Click for next page ( 2

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 1
1 The Nature of the Problem “W e are faced with a really huge problem,” said Clifton A. Poodry, director of the Division of Minority Oppor- tunities in Research (MORE) within the National Institute of General Medical Sciences (NIGMS) at the National Institutes of Health (NIH). The scientific workforce responsible for advances in knowledge needed to improve human health and well- being is not representative of the general population. African Ameri- cans, Hispanics, American Indians, and other minority groups are severely underrepresented among the scientific workforce in general and among biomedical and behavioral researchers in particular. As Orlando L. Taylor, vice provost for research, dean of the gradu- ate school, and professor of communication at Howard University, described the challenge, “research resides at the core of [U.S.] leader- ship and creativity and innovation. . . . As our demographics change radically in the country, and as the competition is changing radically, we know we cannot retain American leadership without greater participation by the full range of American people. . . . We must attract, retain, and graduate more persons from the groups that are the fastest growing groups in the country.” Greater diversity among the scientific workforce serves both abstract and immediate objectives, said Taylor. It demonstrates that the highest levels of achievement are accessible to the members of any group. It also enhances recognition of the full range of chal- lenges affecting those in the United States, since minority research- 

OCR for page 1
 UNDERSTANDING INTERVENTIONS ers bring perspectives to their work that may not be common among non-minority researchers. “It is not just that you are fulfilling the American dream by bringing people in,” said Taylor. Greater diver- sity also “enhances the talents of white and male researchers.” In his keynote address, NIH director Elias A. Zerhouni empha- sized the importance of diversity to NIH’s future. “The diversity of the scientific workforce has to be a reflection of the society within which it resides,” he said. “Otherwise you end up with science becoming a sort of elite activity of a few, rather than the social activity that underpins the strength of society.” In 2050, Zerhouni pointed out, more than half of the U.S. population will consist of minority groups that are currently underrepresented in science. “It is a strategic imperative that we succeed in making sure that we have the scientific body in 20 to 30 years that represents the vitality of our society.” Today’s scientific workforce is a very long way from reflecting the makeup of the broader society. The representation of minorities within the pipeline leading to the PhD and to research careers drops at each successive educational level. In 2004, African Americans, who constitute about 13 percent of the U.S. population, received approximately 9 percent of the baccalaureate degrees at U.S. col- leges and universities but less than 4 percent of the PhDs awarded to U.S. citizens. Hispanics, who constitute more than 14 percent of the U.S. population, received less than 7 percent of the bachelor’s degrees and only a little more than 3 percent of the PhD degrees. Even Asian Americans, though overrepresented in some fields, are underrepresented in others. In short, said Taylor, “there is quite a pool of people we are losing [from the pipeline]. We could double doctorate production by getting a good plumber.” Furthermore, even when underrepresented minorities earn PhDs, they appear to be less likely than white doctorate recipients to conduct research at elite research universities. In the 50 biology departments that have recently received the most federal funding, for example, the percentage of faculty members who are underrep- resented minorities is less than the percentage of underrepresented   Baccalaureate data: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics. 2005. Digest of Education Statistics: 2005. NCES 0006-030. June 2006. Doctorate data: T.B. Hoffer, V. Welch, Jr., K. Webber, K. Wil- liams, B. Lisek, M. Hess, D. Loew, and I. Guzman-Barron. 2006. Doctorate Recipients from United States Universities: Summary Report 2005. Chicago: National Opinion Re- search Center. (The report gives the results of data collected in the Survey of Earned Doctorates, conducted for six federal agencies—NSF, NIH, USED, NEH, USDA, and NASA—by NORC.)

OCR for page 1
THE NATURE OF THE PROBLEM  minorities who received biology PhDs from 1983 to 1999, the pool from which most of the faculty is drawn. As described below, fac- ulty members from underrepresented groups may be more likely to “give back” to their communities and choose to work at a wider range of institutions. Yet as small as these numbers are, a strong argument can be made that they would be even smaller without programs designed to encourage minorities to pursue advanced degrees in the biologi- cal, biomedical, and behavioral sciences. While the number of bio- logical sciences doctorates awarded to white U.S. citizens and per- manent residents stayed about the same from 1995 to 2005, growing from 3,115 to 3,337, the number of doctorates awarded to blacks rose from 107 to 158 (2.5 to 3.6 percent of the total), and the number to Hispanics from 127 to 227 (2.9 to 5.2 percent). As David R. Burgess, professor of biology at Boston College, told the workshop partici- pants, “a lot of the intervention programs you have participated in and directed and led have been very successful.” Zerhouni agreed: “There is a sense that nothing works,” he said. “I would submit to you that this is like saying we would have obtained the same results if those programs had not existed, and I disagree. I really believe that these programs have, in fact, facilitated the careers of many scien- tists who would not be successful today without these programs.” But there is a need to move beyond belief and anecdote, to con- duct rigorous research that will identify the key elements that lead to successful programs. The real question, according to Zerhouni, is how to optimize the nation’s investments in educational interven- tions. What interventions will have the greatest effect, and how much will those investments cost? Both past and current programs have incorporated a wide vari- ety of strategies. Some have been designed to remediate under- preparation. Others have sought to build the skills needed for suc- cess in research. Some have focused on building supportive learning environments. Many have provided financial assistance, and many have used research experiences as a way to achieve more than one of these objectives. The obvious question to ask of these programs is “what works?” But that simple question is not very meaningful, according to Poodry,   Donna J. Nelson. 2004. “Nelson Diversity Surveys.” Norman, OK: Diversity in Sci- ence Association.   National Science Foundation, Division of Science Resources Statistics. 2006. S&E Degrees, By Race/Ethnicity of Recipients: 1995–2004. NSF 07-308. January 2007. Susan T. Hill and Maurya M. Green, project officers. Arlington. VA.

OCR for page 1
 UNDERSTANDING INTERVENTIONS “because the problem is a complex, multidimensional problem. What works to do what?” In seeking to approach these issues, Poodry warned against what he called the “n of one”—each individual’s preconceptions and understandings based on his or her own personal experiences. These experiences inevitably shape how a person sees the world. “If your conclusions concur with what people already believe, you are a genius,” said Poodry. “If you contradict their n of one, you’re a bum. I’ll give you an example”: Many faculty [members] and administrators tell me that minority science students, especially the financially disadvantaged, are far more attracted to professions—the MD—than to research careers, because of the potential earning power. What they tell me is that poor people, having been poor, are more concerned about making the most money. From my n of one, which includes growing up extremely poor on an Indian reservation, that is pure hogwash. I never knew a single poor person from a minority community who actually felt that way. Getting a good job as a nurse, a teacher, maybe an engineer, those were stretch goals. But the notion that we are going to do this to make the most money just wasn’t part of my experience. Overturning these “n of one” biases requires “some fantastic data and thoughtful analyses,” according to Poodry. Asking interesting and answerable questions, designing persuasive research studies, and interpreting data appropriately will be necessary for convincing audiences that personal experiences may not always be correct. In many ways, the existing programs have been “borne out of what we intuitively think,” according to DePass. “We think if we put our students in a lab in the summer, it will make a scientist out of them—or it is going to move them to science. Have we studied exactly ‘does it really work? Is that the best way to do it?’ What envi- ronment, what other factors does one include with that to enhance success?” Part of the problem in analyzing the development of the sci- entific workforce is the complex network of paths that people can follow to a research career, according to Jeremy M. Berg, the direc- tor of NIGMS. Prospective researchers hail from different kinds of undergraduate institutions and have had a wide range of experi- ences before college. They may become researchers through PhD programs, MD programs, or combined PhD-MD programs. More- over, only some of those who earn PhDs in the biological sciences pursue research careers.

OCR for page 1
THE NATURE OF THE PROBLEM  To fully understand how this network functions, three funda- mental questions need to be answered, Berg said: • First, what are the probabilities that an individual with a given set of characteristics will make the transition from one stage to another? Many kinds of transitions are possible, because of the multiplicity of paths people can follow to a career in research. And the probabilities differ for people of different races, ethnicities, and gender as they move along these paths. • Second, why do people with different characteristics make the decisions that they do? Though many hypotheses seek to answer this question (as the next section of this report points out), only research can be expected to produce solid answers. • Third, how can these probabilities be changed? Answering this question requires “a combination of understanding the ‘why’ and a very pragmatic understanding of what really works empiri- cally,” said Berg. “What sorts of interventions really have an impact? Are there interventions, say, at the college level, which not only influence the probability of going on to graduate school but also persist in encouraging increasing probabilities for an individual stu- dent going on to a particular path? . . . We need to understand much better what works. It is not just a question of program evaluation, but really understanding what interventions have real impact, how long it lasts, and so on.” A major challenge in answering these questions involves time- scales, said Berg. Once an intervention is implemented, it takes a long time to gauge the effects of that intervention on the composition of university faculty. As a result, intermediate measures are needed to assess the program’s effectiveness. “Five or six years along the way, even though you have not seen the impact at the end, you [need to] have some sense of what benchmarks to expect,” said Berg. Having such benchmarks makes it possible to “monitor progress so that you are as confident as you can be that the program is likely to have the sort of impact that you intended in the first place.” Another major challenge is how to define success. Is a PhD stu- dent who goes to medical school a failure? “How can we be clear as to what our expectations are but also be realistic?” asked Berg. “If you say a hundred percent of your students should go on to PhDs, that is going to distort the program in such a way that you may end up hurting the overall outcomes.” A final challenge is determining the elements of a program that are working and those that are not. Researchers have to try to under-

OCR for page 1
 UNDERSTANDING INTERVENTIONS stand “what aspects of the program are the major and important contributors to that sort of success,” according to Berg. This chal- lenge is especially difficult, noted Zerhouni, given that social science research needs to deal with systems, and systems are often difficult to simplify in such a way that the effects of one part of the system can be isolated from other parts. “We cannot use the simple anal- ogy of ‘let’s reduce the problem,’ because by reducing it to certain parameters that everybody agrees to, you are losing, in fact, the essence of what the issue can be,” Zerhouni said. Today, social science researchers have not answered the three questions Berg specified. Even knowing what questions to ask and how to answer those questions can be challenging. “I am not so sure that I am hearing cogent analyses [or] powerful arguments about what are the real drivers,” Zerhouni said. People have good will and want to have a fair and just representation of population groups involved in science, he continued, but the numbers are not changing as fast as many wish they would. “So the charge [to the workshop] is simple,” said Poodry. “What do we need to know in order to define effective interventions? What are the important questions? And what are the researchable ques- tions? There are a lot of questions, but which ones can actually make progress with research? What kinds of research and what kinds of methodologies are needed to guide and test promising new inter- ventions? . . . Today we want your help in focusing the questions that should be asked and your guidance as to the appropriate meth- ods to answer them.”