MINUTES OF THE MEETING OF THE COMMITTEE ON
EQUAL OPPORTUNITIES IN SCIENCE AND ENGINEERING

February 7-8, 2002

National Science Foundation
4201 Wilson Boulevard
Arlington, VA 22230

Members Present:

Dr. Suzanne G. Brainard, Chair, University of Washington, Seattle, WA
Dr. Willie Pearson, Jr., Vice-Chair, Georgia Tech, Atlanta, GA
Dr. Marian Johnson-Thompson, National Institute of Environmental Health Sciences, RTP, NC
Dr. Indira Nair, Carnegie Mellon University, Pittsburgh, PA
Dr. Gustavo Roig, Florida International University, Miami, FL
Dr. Paula E. Stephan, Georgia State University, Atlanta, GA
Dr. Gary Toranzos, University of Puerto Rico, Rio Piedras, Puerto Rico
Dr. Thomas Windham, University Corporation for Atmospheric Research, Boulder, CO

Members Absent:

Dr. Kenneth E. Barner, University of Delaware, Newark, DE
Dr. David R. Burgess, Boston College, Chestnut Hill, MA
Dr. Bruce A. Jackson, Boston University School of Medicine, Wellesley Hills, MA
Dr. Lillian Shiao-Yen Wu, IBM Corporation, Somers, NY

Executive Liaison and Acting Executive Secretary:

Mr. John F. Wilkinson, Staff Associate for Workforce Development, NSF


The winter meeting of the Committee on Equal Opportunities in Science and Engineering (CEOSE) was held February 7-8, 2002 at the National Science Foundation in Arlington, Virginia.

Thursday, February 7, 2002

Welcome and Introductions

Dr. Suzanne Brainard, Chair, called the meeting to order at 8:41 a.m. The October 16-17, 2001 meeting minutes were approved. Dr. Brainard reviewed the meeting agenda.

Report of Executive Council Liaison

Mr. Wilkinson, NSF, reviewed conflict of interest guidelines and other administrative issues. He pointed out two publications in the meeting books, the Division of Human Resource Development (HRD) Summary of Programs and a report called "In Pursuit of a Diverse Science, Technology, Engineering, and Mathematics Workforce". He provided a brief update on activities at NSF. The FY2002 budget had an 8.5 percent increase. The FY2003 budget has a proposed 5 percent increase. NSF is pleased with this initial budget increase given all the activities going on. In the budget language, the President is recognizing NSF for its excellence and supports the priorities that NSF has laid out in their request. Stipends will be increased to $25,000 in the proposed budget. There is a "score card" being used by the Administration and NSF is the only federal agency (out of all agencies) that received a green mark for Financial Management. A stated long-term goal of NSF is a doubling of the budget. There is bi-partisan support for NSF and NSF is still working toward this agenda.

A $5.9M budget for bioterrorism was awarded to NIH. It was noted that some of these funds may be awarded to NSF and several directorates have responded to concerns related to terrorism.

Dr. Nair provided an overview of the EHR Advisory Committee meeting which she attended. The focus of the meeting was how to make science education accessible for all so that “no child is left behind”. It would be helpful to redefine this. How do you incorporate science education and communicate the methods of science? One thing that emerged is a meeting in June for Vice Provosts/Deans by invitation to look at some of the new models for science education. There is a need to look across the spectrum of K-graduate education, not just at the outcomes of individual programs. Dr. Nair has been asked to be the CEOSE liaison for the EHR advisory committee.

Discussion of NSF Data Collection and Reporting

At the last CEOSE meeting, a dialog was started with NSF’s Science Resource and Statistics (SRS) Directorate about data collection. SRS staff was asked to resume this discussion and a panel was put together to comment on the discussion.

SESTAT Survey

SRS conducts surveys of individuals educated or working in science and engineering. These surveys are collectively called SESTAT. Surveys included are: National Survey of Recent College Graduates (NSRCG), National Survey of College Graduates (NSCG), Survey of Doctorate Recipients (SDR), and Survey of Earned Doctorates (SED). Samples are taken from the individual surveys to make up the SESTAT survey. The surveys include major degree fields to include: Computer Science/Mathematics, Life Sciences, Physical Sciences, Social Sciences, and Engineering. It also includes individuals in S&E; occupations, but with non-S&E; degrees, though it is more challenging to identify these survey participants. The decennial census provides an opportunity to improve coverage of populations they can’t get any other way and provides motivation to do an overall design review.

Lynda Carlson, Division Director for SRS, talked about the SESTAT surveys redesign. The content for FY2003 will have only minor changes. SRS received an increase in funding to cover the cost of the survey redesign and would like an opportunity to come back to CEOSE and talk about the modules they are thinking about for the next decade. Mr. Ronald Fecso, Chief Statistician, talked about the SESTAT Sample Redesign and Mr. Fecso talked about how they get access to the people they include in the survey.

There are several issues SRS continues to face over the years:

Demographic characteristics will be included in the samples and occupational fields and degrees used in the past will continue to be used. With the different demographics, it is difficult to get significant sample sizes for all possible data cells and they need to be collapsed out of operation necessity. For example:

Demographics (2 sex) (5 race/eth.) (2 disability) (2 citizenship) = 40
8 degree fields x 2 degree levels x 40 demographics x 19 occupations = 18,240 cells!

SRS is in the process of the Sample Design Review with two main questions: 1) how to identify samples and 2) how to get the samples. SRS included other agencies in workshop discussions to identify if there were any other frames available to be included in the survey. The outcome concluded that the current modes now are really the best ways to get at the data. When the American Community Survey is implemented in 2007, there might be another mode. Until then, access modes will remain essentially the same, except minor tweaks in how the survey is conducted.

A chart was displayed depicting the evolution of the SESTAT surveys in the 1990s. The original sample from the decades shrinks down. SRS is constantly bringing in people from the prior Recent College Graduate Survey into the next survey data. They tend to sample down groups that are larger and keep the smaller sample size-groups to keep the hard to fill cells as filled as possible. Weighting is adjusted accordingly to make sure it is represented in a statistically viable way.

CEOSE asked if the data was linked to other data sources such as surveys at NSF or other agency data. The answer was no.

Complexity in the SESTAT survey design was demonstrated in a chart of the pros and cons for options SRS is exploring for the 2003 SESTAT Data Collection. The challenge is to determine the best option for data collection:

Over time there are non-sampling errors that creep up in survey sample (i.e., more drop out). Split sampling (Option 3) will help compare some of the major sources of error between the data. SRS is also hoping to move towards decentralized CATI data collection, which should result in a higher response rate. The split-sample approach is recommended as SRS feels it offers a win-win situation. Additional information SRS is looking at is below the bachelors and CPS analysis. There will be a 2003 modification to get “field of degree” added to Census survey. SRS would also like to cooperate with the American Academy of Arts and Sciences (AAAS) to work towards a much larger base Census survey from which to target the selection of S&E; cases. They would expect more minority individuals with more diverse educational backgrounds for sampling and for researchers to study.

Responses and Reactions from Panelists

SRS was asked how they deal with margins of error and non-response. The Census long form starts with an 80% response rate. This is weighted by Census to adjust for numbers that aren’t there. Regular on-going samples have a 90-95% response rate. SRS is currently doing some research on ways to bring back non-responses such as contacting them the following year, not just removing them from the list.

One additional point raised is the importance of being able to mine the data and longitudinal studies are important for this. There needs to be resources to really analyze and disseminate the survey results. SRS now has someone on staff now that will help identify number of graduate and post-docs with disabilities. Going forward for FY2003, they will have more detailed race/ethnicity categories.

Issues Related to Reporting

The CEOSE group and the panelists talked about issues related to reporting:

CEOSE stressed the criticality of getting data for the CEOSE report by the June 2002 meeting. Tables with empty cells would be useful. SRS asked for specific examples of the kind of tables they wanted (like the Engineering sample provided to the group earlier).

Women and Minorities and Persons with Disabilities Report

Dr. Joan Burrelli, Senior Analyst, SRS, thanked those who provided input on the draft chapters and stated that the Women and Minorities and Persons with Disabilities Report will be published around April 2002. The report data would be discussed in more detail later in the meeting.

Preliminary Discussion on Research on Minorities

Dr. Brainard provided a brief overview of cross-cutting issues related to research on minorities and things that are inhibitors in terms of being able to make policy recommendations.

What they do know:

What they need to know:

The CEOSE raised several additional concerns/issues related to research on minorities:

Diversity is often interpreted as a focus on a specific group. Very often it is women as opposed to underrepresented minorities and persons with disabilities. If programs are funded for women, the institution often claims they are supporting diversity. Programs for women are more often integrated into departments where programs for minorities/persons with disabilities are still at an administrative office of an institution.

There are a few institutions that are producing a higher percentage of women and minority graduates. They can be identified and their practices reviewed for some insight. CEOSE needs to be able to prioritize these areas where research and data is needed. They hope to continue to engage NSF staffers and the research community in these discussions at the next meeting.

Discussion of Use of Supplements to Broaden Participation

Dr. Brainard welcomed Dr. Judith Ramaley, Assistant Director, Education and Human Resources Directorate. The CEOSE asked Dr. Ramaley and her staff to help identify ideas that could be implemented to bring about change and broaden participation by underrepresented groups now.

Dr. Ramaley commented that in the past few months EHR has been evaluating their portfolio in terms of “themes”, rather than individual programs. For example, EHR is looking at each stage in the path to a career in science, technology, engineering and mathematics and the factors that influence whether an individual proceeds to the next level or goes in a different direction. In this evaluation process, they are asking three questions:

A larger issue is the need for ALL citizens to have a deeper understanding of science and mathematics and identifying ways to ensure effective participation.

Dr. Roosevelt Johnson, Acting Division Director, HRD/EHR said that broadening participation is a primary focus of HRD. The NSF/CEOSE had made two recommendations to HRD:

  1. Create fellowships and scholarships for minorities on existing grants in NSF programs.
  2. Provide supplements for undergraduates who have graduated to do a year of training in research (after BA/BS)

HRD is looking at ways for leverage/synergy using the fellowship and supplement mechanisms.

Dr. James Lightbourne, Acting Division Director, DGE/EHR stated that eligibility criteria that are exclusionary are vulnerable to legal challenge. A good program to discuss is ADVANCE. Goals for the program are specific to increase participation for women, but anyone can apply. Observations from research tell us that characteristics of students likely to major in S&E; include: 1) they have taken advanced high school science courses, 2) they are self motivated to study science, 3) they have parents with a relatively high level of educational achievement, and 4) their parents have high expectations for the child’s education. With these factors held constant, racial/ethnic and gender differences among S&E; majors diminish. Another study shows that targeted retention efforts such as pre-college bridging, mentoring, tutoring, individual support, and scholarships have a place, but they are only a temporary remedy. Such services tend to stigmatize the participants and some students elect not to use these programs for this reason.

Current mechanisms at NSF include Research Experience for Undergraduates (REU) supplements to individual researchers to support individual students and REU site grants to support cohorts of students. From 1990-1999, there were approximately 4,300 REU supplements ($18M), over 500 institutions and approximately 380 sites in the summer of 1998 ($31M). This is a fairly large funding mechanism at NSF.

Over the years, several REU Program evaluations have been conducted:

Dr. Lightbourne summarized some of the findings of the report tracking students. It started with about 70% wanting to pursue higher degrees and ended with 90% after they participated in an REU. A fairly high percentage of REU participants go into science careers. Dr. Roosevelt noted one of the take home messages is that the REU mechanism appears to have some value. To maximize this value, it can be put into a different context.

A number of other programs have been devised at NSF to take advantage of the lessons learned with REUs:

Relevant Lessons Learned:

Dr. Roosevelt shared suggestions for a potential approach to use REU-like programs to help increase diversity:

EHR suggested that CEOSE propose characteristics of a program or approach to increase diversity and participation, and then perhaps EHR can get back ways to implement the approaches within their programs.

Discussion of Elements in Successful Programs

CEOSE talked about elements that might be included in successful programs:

EHR is open for discussions on how to measure success. One measure is the numbers of participating individuals. Another would be to see how many individuals are still in S&E; fields five years after they participate. Support from CEOSE would help EHR with the challenge of integrating those activities across divisions/directorates. CEOSE requested data on the success of these programs like REUs, LSAMP, CREST, etc. How can they benefit from the data they have to be able to model programs? The data they have, however little, may help them draw some conclusions about what works.

Discussion of Approaches to Increase Participation CEOSE and EHR discussed possible approaches to bring about change to increase diversity:

Dr. Johnson commented that infrastructure is needed to bring about these programs. EHR is focusing on projects that already have the infrastructure in place. CEOSE asked for suggestions on how to frame their recommendations.

Dr. Johnson summarized the discussion of approaches:

  1. Going to states/governors and soliciting funds to provide grants to increase numbers. This is a radically different way of doing business, but the argument can be made.
  2. Give funding to institutions that do the job well. They should have a pre-selected universe of institutions, but then have a selective process within that.
  3. Link currently existing programs in such ways to accomplish goals of broadening participation.

There are different degrees of difficulty for each of these, but they are all viable options.

Committee Discussion – Items for the NSF Director

Items for the Director, NSF:

  1. There is a need to recognize the number of underrepresented US students in graduate programs (compared to increasing number of foreign students) in science and technology as a national security issue that impacts the prosperity of the Nation. As such, we cannot conduct “business as usual”. Increasing the number of US citizens as Ph.D.s in science and technology is a long-term goal and multiple approaches will be needed to achieve it.
  2. CEOSE would like to look at ways to combine different programs that NSF already has for increasing diversity, but make them work synergistically. NSF should be proactive about recognizing the need for solicitations that are “selective”.
  3. The group liked the approach of looking at NSF programs along a “pathway” of the student and looking at ways to change the focus of these programs to include broadening participation (and include persons with disabilities).
  4. The data on women, minorities and persons with disabilities need to be disaggregated, even if this means that the sample “cells” are empty.

The meeting adjourned for the day at 5:05 pm.


Friday, February 8, 2002

The meeting was reconvened at 9:00 a.m.

Data from Women and Minorities Report

Dr. Joan Burrelli, SRS, met with the group to identify table layouts for the data they would like to see. She provided a sample report. The level of detail on the last page of the sample was what they are looking for. In looking at Science and Engineering majors at Tribal colleges, there is a problem in that many tribal colleges don’t offer four year degrees, and of those that do, there are few degrees offered in science and engineering.

The group requested to see data, if possible, for:

Dr. Burrelli will also distribute a brief on the Survey of Doctoral Recipients based on the desired sector and actual sector. Dr. Brainard added that once CEOSE gets the data they will need to determine what they will present, what programs will address those gaps, and what recommendations CEOSE might offer. CEOSE requested the data tables prior to the June 20-21 meeting for distribution. The group thanked Dr. Burrelli for providing the sample tables.

Committee Discussion

Dr. Brainard noted that Dr. Pearson and she met with Dr. Colwell who was pleased with progress of the committee. During the prior evening dinner with Dr. Bordogna, he briefly reviewed the budget and those present talked about the issues discussed during the day. There also seemed to be opportunities to connect universities with K-12 system. Dr. Bordogna had helpful suggestions to refine the ideas from the day before.

Dr. Pearson requested additional names of people to recommend for reviewers, etc., be forwarded to him. Dr. Colwell was appreciative of the resumes/names provided by the group.

Several suggestions were made for presentations or topics for discussion at the next CEOSE meeting:

It was suggested that there many be an opportunity for another panel with the invitees.

Discussion with Director of NSF

Dr. Colwell recognized Tom Windham and his SOARS program which was selected for the Presidential Award for Excellence. Dr. Colwell provided an update on the NSF budget. The FY2002 budget was approved with an 8.5% increase. National priorities have changed dramatically with the September 11 attacks, but support for science and engineering has continued. The FY2003 budget has been approved by the President and now goes to Appropriations. NSF currently has a 5% increase in the President’s budget. NSF is trying to find innovative ways to increase participation of underrepresented children and schools in science and engineering. They now have funds to increase the stipends for graduate fellows to $25,000. NSF is proud of recognition given them by Mitch Daniels, Director of OMB, where he said “NSF is one of the true centers for excellence in the government” and that NSF is one of the best managed agencies in the government.

The group presented Dr. Colwell with several of the issues they had identified:

Dr. Colwell thanked the CEOSE for their update.

Hopwood and the Top 10 Percent

Dr. John Kain, Director, The Cecil and Ida Green Center for the Study of Science and Society, talked to the CEOSE about a program in Texas that provided an interesting perspective for CEOSE discussions: Minority Access to Higher Education. This is part of the University of Texas, Dallas Texas Schools Project. Analysis is based on the Texas Schools Microdata Panel which has panel data for more than 12 million individuals attending Texas public schools and colleges/universities for 1990-2001.

Dr. Kain shared preliminary findings from 4 working papers, focused mostly on African American data:

He reviewed potential reasons for low minority test scores. Data indicates that the school quality (average test scores) is related to the achievement of the students and typically, black children are at schools that have poorer school quality. Additional findings were that the Texas legislation for Hopwood and Texas’s Top 10 Percent Law on Minority Enrollment in Texas Selective Public Universities seemed to contribute to the decline of minority enrollment. Hopwood is a ruling of the Supreme Court that it was illegal for Texas public colleges and universities to use race as a factor in making decisions about college admissions and financial aid. The Top 10 percent law provided the automatic admission of all students who graduate in the top 10 percent of their high school graduating class to any Texas public college or university.

Charts illustrated that test scores in Reading and Match for minorities are lower than white students [in Texas schools] and continue to stay that way, if not increase in gaps. Other data charts were shown illustrating that black children in the highest testing groups will be impacted the most by the school quality.

The report recommendations were that efforts need to continue to reduce the racial concentration of schools or they need to find ways to increase the mean achievement of the inner-city schools. One way to do this is to facilitate increased-minority access to suburban schools. They also need to increase level of achievement in the lower grades.

Data showed a decline in Hispanic/African American enrollment at Texas colleges and universities since the Hopwood law has been in effect. In Dr. Kain’s opinion, affirmative action is an important tool to help improve minority achievement/enrollment.

Dr. Brainard thanked Dr. Kain for his presentation.

Committee Wrap Up

The group talked about CEOSE tasks over the next few months.

CEOSE Congressional Report. There should be an intense discussion of the CEOSE Congressional report at the June 2002 CEOSE meeting to include data that should be included in the report and dissemination plans. The report should also be published on the web.

Workshop Activities. CEOSE also talked about a possible workshop activities to help establish links between successful programs to increase diversity.

Drs. Brainard and Pearson will talk about the structures of the workshop(s) and who should be included and then would provide input to Marilyn [from NSF]. It was suggested that 1-2 people from CEOSE be on a workshop subcommittee.

The dates for the next CEOSE meeting will be June 20-21, 2001. With no further discussion, the meeting was adjourned at 1:00 p.m.

Action Items



 John F. Wilkinson
 Executive Secretary


[Return to top of page]

[Return to the CEOSE home page]