Inconsistent Search Results
Users may experience issues with the search function. We encourage you to browse our pages manually through the navigation until this is resolved. Thank you for your patience.

2014 Panel Wave 1: Household-Level Recodes for Vehicles, Housing, and Education Savings Accounts

The household-level wealth recodes for vehicles, housing, and education savings accounts were constructed incorrectly for a small number of households with a particular type of family composition.  These are the variables THEQ_VEH, THVAL_VEH, THDEBT_VEH, THEQ_HOME, THVAL_HOME, THDEBT_HOME, THVAL_ESAV.  However, the person-level counterparts of these variables (TVAL_VEH, TDEBT_VEH, TEQ_VEH, TEQ_HOME, TDEBT_HOME, TVAL_ESAV, respectively) are constructed correctly for all individuals.  Because of this, data users who want to use these household-level variables should recreate them by summing the person-level variables within a household.  In Stata, this can be done with the egen command, such as “by ssuid shhadid, egen theq_veh = total(teq_veh).”  In SAS, this can be done with the lines “select sum(teq_veh) as theq_veh” and “group by ssuid, shhadid” within a proc sql command.  Note that this issue does not affect the net worth recodes, so THNETWORTH is constructed correctly on the file.

Page Last Revised - October 8, 2021