The estimates presented in this report were produced using the BPS:1996/2001 Data Analysis Systems (DAS). The BPS:1996/2001 Data Analysis System includes all of the variables from the BPS:1996/1998 study and several sample weights for a cross-sectional analysis of the students in either 1995–96, 1998, or 2001 as well as weights for the longitudinal analysis of students who responded in any two or in all three of the survey years. All of the tables and estimates in this report used the longitudinal analysis weight B01LWT2 for about 9,000 sample students who responded in both the first (1996) and the last year (2001).

The DAS software makes it possible for users to specify and generate their own tables. With the DAS, users can replicate or expand upon the tables presented in this report; the table parameter files (tpf) that produced these tables are available to users on the NCES website. In addition to the table estimates, the DAS calculates proper standard errors⁵ and weighted sample sizes for these estimates. If the number of valid cases is too small to produce a reliable estimate (less than 30 cases), the DAS prints the message “low-N” instead of the estimate.

In addition to tables, the DAS can also produce a correlation matrix of selected variables to be used for linear regression models. Included in the output with the correlation matrix are the design effects (DEFTs) for each variable in the matrix. Since statistical procedures generally compute regression coefficients based on simple random sample assumptions, the standard errors must be adjusted with the design effects to take into account the BPS:1996/2001 sample design.

The DAS can be accessed electronically at http://nces.ed.gov/das. For more information about the BPS:1996/2001 Data Analysis System, contact:

Aurora D’Amico
National Center for Education Statistics
1990 K Street, NW
Room 8115
Washington, DC 20006
(202) 502-7334
Internet address: Aurora.D’Amico@ed.gov