chapter. The reader is referred to Appendix A for an explanation of the study approach, which included extensive review of the literature and various background documents, as well as interviews with key personnel at both the federal and state levels.
Earlier chapters of this report have assessed the state of knowledge about the linguistic, cognitive, and social development of English-language learners and about the programs and teachers that educate them and data collected on them; they have also offered observations on the quality of the research. This chapter assesses the infrastructure that produced much of that research and identifies the characteristics that seem to have facilitated or inhibited good research. Our principal judgment, resting largely on the reviews included in previous chapters, is that the infrastructure has often failed to produce the high-quality and relevant research needed, this despite a great expansion of research on LEP issues in the past 15 years and the strenuous and skilled efforts of many researchers and agency officials. The effectiveness of the infrastructure has been strongly influenced by some factors we cannot hope to change, such as the politics of bilingual education. But we can recommend changes in organization, procedures, and allocation of resources that might improve the infrastructure, and changes in training that might strengthen the skills of the people within that infrastructure in the future. The final section of this chapter, then, presents a set of recommendations for addressing the issues listed above, and thereby improving the infrastructure for research on English-language learners and bilingual education.
Federal research funds for the study of education have always been very modest and unpredictable. Thus, the possibilities for rational agenda setting are constrained. Agenda setting in education research is always tentative; the major players are always changing; and the process is always vulnerable to interruption, undue haste, politics, and controversy. Even during periods when funding has been fairly level, as with the laboratories and centers, the agenda-setting process has been haphazard, sometimes mandated by Congress, sometimes left to internal agency staff, sometimes involving extensive participation by practitioners and other stakeholders, and sometimes left largely to the discretion of research center directors.
Congressional mandates relevant to agenda setting are of two sorts: substantive and procedural. An example of a substantive agenda provided by Congress is the 1978 reauthorization of the Bilingual Education Act, which specified eight areas of research to be conducted by the new Title VII: studies to determine and evaluate effective models for bilingual-bicultural programs; studies to determine language acquisition characteristics and the most effective method of teaching