There’s more in the data! Using month-specific information to estimate fine-grained trajectories of life satisfaction around life events
*Ansgar Hudde & Marita Jacob
Universität zu Köln
Life course research has increasingly used panel data to examine changes in outcomes, such as life satisfaction, over the course of life course events. Such analyses typically estimate fixed- effects regression models in which observations are grouped into one-year intervals. However, this approach discards valuable information and hides any relevant changes within these twelve-month intervals. The contribution of this paper is to show how such changes can be modelled as continuous, fine-grained trajectories using available month-specific information on the time between live events and surveys. We illustrate our approach and test the underlying assumptions by examining changes in life satisfaction in the two years before and after the family life events, first birth and bereavement, using SOEP data. The results identify anticipation effects and discontinuities in life events and show that changes in people's satisfaction are much larger and faster than "standard" fixed effects models suggest. The approach presented here can be applied to a wide range of outcomes and life events and produces new substantive findings by exploiting the full potential of the available data. For researchers seeking to estimate precise trajectories of short-term change, it is a powerful addition to the methods toolbox.