference indicates the optimal number of clusters. In our analysis, ABC peaked at 2 and 3 clusters from full and pruned (r2 = 0.1) sets of SNPs,
