Best & Brightest: Education and Career Paths of Top Science and Engineering Students

Executive Summary

The strength of the U.S. scientific and technical enterprise depends in part on the career choices of academically talented students. Who will educate the scientists and engineers of the next century? Will the quality and quantity of students already in the science and engineering (S&E) educational pipeline be adequate for training future scientists and engineers?

Ten years ago, there was major concern about a possible "brain drain" from science into other professions. Since that time, significant changes have taken place, for example, in the labor markets for scientists and engineers.

Therefore, this study, performed by the Commission on Professionals in Science and Technology (CPST) for the Alfred P. Sloan Foundation, looks at the questions of what fields promising students are choosing and why, and specifically whether science and engineering are attracting a larger or smaller share of these promising students than in the past. Data were analyzed all along the educational pipeline from the choices of high school graduates, through college and graduate/professional school, and out into the labor force. In addition, a workshop of experts and interested parties was held to gather additional data and perspectives.

Study Objectives

The study focuses on the following five questions:

Is there a change in the quality of talent flow into S&E undergraduate programs? Are these programs attracting a larger or smaller share of top students now than in the past?
Do top students persist in science majors or switch to other majors?
Is there a change in the quality of talent flow into S&E graduate programs?
Are top students still interested in doctoral programs?
After graduation, do the top students in science fields persist in science or pursue other employment?

Findings

Our general findings related to the study questions are as follows:

1. Is there a change in the quality of talent flow into S&E undergraduate programs? Are these programs attracting a larger or smaller share of top students now than in the past?

The quality of incoming freshmen is higher.

Of the available pool of freshmen, top students made up a larger percentage in 1995 than in 1985. Thus, of the students choosing S&E, more of them are top students, indicating that the quality is higher.

On the other hand, the S&E fields’ share of the talent pool of top students has dropped in engineering and math. Biological sciences’ share has increased, while the physical sciences’ share has remained the same.

The number of National Merit Scholars choosing S&E majors is consistently higher than other majors.

2. Do top students persist in science majors or switch to other majors?

Many top freshmen do not stay in S&E.

There are some losses in all S&E fields with large losses in biological sciences and mathematical sciences. Recruitment of other top students does not fully compensate for these losses. In addition, top women students’ persistence rates are lower than men’s in all fields.

3. Is there a change in the quality of talent flow into S&E graduate programs?

The GRE quantitative scores are decreasing slightly, but top S&E majors still score consistently higher than other fields.

4. Are top students still interested in doctoral programs?

Almost one-third of top S&E students pursue doctorates.

The majority of S&E majors go to graduate school including 67% of top biological science students. Nine years later, almost one-third of top S&E majors in the 1985 freshman cohort had doctoral degrees, doctoral plans, or medical degrees.

5. After graduation, do the top students in science fields persist in science or pursue other employment?

Nine years after freshman year, employment was highest among top engineers and lowest among top biological science majors, who were more likely to be in graduate school. Engineers pursued engineering careers, while there was a great deal more changing to different fields among the other three majors. However, the potential pool of top talent who will eventually obtain their doctoral degrees and enter the academic or scientific workforce may be underestimated at this early point.

Related to Questions 4 and 5 is our finding regarding the flow of S&E talent into the professional schools, which resulted from the data presented at a workshop held during the study:

Except for many biology majors going into medicine, S&E are not losing top students to professional schools.

S&E majors are not choosing professional schools over graduate school and doctoral programs with the expected exception of biological science majors, the majority of whom pursue medical degrees.

Finally, our campus focus groups provided qualitative data, which resulted in the following finding:

S&E majors have a passion for the subject.

In general, the job market has had little or no effect on S&E majors’ decisions to pursue or persist in these majors through graduate school. These students all expressed a passion for the subject matter.

However, is there still cause for concern? The answer is yes in light of the importance of S&E in producing a citizenry that is scientifically literate and globally competitive, and the importance of research. Consider these specific findings:

A number of concerns still need to be addressed.

	The numbers of women and minorities entering S&E are increasing, but only slightly.
	Biological science and mathematics departments are doing poorly at keeping top undergraduate students, who persist at rates of 33% and 24%, respectively.
	S&E departments are not retaining top women undergraduates. The proportion of top women who persisted in their S&E majors was lower than men in all fields.
	In engineering, only 29% of top undergraduate women persist, while 82% of top undergraduate men do.
	Considering losses and gains during 1985-89, the number of top students majoring in mathematics dropped by one-half, and the total number of mathematics majors dropped by one-third.

In light of these results, much more needs to be learned, and there is much need for improvement, particularly in the area of nurturing our top undergraduates. Recommendations are presented in the next section.

Recommendations

The following are recommended:

	Longitudinal studies covering longer periods of time after college entry than currently available are needed to determine fully the extent of talent flow in academic and scientific sectors.
	At the same time, cross-sectional studies showing peaks and valleys in, for example, professional school admissions, should recognize external factors (such as popularity brought on by hit television programs and other cultural phenomena).
	Individual institutions should consider conducting focus groups and case studies to look at their retention of top talent and share lessons learned with other institutions.

Are U.S. citizens being displaced in graduate schools?

	The data used in this study did not enable the authors to contrast numbers or proportions of U.S. versus foreign citizens. At the undergraduate level, 97% of incoming freshmen are U.S. citizens (CIRP, 1996). However, at the graduate level, of the S&E doctorates (Ph.D.’s) awarded in 1995, only 53% were awarded to U.S. citizens (National Science Foundation, 1996). Therefore, it is important to determine the effect of the increasing numbers of foreign students entering S&E programs and the professional schools. Specifically, should displacement of U.S. citizens be of any concern?
	In addition to the citizenship composition of the S&E talent pool, the balance among other pools of talent based on sex and race/ethnicity should be monitored to ensure proportions that are appropriate, given considerations such as representation in the population at large.
	Data limitations are mentioned several times in this report. These limitations, that is, small cell sizes, were particularly evident in our attempt to analyze trends by gender and race. We recommend that larger numbers of students be surveyed or that the study be designed differently to ensure sufficient data from which to discern a pattern.

In response to some of the specific findings:

Persistence issues need to be examined closely.

	Undergraduate biological science and mathematics departments must examine their programs to determine why the persistence of top students is so low, and what can be done to encourage top undergraduates to stay in their original field choice.
	Analysis of the factors causing top students to switch into an S&E discipline may produce insights into how to retain those who chose S&E at the outset.
	The persistence rates of top women in all the S&E fields, but particularly engineering, were low compared to top men. These disciplines need to examine their student advising programs, course offerings, and other factors to determine what can be done to improve on these numbers. Top women want to major in these fields, but something is keeping them from staying.

The number of minorities in S&E majors is still low.

	The numbers of minorities in some scientific disciplines are quite small. Similar to top women, those top minority students who do choose to major in S&E should be encouraged to stay.
	As expected, the potential for a rewarding career affects top students’ choices, particularly at the graduate level. It was suggested at the December 1996 workshop that a graduate, faculty, and post-doctorate placement service be developed. The assistance in career placement currently given at this level needs to be examined as an avenue to improving the job market for S&E majors.

In conclusion, it is difficult to simply step in at one point in time and evaluate the status of top students and S&E. No matter the study field, to provide an accurate portrait of what fields are attracting and retaining students through career selection, we recommend periodic monitoring of key indicators such as those described in this study. Only by trending these data over time can we assess the educational and career paths of our nation’s best and brightest students.

1. Introduction

This report describes a study performed by the Commission on Professionals in Science and Technology (CPST) for the Alfred P. Sloan Foundation to assess whether top students are choosing science and engineering as an academic field and career path. This introductory chapter provides the background of the study, presents the study objectives, and describes the organization of the remainder of the report.

Background

The strength of the U.S. scientific and technical enterprise depends in part on the career choices of academically talented students. Who will educate the scientists and engineers of the next century? Will the quality and quantity of students already in the science and engineering (S&E) educational pipeline be adequate for training future scientists and engineers?
Ten years ago, there was major concern about a possible "brain drain" from science into other professions. The National Academy of Sciences commissioned Engin Holmstrom to examine the study field and career choices of top students interested in science and engineering. Her findings, using data through 1985, gave no indication of a brain drain, nor did they show any nationwide decline in the quality of science students (Holmstrom, 1987).

Since that time, significant changes have taken place, for example, in the labor markets for scientists and engineers. Therefore, it is useful to revisit the questions of what fields promising students are choosing and why, and, specifically, to determine whether science and engineering are attracting a larger or smaller share of these promising students than in the past.

These and related questions are explored in this study. Data were analyzed all along the educational pipeline from the choices of high school graduates, through college and graduate/professional school, and out into the labor force.

Study Objectives

The study focuses on the following five questions:

Is there a change in the quality of talent flow into S&E undergraduate programs? Are these programs attracting a larger or smaller share of top students now than in the past?

Do top students persist in science majors or switch to other majors?

Is there a change in the quality of talent flow into S&E graduate programs?

Are top students still interested in doctoral programs?

After graduation, do the top students in science fields persist in science or pursue other employment?

Report Organization

The remainder of this report is organized as follows:

Chapter 2 describes the study methodology including definitions, data sources, and the types of analyses performed.

Chapter 3 presents the study findings, organized by the five questions presented above.

Chapter 4 discusses the findings and their implications and presents recommendations.

The appendix contains the results of the study’s literature review.

2. Methodology

This study focused on the most academically talented students, analyzing the most recent data available on the flow of top talent into and out of science and engineering fields at different points along the educational pipeline.
The study methodology steps may be summarized as follows:

Determine data sources.

Set definitions.

Conduct initial analyses.

Conduct campus focus groups.

Conduct workshop.

Conduct additional analyses based on workshop feedback.

The sections in this chapter describe each of these steps.

Data Sources and Points of Inquiry

An initial literature review was conducted to determine whether any similar studies have been performed. The results of this review are presented in the appendix. It was concluded from the review that rarely are academically talented students examined as a group, and that there have been no recent studies on the college and career paths of S&E top students.
After review of potential data sources, the following were selected for use:

Cooperative Institutional Research Program (CIRP) of the Higher Education Research Institute at the University of California, Los Angeles (UCLA)

National Merit Scholarship Corporation

Educational Testing Service’s Graduate Record Examination Program

CIRP

CIRP has collected data on entering freshmen since 1966. The following CIRP data were used:

Freshman surveys of 1985, 1990, and 1995 classes (cross-sectional data)

Follow-up surveys of the 1985 freshman class, taken in 1989 and again in 1994 (longitudinal data)

All the CIRP data are weighted by CIRP to represent the national population of college students (Astin, Panos, & Creager, 1966; Astin & Astin, 1992). The number of freshman surveys used varies each year, but as an example, the 1995 data are based on 240,082 surveys. The 1994 follow-up study includes more than 50,000 students who responded to the 1985 freshman survey, including 27,000 who responded to the 1989 follow-up survey.

National Merit Scholarship Corporation

Data on National Merit Scholars from 1982 through 1994 were obtained from the National Merit Scholarship Corporation (NMSC). This information was used to determine the intended college majors of these students.

Educational Testing Service

The Educational Testing Service furnished Graduate Record Examination (GRE) results from selected testing years (1988-89, 1991-92, and 1994-95). These data provided insights into the scores of students intending to pursue S&E fields at the graduate level versus the scores in other fields.

Points of Inquiry

As mentioned earlier, the study examined top S&E students at various points in a top S&E student’s academic program and career. Figure 1 shows how the data sources just described relate to this timeline.

Figure 1. Points of Inquiry and Data Sources

The data provided by CIRP are of two types: cross-sectional and longitudinal. Figure 2 shows this concept. The cross-sectional data are "snapshots" of the freshman classes of 1985, 1990, and 1995. The longitudinal data consistently analyze the same 1985 freshman cohort four years later (1989) and nine years later (1994).

Figure 2. Cross-Sectional Data Versus Longitudinal Data

Definitions

The following definitions were established as part of the study design once the data sources were determined:

"Top students." Academically talented students are termed "top" in this study. These students were defined as consistently as possible throughout the study.

- At college entry, top students in the freshman class were defined as those with high school grade point averages (GPAs) of A-, A, or A+.

- At the end of four years of college, "top" was defined as A or A+.

- For the nine-year follow-up of 1985 freshmen, "top" was expanded (due to data limitations) to include B, B+, A-, A, and A+.

"Science and Engineering." These fields were defined as follows:

- Engineering. Aeronautical or astronautical, chemical, civil, electrical or electronic, industrial, mechanical, and "other"

- Biological Sciences. General biology, biochemistry or biophysics, botany, environmental science, marine (life) science, microbiology or bacteriology, zoology, and "other"

- Mathematical Sciences. Mathematics and statistics

- Physical Sciences. Astronomy, atmospheric sciences (including meteorology), chemistry, earth science, marine science (including oceanography), physics, and "other"

These field definitions correspond with the categories used by CIRP.

Initial Analyses

The quality of talent flow into S&E was evaluated at high school graduation, college entry, and graduate school entry. The popularity of S&E among top students, that is, S&E’s share, was also traced. Then, the study field and graduate degree choices of top students were traced through four years of college. Finally, graduate school attendance, doctoral degree rates, and employment status of top S&E students were examined nine years after college entry.

Campus Focus Groups

Focus groups were conducted on university campuses with the following groups of students:

	Undergraduate physics majors
	Undergraduate biochemistry students
	Graduate biology students

The purpose of these focus groups was to collect qualitative information on the students’ knowledge of the current job market in S&E and its influence on their educational and career planning and decision making.

The following questions were posed to the groups:

	In what year are you, and what is your major or field?
	What are your immediate plans after graduation?
	What led you to pursue your major or field?
	Has your knowledge of the job market affected your educational and/or career planning?
	(For undergraduates only) What do you plan to do after all of your education is completed?

Workshop

The results of the initial analyses were presented at a workshop held by CPST on December 12, 1996, at the headquarters of the American Association for the Advancement of Science (AAAS). Experts and interested parties from education, government, business, industry, and professional societies attended and contributed to this workshop.

Following the presentation of the initial results, perspectives were provided from professional schools:

	Carl Monk, executive director of the Association of American Law Schools, discussed trends in law school admissions.
	Douglas Kelly, Associate Vice President for Biomedical Research of the Association of American Medical Colleges, presented data on the trends in medical school admissions and the medical degrees awarded.
	Frederic McHale, Director, Graduate Management Admission Test (GMAT), Educational Testing Service, discussed trends in the numbers and types of students taking the GMAT, which is generally a requirement for admission to graduate business schools.

Breakout groups were also held on two topics:

	Data and data needs
	Policy issues and recommendations

Jesse Ausubel of the Alfred P. Sloan Foundation closed the workshop with recommendations for further action.

Key findings and recommendations resulting from the workshop are presented in the remaining chapters of this report.

Additional Analyses

A recommendation from the workshop participants was that the CIRP data be reanalyzed by gender and race. CPST requested further data from CIRP that provided further breakdowns by race and by gender. Due to extremely small cell sizes in the race subcategories, it was not possible to perform analyses by racial subgroups such as African-Americans, Asians, and Hispanics. Therefore, the following race definitions were established for this series of analyses:

	White includes anyone who marked "white" on the survey.
	Nonwhite includes all other race categories (African-Americans, Asians, Hispanics, and others)

The majority of the initial analyses of the CIRP data were repeated, broken down by gender and by race. For example, changes in the gender composition of the S&E top talent pool were examined, as was the proportion of nonwhite students among top S&E majors.

Certain analyses were limited by the nature of the data. For example, a high number of respondents in the CIRP nine-year follow-up survey did not indicate race, resulting in questionable results for the career paths of top nonwhite men and women in the S&E fields.

The findings resulting from all the analyses described here are presented in Chapter 3.

3. Findings

This chapter presents our findings from the analyses described in the previous chapter. Supporting data are presented to illustrate these findings.

General Findings

	Our general findings related to the study questions are as follows: 1. Is there a change in the quality of talent flow into S&E undergraduate programs? Are these programs attracting a larger or smaller share of top students now than in the past?
The quality of incoming freshmen is higher.	Of the available pool of freshmen, top students made up a larger percentage in 1995 than in 1985. Thus, of the students choosing S&E, more of them are top students, indicating that the quality is higher. On the other hand, the S&E fields’ share of the talent pool of top students has dropped in engineering and math. Biological sciences’ share has increased, while the physical sciences’ share has remained the same. The number of National Merit Scholars choosing S&E majors is consistently higher than other majors. 2. Do top students persist in science majors or switch to other majors?
Many top freshmen do not stay in S&E.	There are some losses in all S&E fields with large losses in biological sciences and mathematical sciences. Recruitment of other top students does not fully compensate for these losses. In addition, top women students’ persistence rates are lower than men’s in all fields. 3. Is there a change in the quality of talent flow into S&E graduate programs? The GRE quantitative scores are decreasing slightly, but top S&E majors still score consistently higher than other fields. 4. Are top students still interested in doctoral programs?
Almost one-third of top S&E students pursue doctorates.	The majority of S&E majors go to graduate school including 67% of top biological science students. Nine years later, almost one-third of top S&E majors in the 1985 freshman cohort had doctoral degrees, doctoral plans, or medical degrees. 5. After graduation, do the top students in science fields persist in science or pursue other employment? Nine years after freshman year, employment was highest among top engineers and lowest among top biological science majors, who were more likely to be in graduate school. Engineers pursued engineering careers, while there was a great deal more changing to different fields among the other three majors. Related to Questions 4 and 5 are our findings regarding the flow of S&E talent into the professional schools. Our review of the data presented at the December 1996 workshop resulted in the following general finding:
Except for many biology majors going into medicine, S&E are not losing top students to professional schools.	S&E majors are not choosing professional schools over graduate school and doctoral programs with the expected exception of biological science majors, the majority of whom pursue medical degrees.
	Finally, our campus focus groups provided qualitative data, which resulted in the following finding:
S&E majors have a passion for the subject.	In general, the job market has had little or no effect on S&E majors’ decisions to pursue or persist in these majors through graduate school. These students all expressed a passion for the subject matter. The following sections provide detailed findings related to these statements.

The Talent Pool for Undergraduate Study

Finding

Of the available pool of freshmen, top students made up a larger percentage in 1995 than in 1985. Thus, of the students choosing S&E, more of them are top students, indicating that the quality is higher.

On the other hand, the S&E fields’ share of the talent pool of top students has dropped in engineering and math. Biological sciences’ share has increased, while the physical sciences’ share has remained the same.

The number of National Merit Scholars choosing S&E majors is consistently higher than other majors.

Top and Total Freshmen, 1985 Versus 1995

As shown in Figure 3, between 1985 and 1995, the number of freshmen entering the nation’s colleges and universities declined by about 11%, while the number of top students, that is, those with high school GPAs of A-, A, or A+, increased by 17%.

To some degree, S&E fields reflected these national trends from 1985 to 1995:

With the exception of biological sciences, the popularity of S&E has declined among students in general.

	Engineering. Freshmen choosing engineering declined by 45%. Top talent flow into engineering also slowed with a decline of 33%.
	Biological Sciences. In contrast to other S&E fields, the popularity of biological sciences went up, with the freshmen choosing biological sciences increasing by 70% and the number of top students interested in biological sciences increasing by 96%.
	Mathematical Sciences. Freshmen interested in mathematical sciences declined by 32%. Similarly, the number of top students declined by 23%.
	Physical Sciences. Freshmen interested in physical sciences declined by 8%, while the number of top students increased by 6%.

Although the pool of freshmen interested in majoring in S&E programs was smaller in 1995 than in 1985, more academically able students were among the ranks of these S&E majors. In each case, there were proportionately more top students among S&E aspirants in 1995 than in 1985, as shown in Figure 4.

To specifically respond to one of the study objectives, S&E’s share of top students was analyzed. The total S&E share of top students in 1995 was 24%, down from 29% in 1985. In the specific majors, the share of top students was as follows:

S&E’s total share of top students declined from 1985 to 1995.

	Engineering’s share of top students declined from 18% to 10%.
	Biological sciences’ share of top students rose from 6% to 10%.
	Mathematical sciences’ share declined from 2% to 1%.
	Physical sciences’ share was stable at 3%.

Thus, the loss in share of top students in S&E overall appears largely due to the drop in share in engineering.

The following sections present our findings regarding the gender and racial breakdown of top students choosing S&E.

Gender Breakdown of Top S&E Majors

			As shown in Table 1, the analyses of top freshmen by gender indicate that engineering’s popularity has declined for both genders, especially men; biological sciences’ popularity has increased; the mathematical sciences’ share has declined, and physical sciences have remained relatively stable.
			Table 1. Percent of Top Men and Women Freshmen Interested in S&E (1985 and 1995)
	Men				Women
Major		1985		1995		1985	1995
Engineering		32%		20%		7%	4%
Biological Sciences		6		10		6	11
Mathematical Sciences		2		1		2	1
Physical Sciences		5		4		2	2

Numerically, most S&E fields lost top students from 1985 to 1995, as shown in Table 2. The only exceptions were biological sciences, where the number of both top men and women increased, and physical sciences, where only the number of top women increased.

Although the number of top students increased, engineering and math lost top students of both genders.		The losses of top students interested in engineering and mathematical sciences, as well as top men interested in physical sciences, are even more significant in view of the fact that from 1985 to 1995 the number of top students among both men and women increased by 8% and 24%, respectively. Table 2. Percent Change in Number of Top Students Interested in S&E Among Men and Women Freshmen from 1985 to 1995
Major	Men		Women
Engineering	-33%		-34%
Biological Sciences	74		112
Mathematical Sciences	-20		-26
Physical Sciences	-11		33
Total Freshmen	8		24

Top women students consistently had better grades than men at the time of college entry.		One trend remained stable from 1985 to 1995: At the time of college entry, as a group, women choosing to major in S&E had better grades than men. For example, as shown in Table 3, top women made up the majority of female freshmen majoring in engineering (52% in 1985 and 62% in 1995). The difference between men and women is significant compared to the proportion of top students in the total freshman classes in both years. Table 3. Percent of Top Students Among Men and Women Freshmen Interested in S&E Majors in 1985 and 1995
	1985			1995
Major	Men		Women	Men	Women
Engineering	29%		52%	35%	62%
Biological Sciences	31		42	38	46
Mathematical Sciences	44		55	48	65
Physical Sciences	39		49	46	51
Total	18		24	24	32

	Clearly, students are aware that better academic credentials are needed to enter S&E majors. Table 3 shows that in each year, the proportion of top students stating their preference for S&E majors is higher than the proportion of top students in the total freshman class.
Three out of five top students are women.	The data were also analyzed for changes in gender composition of the top students choosing S&E majors. As shown in Figure 5, women make up roughly 60% of all top students in 1985 and 1995. Their proportion among top students interested in biological and mathematical sciences remained more or less equal to their proportion in the total freshman classes in both years. In other words about three in five top students declaring their intentions to study biological sciences and mathematical sciences are women.
Engineering remains a male-dominated field.	On the other hand, engineering remains a male study field. Fully three-fourths of all top students interested in engineering at the time of college entry are men. The proportion of women among top students interested in engineering remains less than one-half of the proportion of women among top students in the 1985 and 1995 freshman classes. In physical sciences, the composition of top students has changed over the past ten years, with more women indicating an interest in physical sciences: 47% of top women students in 1995 versus 37% in 1985.

Figure 5. Percentage of Women Among Top Freshmen Interested in S&E (1985 and 1995)

Racial Breakdown of Top S&E Majors

For this study, the categories of white and nonwhite were used. The nonwhite classification refers to all minorities including Asians. Separate analyses by race (such as African-American, Hispanic, or Asian) were attempted, but no meaningful results could be determined from the small cell sizes.

As shown in Figure 6, there have been slight increases of minority students entering S&E fields.

Figure 6. Proportion of Nonwhites Among Top Students by Gender (1985 and 1995)

Among top men students, the proportion of nonwhites has increased slightly among freshmen interested in engineering (up from 12% in 1985 to 18% in 1995) and in biological sciences (up from 17% in 1985 to 21% in 1995) but has remained about the same in mathematical and physical sciences.

Although the percentages of nonwhites show an increase, the actual numbers are very small.	Among top women students, all but physical sciences have registered slight increases in the proportion of nonwhites. The largest increase was in mathematical sciences from 6% in 1985 to 15% in 1995, but as seen in Table 4, the actual increase was from 247 to 432 students.
	Table 4. Top Freshmen’s Choices in 1985 and 1995 by Gender and Race
Top Students		1985	1995	% Change
Engineering
Men White		40,728	25,656	-37
Nonwhite		5,632	5,581	-1
Women White		12,161	7,526	-38
Nonwhite		2,754	2,296	-17
Biological Sciences
Men White		7,189	11,934	66
Nonwhite		1,437	3,108	116
Women White		10,787	21,625	100
Nonwhite		1,799	5,050	181
Mathematics
Men White		2,490	1,998	-20
Nonwhite		274	222	-19
Women White		3,607	2,427	-33
Nonwhite		247	432	75
Physical Sciences
Men White		6,552	5,771	-12
Nonwhite		872	866	-7
Women White		3,605	4,903	36
Nonwhite		749	899	20
Total Top Students
Men White		129,107	134,040	4
Nonwhite		16,102	22,974	43
Women White		181,694	212,776	17
Nonwhite		20,255	37,649	86

Table 4 also shows that increases in top nonwhites are not being accompanied by decreases in top white students in S&E fields. In fact, some fields show a decrease in top students in virtually all gender and race categories.

	Engineering. It was shown earlier in Table 1 that the percentage of top men interested in engineering had declined from 32% among 1985 freshmen to 20% in 1995. The number of top white men declined by 37%, from 40,728 in 1985 to 25,656 in 1995 (Table 4). The number of top nonwhite men declined by only 1% (from 5,632 to 5,581). The number of top white women declined by 38%, and nonwhites by 17%. Compared to the total freshman class where numbers of top students increased in all categories, the popularity of engineering among top students is down, but less so among nonwhite men. Nevertheless, the field remains dominated by white men.
	Biological Sciences. The popularity of biological sciences has increased, particularly among top nonwhite students.
	Mathematical Sciences. The popularity of math has decreased except among top nonwhite women.
	Physical Sciences. The popularity of physical sciences has decreased among top men but increased among top white and nonwhite women.

It appears that top nonwhite women compensate to some degree for the losses among top students in math, while top women in general compensate for losses in physical sciences.

National Merit Scholars

National Merit Scholars continue to prefer S&E fields.

Since National Merit Scholars represent the top 0.5% of the nation’s high school graduates in terms of academic achievement, they provide a good indicator of the field preferences of top students. Over the years, these top high school graduates have consistently shown a stronger preference for S&E fields as a college major than for other programs. This trend continues as shown in Figure 7. In 1994, S&E fields still attracted larger numbers of National Merit Scholars than did business and health-related fields.

Figure 7. Number of National Merit Scholars
Choosing S&E Majors Versus Other Fields
(1982 through 1994)

Changes in Major Among Undergraduates

Finding

There are some losses in all S&E fields with large losses in biological sciences and mathematical sciences. Recruitment of other top students does not fully compensate for these losses. In addition, top women students’ persistence rates are lower than men’s in all fields.

Changes in Major Among All Top Students

		During four or more years of undergraduate education, students often change their initial study field plans. By 1989, all the S&E fields experienced losses of the 1985 freshmen who had intended to major in S&E. At the same time, these fields differed in their ability to retain top talent (those with college GPAs of A or A+). As shown in Table 5, the largest outflow of talent was from the mathematical sciences, where the number of top students decreased by 38%, followed by engineering (-13%) and biological sciences (-7%). In contrast, the number of top students majoring in physical sciences increased by 64%. Table 5. Four-Year Change in Major Plans of the 1985 Freshman Cohort
Major	A, A+		Total in Major
Engineering
1985 Aspirants	9,563		158,877
1989 Majors	8,322		123,766
% Change	-13		-22
Biological Sciences
1985 Aspirants	8,908		57,653
1989 Majors	8,260		63,652
% Change	-7		10
Mathematical Sciences
1985 Aspirants	2,459		16,506
1989 Majors	1,516		12,429
% Change	-38		-24
Physical Sciences
1985 Aspirants	2,260		28,255
1989 Majors	3,701		25,981
% Change	64		-8

Figure 8 shows the persistence rates among top students in their S&E majors, that is, the percentage who remained in their freshman plans. Persistence was highest among physical science majors (78%), followed by engineers (69%), biological science majors (33%), and mathematical science majors (24%).

Figure 8. Percent of Top Freshmen Who Persisted
in Their S&E Major

The inflow of top talent (those with college GPAs of A or A+) also differed among the fields. In 1989, 20% of engineering top students, 65% of top biological science students, 61% of top mathematical science students, and 52% of physical science top students had been recruited into the field during the four undergraduate years
(Figure 9).

		Figure 9. Percent Top Students Recruited by S&E Major
		Table 6 shows the net result of the losses of top talent and the gains of new top talent between 1985 and 1989.
		Table 6. Outflow and Inflow of Talent
Major	Changers		Recruits	Net Result
Engineering
A, A+	2,942		1,701	-1,241
Total	71,687		36,575	-35,112
Biological Sciences
A, A+	6,002		5,352	-650
Total	31,245		37,190	5,945
Mathematical Sciences
A, A+	1,868		924	-944
Total	13,254		9,177	-4,077
Physical Sciences
A, A+	485		1,926	1,441
Total	17,496		15,226	-2,270

	With the exception of biological sciences, all S&E fields lost students during four years of college. With regard to top students, however, biological sciences, mathematical sciences, and engineering all experienced losses during four years of college; only physical sciences gained top students. Top talent exchange was heaviest in the biological science programs, which lost two-thirds of their top students during four years of college but compensated for most of the losses by recruiting large numbers of top students from other programs.
The increase in top students in physical sciences reflects their ability to retain and recruit top students.	As stated above, physical science programs were unique in their ability to keep their top talent while at the same time managing to recruit top students from other programs. As a result, the number of top physical science majors increased by 64% after four years of college. Engineering programs also were successful, although to a lesser degree than physical science programs, in keeping their top students.
Mathematical sciences suffered a net loss of 40% of their top students.	In contrast, there was heavy traffic into and out of mathematical science programs. Fully three-fourths of top freshmen interested in mathematical sciences in 1985 had switched to other programs by 1989. After four years of college, 61% of top students majoring in mathematical sciences were recruits from other fields. But in the process, mathematical science programs lost nearly two-fifths of their top students. In total, S&E majors lost nearly one-half of their top talent but compensated 88% of this loss by recruiting other top students. Just over 10% were recruits from other S&E fields. Nearly 90%, however, came from outside S&E disciplines.

Changes in Major by Gender and Race

	The major finding of the analysis by gender and race is that top women’s persistence rates are lower than men’s in all S&E fields. See Table 7. Table 7. Four-Year Persistence Rates of Top Women Versus Top Men
Major		Top Men Persisting	Top Women Persisting
Engineering		82%	29%
Biological Sciences		40%	22%
Mathematical Sciences		33%	25%
Physical Sciences		99%	31%

Best & Brightest: Education and Career Paths of Top Science and Engineering Students

1. Introduction

2. Methodology

Figure 5. Percentage of Women Among Top Freshmen Interested in S&E (1985 and 1995)

Figure 6. Proportion of Nonwhites Among Top Students by Gender (1985 and 1995)

Figure 7. Number of National Merit Scholars Choosing S&E Majors Versus Other Fields (1982 through 1994)

Figure 7. Number of National Merit Scholars
Choosing S&E Majors Versus Other Fields
(1982 through 1994)