Working Paper: Love and Machine! Applying Machine Learning to Investigate Mating of Single Parents in Large-Scale Online Dating Platform Data

Single parenthood emerged as a common family model in western societies. Congruently, the comprehension of mating behavior of (and with regard to) single parents seems to be pertinent not only because single parents typify a significant proportion in (contemporary) mating markets (online dating platforms) but also because mating behavior of (or with regard to) single parents is expected to enfold downstream consequences on both respective single parents and associated children. However, only little is known about mating behavior of (and with regard to) single parents in the early stage of mate selection. In this research, we apply causal machine learning methods (Causal Forests) to examine mating behavior of (and with regard to) single parents in the early stage of mate selection based on high-dimensional field data. The findings of this research provide important implications for theory and practice. First, the insights of this research contribute to most recent research on mating behavior of (and with regard to) single parents. In particular, this research shows that single parents relative to childless singles are punished in the mating market and that the punishment of single parents in the mating market is particularly significant for female relative to male single parents. Second, the insights of this research advice econometricians and business analysts on the application of causal machine learning methods (Causal Forests) to derive valuable insights in the context of high-dimensional field data. Finally, the insights of this research support market designers and / or software developers in improving existing matching algorithms.