There has been an increase in the use of partnerships as a policy prescription for improving education since the mid 1980's. This trend builds on nearly a century of reform movements in education, has been noted in the professional literature and can also be seen in the growing trend to add partnership requirements to federal grant programs in a variety of fields as reported in the "Federal Register". Policy makers include partnership requirements as a condition for receiving the funds in grant programs, as an inducement to engage in collaborative behavior, assuming that there will be some benefit to the inclusion of partners in program implementation that will lead to improved policy outcomes. The motivation for this study is the desire to answer the question of whether the partnership requirement included in these federal programs constitutes good public policy. In particular, this study focuses on whether these partnerships as typically constituted provide the necessary conditions for the adaptive management (sustainability and locally adaptable action) of education problems by the local community partners. The results of this study suggest that the implementation and content requirements built into grant programs, that include partners as a condition in aid, most often result in a narrow programmatic focus among the participants. Organizations choose to participate in disjointed serial interventions that support organizational needs or goals based on the availability of funding and partners for particular programmatic activities. They choose partners from among those who are interested in similar or complementary activities. The primary focus of STEM education partnerships is therefore on implementing and sometimes evaluating the funded programmatic activities and not on building a broader learning community. Activities or education problems that are not funded tend to be excluded from the activities and dialog of the policy-induced partnership. By limiting the scope of the collaboration we are limiting the value of these partnerships. An important theoretical benefit of partnerships, found in both the social capital and learning community literature, is as a venue for sustainable dialog and adaptive decision making (i.e., adaptive management). In our current environment of programmatic accountability, education evaluators are underutilizing this large established literature. Building off this older literature, this study reminds educators and evaluators that partnerships can be a meaningful end and not simply serve as an implementation tool. Investment in partnerships, and the social networks that they are developed from, creates valuable a resource that can be used to enhance future programmatic work. Public policies that focus too narrowly on programmatic goals, even when they require partnered implementation, are unlikely to provide a stable, sustainable, value generating core that will last beyond a funded intervention. Wise grant seekers naturally look for funding opportunities that may or may not include current partners. Instead of building from a stable partnership core, organizations interested in a policy problem or choice of intervention seem to form a loose issue oriented network, parts of which are activated as available funding is identified. Once an intervention's funding is exhausted the participating organizations gradually slip back into a loosely connected network of potential partners. This finding has led to conceptualizing policy-induced partnerships as relational events that are bounded by funding cycles. And, any attempt at adaptive management within the partnerships developed under these conditions will be far less deliberative, less adaptive, less sustainable, and more opportunistic in nature. Policy prescriptions, with a wider focus, are more likely to provide a positive return on our education investment in the form of greater opportunities for sustainable collaborative relationships. Education partnerships as currently implemented are unlikely to provide a stable venue for adaptive management and the consideration of broader education problems. As long as the policy focus is on delivering desired interventions or programs, partnership will be just one of many implementation options. Further, the requirement for specific kinds of partners as a condition in aid for grant programs has the potential to crowd out other beneficial collaborative interactions. However, the broader networks of those interested in a policy issue do seem to be engaging in a loose form of adaptive management. Elements of a wider issue network continually explore local needs and options becoming focused and activated around opportunities to address problems through partnered interventions. Policy makers need to reconsider the literature on building learning communities and support the development of a venue for greater network connectivity when designing education policy and programs. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com.bibliotheek.ehb.be/en-US/products/dissertations/individuals.shtml.]