Crises can spur improvement and lead to novel approaches to find solutions. Such a crisis occurred in the area of vocational education information in the early 1980s. In the featured article for this issue of the Quarterly, Lisa Hudson recounts how the National Center for Education Statistics (NCES) developed the Data on Career/Technical Education Statistics (CTES) in the mid-1980s to address this crisis. Since that time, CTES has evolved into a system that draws its data from several national data sources, supplemented by special studies to fill information gaps. As such, it has developed into a possible model for other NCES efforts that necessarily cut across institutional boundaries, levels of education, and cross-sectional and longitudinal data collections. In a division responsible for crosscutting work at NCES, I am keenly aware of the potential for synergy that such an approach can offer. I also recognize the challenges that this type of system poses for coordination, analysis, and measurement.

The CTES evolved to meet a clear need. Policymakers at all levels needed a reliable, accurate source for information on vocational education, an important segment of American education at both the secondary and postsecondary levels. To bridge different types of vocational education providers and participants as well as levels of schooling and training required a new approach. CTES diverges from regular NCES data collections in that it synthesizes data from a wide range of preexisting education surveys. It works as a "derived," or "synthetic," system, pooling information from a host of sources. The originators of CTES found that nationally representative sample surveys conducted by NCES and other federal offices could respond to several of the key policy issues that Congress and other policymakers needed addressed. Much of this information was already available, although work was needed to fill information gaps and apply consistent concepts and definitions.

The CTES approach also allows NCES to relate vocational education to the larger education system or to link experiences at the secondary level to those at the postsecondary level. Furthermore, the variety of cross-sectional and longitudinal surveys in the CTES provides a rich data source for policy analysis and research, as well as for basic descriptive purposes. Nonetheless, the CTES must continue to evolve in the face of challenges—challenges that may confront any derived system that is so dependent on cooperation across different offices and coordination across different data sets. These challenges fall into three areas: institutional, conceptual, and methodological.

Institutional challenges. Working across different data sets requires working across different offices and greater collaboration across the Center and the Department of Education. This institutional challenge is common to much of the crosscutting work that our division undertakes. More specifically, bringing together studies that began life as separate surveys presents analytical challenges. Modifying surveys to better address vocational education can also conflict with institutional patterns of behavior; because regular data collections are often conducted to report on trends, a premium is placed on maintaining consistency in data elements over time. To address some of these issues, NCES has institutionalized for CTES some of the features that characterize a major data collection system, including a Technical Review Panel and mechanisms such as a written planning document and a system for survey review designed to encourage collaboration. At the same time, CTES does not have the visibility of a regular data collection; for example, it does not appear in the NCES budget, and so must continuously press for recognition and resources.

Conceptual challenges. Conceptual challenges arise from the ongoing debate on what constitutes vocational education, its role in education as a whole, and the appropriate goals and outcomes for vocational education. This is true as well for other topical areas of inquiry that garner public attention and debate. Conceptual challenges will vary depending on the purpose of the data collection system. For example, vocational education has its unique conceptual challenges; a data collection system for monitoring the progress of minorities in education would face a different set of conceptual challenges, as would a system to assess education finance. Vocational education itself, particularly at the secondary level, is also undergoing profound changes. Reforms such as the integration of vocational and academic education, the articulation of secondary and postsecondary education, and the adoption of high school career clusters, service learning programs, and applied academic courses are further blurring distinctions between vocational and academic education, thereby making vocational education harder to define and identify. Again, keeping up with the pace of change confronts other topical areas as well. For example, new forms of telecommunications and distance learning require that we continuously upgrade our definitions of advanced technology (while also maintaining the ability to monitor change over time). In response, we often rely upon quicker vehicles, such as the NCES Fast Response Survey System (FRSS) and Postsecondary Education Quick Information System (PEQIS), to gather such information.

Methodological challenges. Methodological challenges include the need to balance basic descriptive information and policy-relevant information. This balance affects not just the questions asked but also sampling procedures, the timing of surveys, and the type of survey that is conducted. Sampling issues also arise because the populations that are often of the most interest to policymakers (e.g., disabled students or limited-English-proficient students) are those for whom information is most difficult to collect in general-purpose surveys. Finally, we need to refine both our measures of outcomes of vocational education and our measures of processes, such as the quality of instruction in vocational courses.

Some of these challenges will surely be overcome in the near future, while others may prove to be more resistant to change. Nonetheless, the CTES could someday be the model for other data collection systems that focus on specific sectors of education, such as mathematics and science education, special education, or lifelong learning. Advantages that could ensue include increasing the analytic potential to inform education decision-makers and researchers; reducing the data collection burden on survey respondents; and, potentially, lowering costs. CTES has shown the possibilities that a synthetic system can offer. It has also demonstrated how such a system—through the continuous cooperation of our sponsors, respondents, and data users—can work.