Plenary, panel, and breakout discussions all included lively and sometimes sharp exchanges about government, university, industry, and school system initiatives to prepare, recruit, and retain underrepresented populations for the STEM workforce. These discussions allowed participants to draw the following generalizations about the characteristics of successful programs.

Lessons Learned from Prior/Existing Programs

Leadership: Strong leadership at all levels is a key to success of any program. Successful programs commonly are championed by individuals dedicated to long-term improvements who can obtain buy-in from others in leadership positions (BEST, 2004). The importance of champions points to the need for succession planning for leadership (in part, to prevent burn-out and encourage new ideas) and for the continuity of institutional commitment. In addition, leadership must come from multiple levels and a range of communities such as academe, industry, professional societies and government.

Mentoring: A strong mentoring or coaching component, whether between researchers and teachers, between graduate and undergraduate students, among peers, or involving other groups, is characteristic of many successful programs.

Institutional Support: The long-term success of programs depends on engaging institutional support and commitments from a broad network of partners among faculty, schools and departments, students, industry, and the community. Programs with goals that are aligned with those of their home institution are more likely to thrive and become a priority for local funds.

Funding: Many successful programs have diverse sources of funding to ensure program continuity. Although the highly competitive nature of the grant-seeking process may help clarify and strengthen program goals, funders need to be flexible to allow for innovation and risk-taking based on program evaluations or external peer review. To that end, it is essential that proposal review panels reflect the personal and institutional diversity desired for the project outcomes.

Measures of Success: Measurable objectives and formal evaluation are crucial for assessing program success (BEST, 2004). Assessments should provide continuous feedback to guide program design, planning, and implementation and offer opportunity for partnerships with social scientists engaged in evaluative research on STEM workforce issues. Social science and education researchers should be called upon to play a key role in developing meaningful metrics, designing robust evaluation protocols, and implementing the assessment process for projects (Levine, Abler, and Rosich, 2004).

Focus on Transition Points: Students take multiple and intersecting career paths to STEM careers, many of which can be idiosyncratic and therefore hard to predict. Innovative solutions to recruiting students into STEM fields can come from looking at intersections and transitional points in the educational and workforce system. In addition, a focus on transitions can enhance an individual’s successful academic and social integration into a scientific discipline.

Focus on Communities: Many programs are successful because they are targeted at a specific underrepresented group, educational level, or problem. At the same time, this focused approach can serve broader groups — for example, programs designed to encourage females or underrepresented minorities to pursue STEM careers have often proven to be equally helpful for all demographic groups of students and workers.

The Dangers of Isolation of Programs: While workshop participants acknowledged that many institutions have implemented activities that can create pathways to STEM careers, they also pointed out that many existing programs have inherent and often severe limitations. The underlying problem with many programs is that they do not build on other successful programs to create a whole that is greater than the sum of the parts. “We have numerous examples of projects that are successful,” said one workshop participant. “But these are individual projects. We need to change the overall system if we are to achieve critical mass.” Many programs have been developed by trial-and-error and in relative isolation. As a result, programs tend to engage in a phenomenon of “parallel discovery,” where similar programs have evolved less efficiently than would have been the case had better communication existed. These isolated programs have few ways to communicate with other programs to share experiences and incorporate improvements. Overall, targeted programs that address diversity issues often suffer from the same shortcomings:

1 Loss of a champion for a program, which can undercut its effectiveness or even lead to its discontinuation.

2 Being developed and operated in a piecemeal fashion without links to other programs or evaluations that can guide future activities. Because of the isolation of such programs, they are not viewed as part of a larger whole.

3 Despite their successes, many programs have remained marginal to the core activities of institutions. Such programs can be eliminated without affecting the institution of which they are a part, and in times of budget constraints these programs are prime candidates to be eliminated.


.