Panel data
========
Files: elmps 2012 panel 98_12 v2.1.dta
NOTE: The data are constructed as panel data (wide format) directly from the repeated cross section. Variables repeat as 'variable'_# where # is round, i.e. pn_98, pn_06, pn_12 for the person numbers in 1998, 2006, and 2012 for an individual. If an individual is not present in a round, the variables for that round will be missing. In addition, an indicator variable attr_98_06_12 categorizes individuals based on when they were observed (i.e 98_06_12, 98_06 only, 06_12 only, etc.).
The elmps 2012 98_12 panel dataset includes unique individuals from all rounds, regardless of whether or not they were successfully followed in all rounds. For example, people who were added in 2012 are included, but have missing data for their 98 and 06 variables. If you only wish to work with individuals followed across two or more rounds, use the attr_98_06_12 variable to delete the observations you do not need.
NOTE: The datasets include the expansion factors for each wave as a cross section (e.g. expan_12, expan_06, expan_98) and panel weights (panel_wt_98_06, panel_wt_98_06_12, panel_wt_06_12). Make sure to use the appropriate weight for the specific panel you are working with.
Important Variables:
============
hhid_#: unique household identifier, begins with two-digit round number (88, 98, 06, 12)
indid_#: unique individual identifier, begins with two-digit round number (88, 98, 06, 12)
expan_#: expansion factor for cross-sections ONLY (must be used as an analytic weight)
panel_wt_98_06: panel weight for analyzing panel data with individuals observed in 1998 and 2006
panel_wt_98_06_12: panel weight for analyzing panel data with individuals observed in 1998, 2006, and 2012
panel_wt_06_12: panel weight for analyzing panel data with individuals observed in 2006 and 2012
attr_98_06_12: categorizes individuals based on whether they were in 1998, 2006, and 2012
Cases: | 64192 |
Variables: | 781 |