Sampling design overview

Stratified random sampling

1. Stratified random samples according to sex, race/ethnicity, and socioeconomic status (measured by regionally specific, US Housing and Urban Development HUD adjusted income of the parents as a surrogate variable) were obtained at each PSC in order to match total sample proportions of these factors, marginally across all PSCs, to those of the entire U.S. population as determined by cross-tabulated US Census 2000 family income by race/ethnicity proportions.

2. In order to address concerns regarding a “super-normal” sample population, we analyzed groups of data stratified according to the full-scale WASI IQ of the subject are available to also match nationwide IQ distributions.

Age targets, limits, and groups

Target ages were staggered at entry with 20% compounded attrition for both Objectives; please see the Sampling Strategies section in the NIHPD Protocol for more detail.

Age at study entry

Objective 1: 4 years 6 months (Minimum) - 18 years 3 months (Maximum)

Objective 2: 10 days (Minimum) - 4 years 5 months (Maximum)