Ga direct naar de inhoud
Ga direct naar de site navigatie
Ga direct naar zoeken
Thomas Pollet, born in Wilrijk (Antwerpen), Belgium, 1981. Ph.D. from Newcastle University, UK. Assistant Professor in Social and Organizational Psychology at VU University Amsterdam.
Fellow (1 September 2015 - 30 June 2016)
Which factors influence women’s decision to have a(nother) baby? To date many approaches have tried to understand and model women’s fertility decisions and outcomes but with varying success. In this project we propose a data mining approach relying on existing large databases, whereby we use machine learning to predict women’s reproductive decisions and outcomes.
A striking feature of 20th-21st century Western industrial societies is that, in general, people have fewer children than at any point in recorded history, and many remain childless altogether. Models derived from Demography, Economics, Sociology, to Evolutionary Biology, have attempted to explain these phenomena. Across a range of studies a broad range of key variables, which should matter for women’s fertility decisions (and outcomes) have been suggested (e.g., relationships to kin, educational attainment,…). The emerging picture is complex, however, and the influence of all these factors is not easily disentangled. A potential first step to providing greater clarity would be to better understand which variables really matter. To this end, we will use machine learning techniques to extract key variables from complex datasets on fertility.
1) Pollet, TV, Stulp, G., Henzi, S. P., & Barrett, L. (2015). Taking the aggravation out of data aggregation: A conceptual guide to dealing with statistical issues related to the pooling of individual‐level observational data. Am J Primatol. doi: 10.1002/ajp.22405
2) Pollet TV, Nettle D (2008) Driving a hard bargain: sex ratio and male marriage success in a historical US population. Biol Lett 4: 31–33. doi:10.1098/rsbl.2007.0543.
3) Nettle D, Pollet TV (2008) Natural selection on male wealth in humans. Am Nat 172: 658–666. doi:10.1086/591690.
Ga terug naar de bovenkant van deze pagina
Ga terug naar de inhoud
Ga terug naar de site navigatie
Ga terug naar zoeken