How to measure parenting styles?

Abstract In this paper, we measure parenting styles through unsupervised machine learning in a panel following children from age 5 to 29 months. The topic model, which is statistical model originally developed discover the latent semantic structures text, classifies parents into two styles: “warm” and “cold”. Parents of warm type tend respond children’s expressions supportive manner, while cold are less likely engage with their an encouraging manner. Warm more amongst educated older mothers. Although reveal some persistence, share style decreases child, particular for boys. Children achieve higher cognitive non-cognitive scores at later ages. We find that estimated on different sample splits, such as by education or child age, additional information maintaining robust overall patterns.