Cognitive Test Battery
The test questions were selected from previous assessments: NELS:88, the National Assessment of Educational Progress (see NAEP chapter), and Program for International Student Assessment (see PISA chapter). Most, but not all, were multiple choice items. Test specifications for ELS:2002 were adapted from frameworks used for NELS:88. Math tests contained items in arithmetic, algebra, geometry, data/probability, and advanced topics were divided into process categories of skill/knowledge, understanding/ comprehension, and problem solving. Through inclusion of items from the PISA, the ELS:2002 math tests placed a somewhat greater emphasis on practical applications and problem solving than did the NELS:88 test forms. Reading tests consisted of reading passages of one paragraph to one page in length, followed by three to six questions based on each passage. The reading passages included literary material as well as topics in the natural and social sciences. Several passages required interpretation of graphs. Questions were categorized as reproduction of detail, comprehension, or inference/evaluation.
A cohort is a group of individuals who have a statistical factor in common; for example, year of birth, grade in school, or year of high school graduation. ELS:2002 is a sophomore-grade cohort based on the spring term of the 2001–02 school year. It also contains, however, a nationally representative sample of high school seniors in the spring term of the 2003–04 school year.
Socioeconomic Status (SES)
A composite variable is constructed through the combination of two or more variables—socioeconomic status, for example, combines mother’s education, father’s education, mother’s occupation, father’s occupation, and family income or an income proxy (household items) or it is calculated through the application of a mathematical function or transformation to a variable (e.g., conversion of raw test scores to percentile ranks).
Dropouts were defined in ELS:2002 as sample members who had been absent from school for 4 or more consecutive weeks at the time of the survey and who were not absent due to accident or illness.
Early graduates were defined as sample members who had graduated from high school or obtained certification of high school equivalency (e.g., obtained a GED credential) on or before March 15, 2004.