Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.authorBoertien, Diederiken_US
dc.contributor.authorvon Scheve, Christianen_US
dc.contributor.authorPark, Monaen_US
dc.description.abstractThe reasons why the lower educated divorce more than the higher educated in many societies today are poorly understood. Distinct divorce risks by education could be caused by variation in pressures to the couple, commitment, or relationship skills. We concentrate on the latter explanation by looking at the distribution of personality traits across society and its impact on the educational gradient in divorce in Germany. Using data on married couples from the German Socio Economic Panel (N = 9 417) we first estimate the effect of several personality traits on divorce: the tendency to forgive, negative reciprocity, positive reciprocity, and the Big Five. We also account for and find non-linear effects of several personality traits on divorce risk, which is relevant for future research on the effects of personality. In addition, effects differ by level of education. We find personality traits that affect divorce risk to be unevenly distributed over educational groups, but contrary to expectation to favor the lower educated. Once taking into account personality the educational gradient in divorce becomes more negative. This is due to especially high scores on openness to experience for the higher educated, which is a very significant predictor of divorce risk. Overall, we find no support for the hypothesis that the lower educated have less relationship skills in Germany.en_US
dc.publisher|aDeutsches Institut für Wirtschaftsforschung (DIW) |cBerlinen_US
dc.relation.ispartofseries|aSOEPpapers on Multidisciplinary Panel Data Research |x487en_US
dc.titleEducation, personality and separation: The distribution of relationship skills across societyen_US
dc.typeWorking Paperen_US

Files in This Item:
232.11 kB

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.