Chung, Hwan
Anthony, James C.
Schafer, Joseph L.
Latent Class Profile Analysis: An Application to Stage Sequential Processes in Early Onset Drinking Behaviours
Journal of the Royal Statistical Society: Series A (Statistics in Society) 174,3 (July 2011): 689–712.
Also: http://onlinelibrary.wiley.com/doi/10.1111/j.1467-985X.2010.00674.x/full
Cohort(s): NLSY97
Publisher: Wiley Online
Keyword(s): Adolescent Behavior; Alcohol Use;

Permission to reprint the abstract has not been received from the publisher.

Summary: In longitudinal research on early onset drinkers, much attention has been paid to the identification of subgroups of individuals who follow similar sequential patterns of drinking behaviours. However, research on the sequential development of drinking behaviour can be challenging in part because it may not be possible to observe the particular drinking behaviour stage at a given point in time directly. To address this difficulty, we can use a latent class analysis, which provides a set of principles for the systematic identification of homogeneous subgroups of individuals. In this work, we apply a latent class analysis in an investigation of stage sequential patterns of drinking behaviours among early onset drinkers, using data from the National Longitudinal Survey of Youth 1997. A latent class analysis approach is used to sort different patterns of drinking behaviours into a small number of classes at each measurement occasion; and the class sequencing of early onset drinkers over the entire set of time points is evaluated to identify two or more homogeneous early onset drinkers who exhibit a similar sequence of class memberships over time. This approach uncovers four common drinking behaviours in early onset drinkers over three measurements from early to late adolescence. The sequences of drinking behaviours can be grouped into three sequential patterns representing the most probable progression of early onset drinking behaviours.

Davey, Adam
Shanahan, Michael J.
Schafer, Joseph L.
Correcting for Selective Nonresponse in the National Longitudinal Survey of Youth Using Multiple Imputation
Journal of Human Resources 36,3 (Summer 2001): 500-519.
Also: http://www.jstor.org/stable/3069628
Cohort(s): Children of the NLSY79
Publisher: University of Wisconsin Press
Keyword(s): Attrition; Behavior Problems Index (BPI); Data Analysis; Data Quality/Consistency; Longitudinal Surveys; Modeling, Growth Curve/Latent Trajectory Analysis; Nonresponse; Poverty; Psychological Effects;

Survey attrition and nonresponse, particularly when selective, present unique challenges to researchers interested in studying developmental processes and longitudinal change. Four distinct patterns of nonresponse on children's psychosocial adjustment and lifetime poverty experiences and family histories are identified using principal components analysis. In turn, membership in these four groups is significantly predicted by the child's demographic characteristics, family experiences, and previous values on adjustment variables, indicating selective nonresponse and raising the possibility of biased estimates based on listwise deletion of missing data. We then examine a set of latent growth curve models that interrelate children?s family experiences and psychosocial adjustment using listwise deletion (LD) and multiple imputation (MI) procedures. Implications for treatment of nonresponse in national longitudinal surveys are discussed.

Davey, Adam
Shanahan, Michael J.
Schafer, Joseph L.
Corrections for Missingness in the NLSY: Longitudinal Patterns in the Mother-Child Files
Presented: Ann Arbor, MI, Health and Retirement Study, Institute for Social Research, Conference on Data Quality Issues in Longitudinal Surveys, 1998
Cohort(s): Children of the NLSY79
Publisher: Institute for Social Research (ISR), University of Michigan
Keyword(s): Attrition; Data Analysis; Data Quality/Consistency; Longitudinal Surveys; Modeling; Modeling, Growth Curve/Latent Trajectory Analysis; Nonresponse; Poverty; Psychological Effects;

Permission to reprint the abstract has not been received from the publisher.

NOTE - See J. of Human Resources 2001 for published version.
Survey attrition and nonresponse, particularly when selective, present unique challenges to researchers interested in studying developmental processes and longitudinal change. Four distinct patterns of nonresponse on children?s psychosocial adjustment and lifetime poverty experiences and family histories are identified using principal components analysis. In turn, membership in these four groups is significantly predicted by the child's demographic characteristics, family experiences, and previous values on adjustment variables, indicating selective nonresponse and raising the possibility of biased estimates based on listwise deletion of missing data. We then examine a set of latent growth curve models that interrelate children's family experiences and psychosocial adjustment using listwise deletion (LD) and multiple imputation (MI) procedures. Implications for treatment of nonresponse in national longitudinal surveys are discussed