hypotheses and used different methodologies to begin to explain why young women were less likely to take elective mathematics courses, with a view to changing the situation once it was better understood. About a decade ago, we began to hear and act upon the findings of this research.
The result was a wealth of intervention programs in science, mathematics, and engineering targeted toward women and focused on improving attitudes and increasing interest and participation. Perhaps more importantly, the design of many of these intervention programs was based upon the recently developed research base. In fact, there was tremendous respect and collegiality and considerable interaction between and among the scholars and the practitioners. This close coupling of research and practice contributed to the way that intervention programs developed and were implemented and evaluated. An appropriate definition of an intervention program should reflect this research-practice connection: Having identified a problem to solve, select and implement a strategy (either to change the situation or to compensate somehow for a situation that you cannot change) and then continually monitor to see if your strategy is successful. The definition is, therefore, simple, straightforward, and empirical.
There is great diversity in the approach taken by intervention program designers, beginning with the way that the problem-to-be-solved is described. First, there is a tension between a recognition of the complexity of the situation and not being overwhelmed by it. A myriad of factors combine to influence a child's interest in science or mathematics studies and related careers. These include both cognitive and affective factors, which vary depending upon the age of the child. Other important influences originate in the support and educational system in which the child functions: school, home, extracurricular activities, peer, and the media, for example. Previous research has shown us that these factors tend to work in combination with each other; therefore, we know that a single intervention event is unlikely to change all of the factors involved. Rather, systemic change in a variety of areas is needed. It is unlikely, however, that one can attempt to simultaneously approach all of these problems. One must define a manageable problem, one on which we can reasonably hope to have impact. This is especially