Table 3a

Testing whether individual and family characteristics at time t predict location at time t+ 1, conditional on rank and occupation interactions. Sample: Children aged 0–5 who moved

VariableSampleNumber of basesMeanStandard DeviationNumber of basesMeanStandard Deviation
Panel A
Children aged 0–5 regardless of location in continental United States who moved
to any base, excluding foreignto any base, including foreign
One equation per year and base, control for rank*pmos
D = 1 if p < 0.05
ages 0–53800.0660.2483860.1060.309
Panel B
Children aged 0–5 in estimation sample who moved (restrict to bases close to monitors and military hospitals)
to any base, excluding foreignto any base, including foreign
One equation per year and base, control for rank*pmos
D = 1 if p < 0.05
ages 0–52650.0000.0002710.0180.135
One equation per year and base, control for rank*pmos
D = 1 if p < 0.05
ages 0–11600.0000.0001660.0060.078
One equation per year and base, control for rank*pmos
D = 1 if p < 0.05
ages 2–52430.0000.0002490.0080.089
  • Linear probability models. Errors clustered at the sponsor level. Control variables tested include age, gender, health variables (whether hospitalized, hospitalized in MTF, hospitalized in MTF for respiratory condition), father/sponsor’s controls (number of months since last enlistment, total active months in the military, age, white dummy, college degree, number of dependents, enlisted in the last five years), and mother’s health (whether hospitalized, hospitalized in MTF, hospitalized for pregnancy-related).