Using National Education Longitudinal Data Sets in School Counseling Research
Bryan, Julia A., Day-Vines, Norma L., Holcomb-McCoy, Cheryl, Moore-Thomas, Cheryl, Counselor Education and Supervision
National longitudinal databases hold much promise for school counseling researchers. Several of the more frequently used data sets, possible professional implications, and strategies for acquiring training in the use of large-scale national data sets are described. A 6-step process for conducting research with the data sets is explicated: determining research questions, accessing the data set, understanding the study's sample design, determining an analytic sample, considering pertinent data analysis issues, and understanding the limitations of using the data sets. Suggestions for preparing school counseling researchers and potential themes for future research are outlined.
Several researchers have documented that the school counseling profession lacks research demonstrating the effectiveness of K-12 school counseling programs and interventions (Sink, 2006; Whiston, 2002; Whiston & Sexton, 1998). School counselors' seemingly limited response to the paucity of evidence that school counseling programs actually increase student achievement and success may further exacerbate this problem. In 1988, Loesch concluded that research has not been "valued, emphasized, or endorsed as an important role function for school counselors" (p. 170). This disturbing conclusion reached over 20 years ago seems to have had little to no effect on the production of research on school counseling effectiveness or practice. This may place the school counseling profession and the children it serves at risk as important yet unanswered questions related to the effectiveness and relevance of the school counseling profession remain.
School counseling research that uses current national longitudinal databases provides one opportunity to rectify the lack of rigorous research that puts the school counseling profession at risk (Whiston, 2002). National longitudinal data sets function as an important source of information about schools, students, and families because studies that draw on national databases have strong external validity and statistical power (Perna, 2007). High response rates (e.g., 94% for the National Education Longitudinal Study of 1988 [NELS:88]; Curtin, Ingels, Wu, & Hever, 2002) and use of weighted samples provide a more representative depiction of the national population and enhance the generalizability of results. Furthermore, the large number of cases, variables, and oversampling of minority students allow researchers to more effectively detect differences (Perna, 2007) and test theoretical models over time and across samples where appropriate (e.g., NELS:88; Education Longitudinal Study of 2002 [ELS:2002]).
The Education Trust's Transforming School Counseling Initiative (Martin, 2002) and the American School Counselor Association (ASCA) National Model (ASCA, 2005) promote data driven program design, implementation, and decision making in school counseling. The use of national longitudinal databases to examine issues relevant to school counseling may stimulate scholarly research that informs school counseling practice and education and school reform policies. These policies are often derived from rigorous research investigations that used large-scale, longitudinal databases (Swanson & Barlage, 2006; St. John, 2004). Examining issues using national databases does not diminish the importance of school counselors collecting school-based outcome data to examine the impact of individual programs nor does it ignore the importance of school counseling professionals collecting and maintaining national longitudinal data. However, school counseling research that uses national longitudinal databases could provide an immediate response to the call of Loesch (1988) and others; would address the lack of rigorous, outcome-based research (Whiston, 2002); and would make audible the relatively inaudible voices of school counseling practitioners and researchers. In the sections that follow, we describe several national longitudinal data sets, explicate a six-step process for conducting research with these databases, and suggest potential areas of future research. …