My career in the federal government began on a high note—I was offered a position that enabled me to work on the Longitudinal Studies Program at the National Center for Education Statistics (NCES). I was being given the opportunity to take an active role in helping to design studies that collected data from nationally representative samples of students over selected periods of time. This is where the action was at NCES—this is where I wanted to be.

Although my job description at NCES has changed several times in the past 22 years—from education statistician to program officer to program director and, most recently, to associate commissioner—I am still closely associated with longitudinal studies. On a daily basis, I review questionnaires, examine participation rates, and make decisions that I hope will help NCES to produce user-friendly data sets that provide researchers with data that can be used to tell the stories of both those students who thrive in the education system and those who fail. There are also the stories of students who, judging by their home and academic background, are projected to fail, but instead choose a path that leads to success.

During my 22 years at NCES, I have been associated with four major longitudinal studies that follow students through high school into postsecondary education and/or the world of work. These are the

• National Longitudinal Study of the High School Class of 1972 (NLS:72)—a cohort of 12th-graders that was followed for 14 years;
• High School and Beyond Longitudinal Study (HS&B)—a cohort of 1980 high school sophomores and a cohort of 1980 high school seniors that were followed through the 1980s;
• National Education Longitudinal Study of 1988 (NELS:88)—a cohort of eighth-graders in 1988 that was followed through the year 2000; and
• Education Longitudinal Study of 2002 (ELS:2002)—a cohort of 10th-graders that will be followed between the years 2002 and 2014.

Unlike most cross-sectional studies, which have a limited life due to the age of the data collected, the usefulness of longitudinal studies for research is extended over time. In fact, longitudinal data can be used to conduct cross-sectional (single point in time), longitudinal (across time with the same individuals), or trend (between different cohorts) analyses. The story told in the featured article of this issue of the Education Statistics Quarterly—an excerpt from the NCES report Coming of Age in the 1990s: The Eighth-Grade Class of 1988 12 Years Later—is longitudinal—a cohort of eighth-graders from NELS:88 is examined in 1988 and then again in the year 2000. Because NELS:88 followed a group of eighth-graders for 12 years, it is possible to associate past events with later educational and occupational outcomes.

The members of this eighth-grade cohort were born at the end of the Vietnam War (1974), when handheld calculators were not used with great frequency and personal computers had just been invented. They grew up, though, in an era that experienced numerous changes in the fields of communication, technology, medicine, and transportation that influenced their day-to-day lives. They experienced explosive growth in the computer industry and participated in secondary and postsecondary education on the threshold of a new millennium dominated by personal computers. Given the rapidly paced society in which this cohort matured, some questions naturally arise: Have these individuals been prepared for the 21st century? How much education do they have? What occupations are they entering? Are they starting families? Coming of Age in the 1990s provides insight into these kinds of questions.

Over 12,000 eighth-graders were surveyed both in the base year (1988) and in the fourth follow-up (2000) of NELS:88. Their responses were coded and then analyzed. Using the findings of these analyses, a story has been told about their lives—both the successes and the failures. In telling such stories, longitudinal studies have an advantage over cross-sectional studies because they provide both background and outcome variables. The background variables (e.g., family characteristics, eighth-grade courses) can be used to predict later outcomes such as college or career success. Background variables do not always work well as predictors, however. For example, there are always groups of students who succeed when background variables suggest a higher risk of failure. There are also groups of students who fail (e.g., drop out of high school) when advantaged backgrounds suggest more favorable outcomes. These kinds of stories (both predictable and non-predictable) can be used by researchers, policymakers, schools, and parents to better inform decisions regarding the education experiences that are selected for our nation's youth. The featured article takes a first look at the year 2000 outcomes experienced by the eighth-grade class of 1988.