U.S. strength in science, technology, engineering, and mathematics (STEM) disciplines has formed the basis of innovations, technologies, and industries that have spurred the nation’s economic growth throughout the last 150 years.1 Despite this, our understanding of how universities receive, interpret, and respond to industry signaling demands for STEM-trained workers is far from complete. While there is anecdotal evidence that regions differ in their ability to link STEM degree production with local workforce needs, there has not yet been a systematic examination of how to measure effectiveness or to identify the best practices for accomplishing this goal. In response to this challenge, a committee organized under the auspices of the Board on Higher Education and Workforce of the National Academies of Sciences, Engineering, and Medicine undertook a 1-year study of the extent to which institutions of higher education and regional businesses and industries in five metropolitan communities (Phoenix, Arizona; Cleveland, Ohio; Montgomery, Alabama; Los Angeles, California; and Fargo, North Dakota) collaborate successfully to align curricula, labs, and other undergraduate educational experiences with current and prospective regional STEM workforce needs.
The committee sought to identify the current state of higher education’s responsiveness to regional workforce needs at the five target sites, identify barriers to stronger collaboration between universities and regional employers, and identify a number of promising practices and model programs that have fostered university-employer partnerships for aligning educational resources with STEM workforce needs. In particular, the study had two key objectives:
- To the extent to which such signals are available, understand how colleges and universities receive, interpret, and respond to industry signals regarding their demand for STEM-educated and STEM-trained workers.
- Understand the nature and scope of local and regional interactions among employers, nearby colleges and universities, and intermediary organizations focused on regional economic development.
It is worth noting at the outset that colleges, universities, and businesses are not entities that can engage with one another; rather, it is the people employed by those institutions who initiate, continue, and in some cases halt those interactions. As such, this study examines the conditions under which such interactions are most likely and least likely to take place, the conditions and circumstances that make those interactions most and least productive, and the types of obstacles that individuals and groups of individuals on campuses and in industries must overcome to create true and sustainable partnerships between their organizations. In addition to examining the importance of individual initiative and entrepreneurship, however, the committee looked at organizational structures, institutional cultures (both on campuses and in businesses), and the character-
1National Research Council (2012). Research Universities and the Future of America. Washington, DC: The National Academies Press.
istics of leadership in seeking to identify what was most likely and least likely to facilitate effective partnerships and alliances.
The outcomes of the study are detailed in this report. The hope of the committee is that the experiences and actions of the universities and employers at the selected sites can serve as models that could be adapted and/or scaled in other communities, regions, and states across the United States.
CHARGE TO THE COMMITTEE
This study sought to explore the effectiveness of educational institutions in producing STEM-trained workers in response to regional workforce needs and to identify promising practices in achieving this goal. The study’s statement of task is outlined in Box 1-1. To address the issues described in the statement of task, the committee organized five fact-finding regional workshops to gather information and ideas on challenges and strategies in building effective workforce development partnerships.
STUDY PROCESS AND METHODOLOGY: ASSUMPTIONS AND CONSTRAINTS
As described in the statement of task, the main focus of the study was on 4-year STEM degrees and the colleges and universities that grant these degrees. However, 2-year institutions are critical in the nation’s STEM education landscape, with nearly 50 percent of individuals
An ad hoc committee will explore the effectiveness of selected higher education institutions in educating STEM-trained workers in response to regional workforce needs and identify effective practices in achieving this. In this context, effectiveness will be judged, in part, on the number of STEM graduates and the extent to which these individuals graduate from educational programs with degrees that address local STEM workforce needs. The committee will conduct a series of regional meetings, commission work to analyze data on STEM degree production, convene a concluding workshop, and produce a report that compiles insights from all these sources to address the following questions:
1. What data and measurements are available to assess the effectiveness of higher education institutions in educating STEM-trained workers in response to regional labor market needs, and what do such measurements say about differences in effectiveness across regions?
2. To what extent can regional profiles be created that link STEM educational resources and postsecondary degree pathways with workforce needs, including retraining of professionals in order to meet current, in-demand needs in STEM fields? Is there sufficient resolution in the data to specify regional STEM degree needs by discipline or degree level?
3. What practices and policies are educational institutions adopting to respond to local industry STEM workforce needs, both individually and as part of regional coalitions, and which effective practices and policies are replicable and scalable?
What barriers, if any, exist that inhibit educational institutions from meeting regional STEM workforce needs? 4. What further actions are needed to assist higher education institutions, industry, and state or federal policy makers in fostering improved linkages between higher education resources and STEM workforce needs at the regional level? While the main focus of the study will be on 4-year STEM degrees, the role of community colleges and graduate institutions in meeting regional STEM workforce needs and transitions between institutions will also be examined.
receiving bachelor’s degrees in science and engineering having attended a community college at some stage of their education.2 Furthermore, within the context of a regional metropolitan area, local community colleges and universities are likely to have existing partnerships (whether formal or informal), and populations of students must navigate transitions between 2-year and 4-year institutions during the course of completing their undergraduate education. Community colleges also tend to have more of a workforce orientation than 4-year colleges and universities, given their critical role in upskilling adult learners, as well as providing terminal associate degrees. This career and technical orientation has resulted in many community colleges developing innovative career preparation programs and workforce development partnerships with industry—some of which are profiled in this report. Given these considerations, and the expectation that the committee and the 4-year community could learn from the important work being done at community colleges, 2-year institutions were included as active planning partners at every regional workshop, and community college administrators and faculty were well represented on workshop panels and in other agenda roles (see Appendix C). Because of this extensive involvement, this report contains many examples of innovative initiatives and practices occurring in the 2-year sector and also includes findings and recommendations drawn from these examples.
Among the primary sources of information for this study are the five regional workshops that the committee held around the country. As described below and in more detail in Box 1-2, the committee worked with partners in each region to identify key stakeholders, plan agendas, identify panelists, and invite a broad array of participants. As with any project of this nature, it is impossible to include every relevant stakeholder at every workshop—scheduling and logistical constraints will always present themselves. Because the absence of certain stakeholders from the five regional workshops may alter the content and conclusions of the committee, it is important to note some of the additional groups we endeavored to include at one or more workshops, but who were unable to participate for various reasons. These include nontraditional education providers, including adult upskilling organizations, representatives of professional and scientific societies, higher education accreditation organizations, and undergraduate students themselves (see Appendix C for more information on workshop participants).
The committee notes that its primary focus was on addressing questions 2, 3, and 4 as described in Box 1-1. In addressing question 2, the committee discovered that it was difficult to systematically assess how professionals already in the STEM workforce were being retrained, due to a lack of available data on the prevalence and scope of the job “upskilling.” Although continuing education and training may be especially important for incumbent workers in fast-evolving, highly skilled STEM fields, some reports have suggested that this kind of training is not very common in the United States.3 Regarding the Statement of Task’s question 1, the committee initially planned on commissioning analyses that would use an econometric approach known as stochastic frontier analysis to identify the efficiency with which institutions of higher education in a defined region produced undergraduate degrees in STEM (question 1 in Box 1-1). Using the Integrated Postsecondary Education Data System (IPEDS)4 data, these analyses calculate “effectiveness,” or the degree to which institutions produce greater or fewer STEM degrees than expected, given the financial and human capital available to them.5 Unfortunately, it became apparent that determinations of effectiveness require a larger number of colleges and universities as inputs to the analyses than are typically available within a given region. To identify those
2National Research Council (2012). Community Colleges in the Evolving STEM Education Landscape: Summary of a Summit. Washington, DC: The National Academies Press.
3National Science Board (2015). Revisiting the STEM Workforce: A Companion to Science and Engineering Indicators 2014. Arlington, VA: National Science Foundation.
4National Center for Education Statistics, U.S. Department of Education.
5Hurtado, S., A. Ruiz Alvardo, and K. Eagan (under review). Metrics, Money, and Degree Attainment: Identifying Engines of Social Mobility. Los Angeles, CA: Higher Education Research Institute, UCLA.
The overarching goal of each workshop was to convene key stakeholders that contribute to the creation and maintenance of regional workforce development partnerships. These stakeholders included faculty, administrators, and students from local colleges and universities, business leaders and company employees (small, medium, and large regional companies and national and global corporations), economic development professionals, policy makers (local, county, regional, and state), and philanthropic organizations. The following list details the key steps in planning each workshop.
1. Identify a motivated regional partner interested in helping to coordinate logistical and programmatic details in the selected city and region. For three workshops, this primary partner was a university (e.g., Arizona State University, Alabama State University, North Dakota State University); for the remaining two workshops, this partner was an intermediary entity interested in facilitating collaborative activities between higher education and industry in the service of regional economic development (e.g., the Ohio Aerospace Institute, the Los Angeles Area Chamber of Commerce).
2. Identify organizations interested in regional economic planning and development. These included city, county, or state chambers of commerce; city, state, or regional economic development organizations; state departments of commerce; and county or state workforce investment boards. Early in its study, the committee learned that these organizations are frequently key connectors and facilitators in their communities and in many instances were critical in helping the committee identify the appropriate individuals to invite to the workshop. These insights, and others, led the committee to make findings and recommendations about these critical intermediary organizations.
3. Using insights from intermediary organizations, identify and recruit key stakeholders to serve as speakers and panelists (see Appendix C for workshop agendas and participant lists). Most often, these individuals were deans and faculty from community colleges and universities, chief executive officers and department heads from industry (representing firms of all different sizes), economic development professionals, and philanthropic leaders.
4. Using the networks and insights from intermediary organizations and speakers/panelists, advertise the workshop and invite participants. Importantly, the committee viewed all workshop attendees as integral to an effective convening, and agendas were designed to solicit comments and suggestions from all workshop participants via ample discussion time and small breakout group sessions.
institutions most effective at producing undergraduate STEM degrees, these analyses must consider all of the nation’s colleges and universities, not just the handful of colleges and universities within a metropolitan statistical area, or even an entire state.6 Furthermore, the committee found that, for purposes of understanding issues of STEM education and workforce preparation at the regional level, data are inadequate. There is confusion about how to define a region and its boundaries: is it a metropolitan statistical area (as defined by the U.S. Census Bureau and used by other federal agencies) or something else? The U.S. Department of Education tracks graduation rates through its IPEDS system, and some institutions track subsequent employment of graduates, but there is not a systematic way to link disparate data sources to provide universities and employers with the rich and robust data they need to help meet local STEM workforce needs. For these reasons, the committee commissioned real-time labor market information (RTLMI) analyses to characterize the key occupations and skills that employers in a given metropolitan statistical area are seeking in new hires. The committee commissioned RTLMI analyses
for four of the five regions7 it visited. These results and their implications for workforce development partnerships in each region are described in the report.
At the outset, the committee took stock of current efforts to examine the effectiveness of educational institutions and programs in producing STEM-trained workers and how educational outputs are coupled with workforce needs. Given that recent work has demonstrated that a region’s economic prosperity is related to the educational attainment of its inhabitants,8,9,10,11 the committee took a regional approach to understanding these issues. The committee identified educational institutions and geographical regions within the United States that appear to be effective in producing STEM-capable workers in response to local industry signaling—settling on five regions to undertake its investigation and analysis: Phoenix, Arizona; Montgomery, Alabama; Cleveland, Ohio; Los Angeles, California; and Fargo, North Dakota. The rationale for selecting those sites is outlined in Chapter 3. Briefly, regions were selected based on the committee’s evaluation of the complement of higher education institutions (both 2-year and 4-year colleges and universities), the diversity and strengths of various industries (i.e., some regions have stronger representation in aerospace or advanced manufacturing, others in health care or agriculture), the region’s population size, and whether it was urban (Phoenix, Cleveland, Los Angeles) or rural (Montgomery, Fargo). The committee opted to use an inclusive definition of region, so as to capture the differences in how industry and higher education partners interact in the different communities visited.
The committee then worked with regional partners—starting with one or more universities or intermediary organizations in each region—to organize a workshop to highlight examples of promising practices and policies that institutions and industries are using to align curricula, labs, and other educational experiences with the needs of local employers, particularly those practices that might be scalable and sharable (see Box 1-2 for more details on the methods used to plan each regional workshop). The committee identified and invited key regional stakeholders to participate in each meeting, including higher education institutions, scholars, local industry representatives, regional economic experts, regional and state policy makers, and local foundations. Appendix C contains the agenda and participant lists for each regional workshop.
During the course of the study, it became apparent that a concluding workshop would likely not add any additional insights or observations that would help the committee meet its charge, and so was not held. This report represents the culmination of the committee’s effort, and the findings and recommendations contained in this report represent a call to action for the industry
7The committee commissioned the nonprofit organization Jobs for the Future (JFF) to conduct RTLMI analyses for Phoenix, Cleveland, Montgomery, and Fargo. RTLMI is a form of labor market intelligence that is drawn from online job postings. Vendors such as Burning Glass, The Conference Board, Help Wanted OnLine, and Geographic Solutions “scrape” these data from the Internet daily. RTLMI is distributed by vendors and is purchased by a variety of companies, governments and non-profits that desire access to the data to perform any number of analyses. JFF has no special or unique relationship with any vendor that would make the data it uses different from what any other organization would receive. For purposes of the committee’s report, the data that JFF staff used to help frame the conversations at each regional workshop were the same data that any subscriber could have pulled. In fact, several vendors would have likely provided similar analyses for their clients. The committee joined with the Los Angeles Area Chamber of Commerce to plan the Los Angeles workshop, and due to logistical constraints, it was unable to include JFF analysis in that workshop. Workforce and occupation data for Los Angeles was analyzed and presented by the consulting firm Beacon Economics.
8DeVol, R. C., I-L. Shen, A. Bedroussian, and N. Zhang. (2013). A Matter of Degrees: The Effect of Educational Attainment on Regional Economic Prosperity. Santa Monica, CA; Washington, DC: Milken Institute.
9Rothwell, J. (2013). The Hidden STEM Economy. Washington, DC: The Brookings Institution.
10Abel, J. R., and R. Deitz (2011). The Role of Colleges and Universities in Building Local Human Capital. Federal Reserve Bank of New York, Current Issues in Economics and Finance 17, no. 6.
11San Diego Regional Economic Development Corporation (2015). The Economic Impact of San Diego’s Research Institutions.
and academic sectors—especially 4-year universities and research universities—to improve their responsiveness to local and regional STEM workforce needs.
It is important to note the benefits and limitations of taking a regional perspective when looking at issues of STEM education and workforce development. As described above and in more detail in Chapters 2 and 3, universities can have real economic impact in their communities and the production of educated individuals who contribute to the region’s economy is just one of the public goods universities produce. The regional perspective has value to employers and industry, which have long recognized the importance of human capital, and many of the most productive regional economies in the United States are located in close proximity to some of the country’s top educational institutions. However, it is difficult to systematically track graduates who are mobile and may leave the immediate region where they obtained their degree to take a job elsewhere. This difficulty is due to uncoordinated data systems between universities and the various federal agencies who track educational attainment, wages, and migration, namely the U.S. Departments of Education, Labor, and Commerce, and the U.S. Census Bureau. A recent study was among the first to examine this issue, looking at employment and earnings outcomes for Ph.D. recipients by combining university administrative records with confidential data files from the U.S. Census Bureau.12 Given that mobility may be different for different types of graduates, depending on degree field, institution type, or other characteristics, the committee believes that this is an important area for future study.
STRUCTURE OF THE REPORT
The remainder of this report focuses on the activities, findings, and recommendations from the five workshops. Chapter 2 provides context for the study and reviews the literature on STEM workforce development strategies. Chapter 3 provides an overview of the five regions and the rationale for selecting each region as a workshop site, and also provides demographic and labor market information about each site. Chapter 4 summarizes the observations gathered at each workshop and identifies the STEM workforce needs articulated by industry leaders in each region, the barriers to successful partnerships, and promising practices used by both universities and employers to overcome the barriers and establish strong partnerships. It includes the specific mechanisms used by employers and universities to align college and university curricula, courses, labs, and internships/mentorships with the current and anticipated knowledge, skills, and attributes of workers. Chapter 5 includes the committee’s recommendations for actions, practices, and policies among all partners to improve higher education’s responsiveness to regional workforce needs. It also includes a discussion of remaining issues and questions that merit further study to determine even more effective ways to build and sustain strong partnerships between institutions of higher education and regional employers.
12Zolas, N., et al. (2015). Wrapping It Up in a Person: Examining Employment and Earnings Outcomes for Ph.D. Recipients. Science 350(6266):1367–1371.