Heterogeneity in job application decisions in two-adult households under social influence: A latent class mixed logit model

Abstract

Work schedules with their start and end times substantially affect traffic flows, particularly during peak hours. Therefore, the study of work schedules is important to understand daily activity travel patterns. Particularly adults in households with children need to trade-off the number of working hours versus their time expenditure on child care, household tasks and leisure. Work schedule preferences are reflected in job application decisions in the sense that people are less likely to apply if the job profile is incongruent with their time use preferences. Job application decisions may also be influenced by attitudes of members of a household’s social network and peer groups. The aim of this study is to analyze unobserved heterogeneity and the effects of job properties and social influence on the decision to apply for a job in two-adult households with children. To that end, a stated choice experiment, in which respondents are asked whether they would apply for a job of a particular profile and a certain set of attitudes of members of their social network, is constructed and implemented. A latent class mixed logit model with two latent classes is applied to estimate the effects of work schedule attributes, social influence and socio-demographic characteristics on the probability of applying for a particular job. Results show that in both classes the effects of job attributes are more significant than social influence, and that the effects of the number of working hours and salary differ show considerable unobserved heterogeneity.