NSF Grant Number HRD 9817536, A002
( Links to report sections )
The Directorate for Education and Human Resources Programs (EHR) of the American Association for the Advancement of Science (AAAS) is grateful to the National Science Foundation (NSF), Alliances for Graduate Education and the Professoriate Program (AGEP) for its generous support of this research project. We especially appreciate the support of the NSF staff, including Norman Fortenberry, Division of Undergraduate Education; Roosevelt Johnson, Human Resources and Development; and Bernice Anderson, Division of Research, Evaluation and Communications.
We particularly appreciate the work of study group leaders and recorders:
We extend a special thank you to the following individuals for prioritizing the report's recommended new research areas:
Our appreciation to Senayt Assefa, Betty Calinger and Jolene Jesse at AAAS who helped edit the report.
And a thank you to the editorial and design team at Potomac Communications Group, Inc., Washington, DC.
Today the United States is the world leader in the global science, technology, engineering and, mathematics (STEM) enterprise, but other countries stand ready to challenge this economic strength. One of the main reasons is a shortage of U.S. workers to fill STEM jobs. Technically skilled workers on H-1B Visas (guest workers) are now making up for the U.S. worker shortfall. This supply of talent could dwindle in the near future as other nations take steps to increase their own STEM productivity. Another reason is that the majority of the current STEM workforce, White, non-Hispanic men, is shrinking. In 1995, the projected percentage of White men in the overall workforce was 36%. By 2050 White males are projected to be 26% of the overall workforce, while in 1997 they represented nearly 70% of the STEM workforce.1
In our efforts to sustain U.S. productivity and economic strength, underrepresented minorities (URM) (for the purpose of this paper defined as persons of African American, Hispanic American, and Native American racial/ethnic descent), provide an untapped reservoir of talent that could be used to fill technical jobs. Over the past 25 years, educational diversity programs have encouraged and supported URM pursuing STEM degrees. Yet, their representation in STEM still lags far behind that of White, non-Hispanic men.
To understand the reasons why this is occuring, the American Association for the Advancement of Science Directorate for Education and Human Resources Programs convened a study group meeting in September 2000 of 70 leading educators and researchers in the STEM fields. We examined over 150 research efforts related to choice of college majors, retention in STEM college majors, academic mentoring at both the pre-college and higher education levels, and pursuit of a STEM doctorate, as well as faculty positions. At the study group meeting, we discussed key research, identified gaps, and developed a research agenda for the future. Particular attention was paid to the transition process from one level of academic achievement to the next.
We identified three research priorities for URM in STEM from the high school years to the professoriate:
To improve research linkages, we recommend developing data collection guidelines and definitions, using common research methods and developing models that will permit cross-comparison of findings in a wide range of studies. We also recommend establishing a research consortium.
In addition, we encourage organizations like the National Science Foundation, and other government agencies to foster STEM education research coordination. We also recommend that these organizations maintain and build databases to provide information about the education and workforce experiences of URM, women, and persons with disabilities in STEM.In our efforts to sustain U.S. productivity and economic strength, African Americans, Hispanic Americans, and Native Americans provide an untapped reservoir of talent that could be used to fill technical jobs.
Continued collection of critical data. Researchers need to continually collect data from higher education enrollment, STEM course taking, and graduation for different groups, as well as data from different types of colleges and universities. Curriculum alignment between community and four-year colleges also should be monitored. Specific attention should be given to monitoring the impact of changes in higher education admissions, retention, and graduation policies at both the state and national levels.
research to better understand factors that facilitate or limit student
progression towards doctoral degrees and faculty positions in STEM.
We need to study the reasons why able and high achieving URM do not enter
STEM college majors or, if they enter, search for the reasons why they
do not complete STEM higher education degrees or go on in higher education
to pursue doctoral careers in academe. Suggestions for further research
Members of our study group meeting hope that these suggestions will provide guidance, and strategies for policymakers, researchers, educators, and public and private foundation staff who want to build and sustain STEM education research on URM. The better the quality of information we have on what facilitates and what limits URM's progress towards STEM doctorates, the better equipped we will be to create educational policies and programs to address them.
Building a diverse workforce in science, technology, engineering and mathematics (STEM) is increasingly important to sustaining the nation's productivity and economic strength. Evidence already exists that the lack of United States citizens in the STEM workforce is limiting economic growth, and business has looked to H-1B Visas (guest workers) as a way to fill this gap. However, recognizing the connection between sustained economic growth and a technically trained workforce, other nations are aggressively restructuring higher education and workforce policies to keep their nationals at home.2
worldwide education and workforce policy trends can have a negative effect
on the nation's economy, particularly if enough U.S. workers are not technically
Traditionally, STEM workers have been White, non-Hispanic men. In 1997, the general STEM workforce was comprised of nearly 70% White men.3 Their numbers, however, as a percentage of the U.S. population are declining. The Census Bureau reports that the population of White, non-Hispanic men is expected to decrease by 10% by the middle of this century.4 This could leave an enormous gap in the American workforce that must be filled. (See Chart right.)
United States has an untapped reservoir of talent that could be developed
to fill technical jobs. Underrepresented minorities (URM) in STEM (for
the purpose of this paper defined as persons of African American, Hispanic
American, and Native American racial/ethnic descent), should be encouraged
to pursue STEM education from high school to the doctoral level. In 1997,
URM comprised just over 6% of the general STEM workforce.5
URM accounted for only 4.6% of the STEM workforce with doctoral degrees,
compared to nearly 80% for White men.6 (See chart on Racial/Ethnic
Distribution of U.S. STEM Workforce in 1997and chart on The STEM Workforce
with Doctoral Degrees in 1997.)
a strong U.S. STEM workforce is to be ensured, it is imperative that this
nation understand how to encourage and develop the STEM talent of all
U.S. citizens, including all racial/ethnic groups, men and women, and
persons with disabilities.
study group met for seventeen hours, and participants considered a variety
of questions based on current STEM research. The goals of the study group
were to review current research on STEM education, identify gaps, and
recommend research priorities for future studies.
Each of the six categories was assigned to an individual study group. The groups examined a selection of key research papers within their assigned categories, that focused on the STEM experiences of URM, women, and persons with disabilities from high school to the professoriate. They also answered questions developed to stimulate discussion and reviewed bibliographies of current research. (See Appendix B for a complete list of discussion questions.) Each study group had an assigned leader who was responsible for producing a written report by the end of the meeting.
meeting produced several insights into areas of STEM research that call
for further exploration as well as three specific research priorities.
These recommended research priorities, with regard to URM in STEM from
the high school years into the professoriate are:
We hope that it will provide guidance to researchers about methodology, areas of study, and the most effective ways to share information. In addition, we hope these suggestions will provide guidance and strategies for policymakers, educational leaders, and public and private foundation staff who want to build and sustain STEM education research on URM.
The following is an assessment of what we know from the existing research on URM in STEM for the high school, undergraduate, and graduate years, as well as the professoriate. With input from the study group participants, over 300 quantitative and qualitative studies were identified, including ongoing data reporting and studies done by the U.S. Department of Education, the National Science Foundation, private testing groups, and STEM professional associations or groups.
These findings are drawn from large data collection efforts, as well as smaller studies about perceptions of students, faculty, or special program staff. They provide information about URM students as they progress from high school to professional STEM careers. They explain the variables that contribute to successful completion of STEM degrees, the barriers to degree attainment, and, when understood, the reasons for attrition from higher education, as well as barriers to the professoriate. The findings are not necessarily ranked in order of importance.
Findings for the High School Years
Findings for the Undergraduate Years
Findings for the Graduate School Years
Findings for the Faculty Years
Barriers to advancement and retention of URM in post-doctoral and tenure track faculty positions include:
These current research findings provide a snapshot, albeit an incomplete one, of students, faculty, or special program staff within the STEM disciplines. The following is a summary of the factors that facilitate or limit progress of URM students within STEM. These factors present an overview of current research and can be used to identify gaps and new areas of study.
Factors that facilitate progression of URM into STEM post-secondary studies include:
Factors that limit the progression of URM students into STEM post-secondary studies and professorial positions include:
Given the limiting factors to progression of URM in STEM, there is a need to:
Also there are a number of areas limiting student progression that we need to better understand. They include:
A 1999 College Board report states that few of the numerous programs to improve academic outcomes of URM have undergone extensive external evaluation.23
The current research base provides limited potential for thorough analysis for several reasons. First, there are no established data collection guidelines for researchers who study URM in STEM. In other words, most of the data in this research base is not comparable due to differences in methods, definitions of terms (e.g., retention, URM), and other factors. Also, many of the studies on URM in STEM are sample surveys with differing degrees of validity and reliability.
The studies conducted are particularly limited in terms of disaggregated information about different racial/ethnic groups, as well as different STEM disciplines. As a result, it is hard to discern which factors are generic to all STEM students or faculty, and which are group or discipline specific.
In many studies, information on African Americans and Hispanic Americans is included; however, information on Native Americans or on gender within racial/ethnic groups is not reported. This occurs primarily because the cell size of these populations is too small to report statistically meaningful data. Further, no studies were found on minority students with disabilities in STEM.
The studies also do not examine the full spectrum of colleges and universities. For example, many of the studies are conducted at Research Extensive and Research Intensive universities, with few looking at community colleges, Historically Black Colleges and Universities, institutions serving concentrations of Hispanic Americans, tribal colleges, women's colleges, or colleges and universities that target or serve persons with disabilities.
Much of the research has focused on patterns of group differences or underrepresentation. However, very few of these studies focus on the complexity of the undergraduate or graduate educational experience of URM in STEM.24
Recommendations for Improving Methodology
In terms of methodology, the research designed to investigate and explain differences must take into account the complexity of the STEM disciplines. That is, the research should be multivariate and multi-leveled (e.g., path-analysis, structural equations). Researchers should ensure that their attempts to reduce the number of variables and interactions do not oversimplify important research questions.
The research also must be comprehensive, incorporating ecological models that include a complementary set of individual and systemic approaches. Strategies such as cohorting should be used and, ideally, information should be collected from all involved in the educational process, including students, university administrators, department faculty and staff, and STEM intervention program staff.
Specific recommendations for improving methodology include:
One of the biggest problems identified was the lack of data collection guidelines for researchers who are studying human resources in STEM. Also, more opportunities are needed for early communication and information sharing among STEM researchers. Although data are shared in a number of venues upon completion, researchers do not communicate as often during the study development, collection of data and analysis of results phases. Researchers need to communicate more effectively with each other at all stages of the research process so they can share methodology and findings.
Building a community of researchers who confer regularly may help to improve methodol-ogy and bring us closer to understanding factors that limit the STEM talent pool. To this end, recommendations for community building include:
Tables One to Four provide information on addressing the gaps in STEM research. They recommend areas of new and continuing research on the STEM experiences and achievements of URM from the high school years into the professoriate. While all research topics mentioned are important, several study group participants recommended seven specific areas as having the highest priority for funding.
These areas are:
Tables 1 through 4 follow.