The National Center for Early Development and Learning (NCEDL) combined data from two studies of state-funded pre-kindergarten programs in order to examine how variations in the programs affect children's outcomes, both at the end of pre-kindergarten and after the transition to kindergarten. The Multi-State Study of Pre-Kindergarten looked at programs in California, Georgia, Illinois, Kentucky, New York, and Ohio during the 2001-2002 school year. The Study of State-Wide Early Education Programs (SWEEP) collected data on programs in Massachusetts, New Jersey, Texas, Washington, and Wisconsin during the 2003-2004 school year. The states selected for the SWEEP study were chosen in part due to their differences from the states included in the Multi-State study. The studies collected information on pre-kindergarten children, classrooms, and teachers, along with demographic information such as race, teacher and child gender, family income, maternal education level, and teacher edcuation level.
For the Multi-State study, the original pool consisted of 19 states that served a minimum of 15 percent or 15,000 four-year-olds. The number of states was decreased to six states chosen to represent a variety of program characteristics and geographical locations. Twenty zip codes were selected from each state, and two sites from each zip code. From each site, one classroom was selected, with four children selected from each classroom. For the SWEEP study, 100 pre-kindergarten sites were selected from each of five states. One classroom was randomly selected within each site, with the stipulation that at least five eligible children had to be in the selected classroom. Both studies excluded children with Individualized Education Plans and children who did not speak competent English or Spanish.