Demography of Aging Training Seminar- Christina Kamis - Duke University

Measuring Childhood Adversity: Latent Class Analysis with Distal Outcomes

    Date:  12/05/2019 (Thu)

    Time:  3:30pm- 5:00pm

    Location:  Seminar will be held on-site: Gross Hall 270

    Organizer:  Scott Lynch

Meeting Schedule: (Not currently open for scheduling. Please contact the seminar organizer listed above.)

    All meetings will be held in the same location as the seminar unless otherwise noted.

    3:30pm - Seminar Presentation (3:30pm to 5:00pm)

    Additional Comments:  Much of the work connecting childhood adversities and adult health has operationalized adversity as the total number of adversities experienced by an individual. This strategy masks how adversities cluster together, whether certain adversities are more problematic for adult health, and variations within the experience of similar adversities. To better understand the relationship between childhood adversity and adult health, research is increasingly moving toward the use of latent class analysis (LCA) with distal outcomes. LCA is an attractive approach as it can identify unique subpopulations with similar adversity profiles without a priori distinctions made by the researcher and can incorporate more nuanced measurements of adversity. However, there are several available approaches to LCA with distal outcomes, and results may be inconsistent across these approaches. In this talk, I use data from the National Longitudinal Study of Adolescent to Adult Health (Add Health) to evaluate the results from five different approaches to LCA with distal outcomes, including one that is not currently automated in statistical software. I show that different substantive conclusions can be drawn based on the approach chosen by the researcher. This study underscores the need for simulation studies assessing the performance of these approaches against relevant characteristics of social science data.