The workshop’s second day began with two rounds of six breakout groups, with each round lasting 75 minutes. Workshop participants were free to participate in the discussions in any of the six groups during either round. Discussion leaders led and guided the groups’ conversations using a series of questions to draw out relevant ideas and issues. The intent of these groups was not to create a list of specific recommendations for workforce development investments, but to identify important issues and challenges related to workforce readiness, and to create a range of thoughtful options for consideration by federal agencies; research institutions; the National Academies of Sciences, Engineering, and Medicine (the Academies); and other stakeholders in science, technology, engineering, and mathematics (STEM) workforce development.
The six groups and the topics to be addressed by the groups were as follows:
1. Exploring new/innovative pathways and effective articulation strategies. What are examples of new and innovative pathways toward a STEM career beyond the more traditional 2-year and 4-year degree from an accredited institution? What is appealing about the new and innovative education and training providers that have arisen in recent years—and what are legitimate concerns about these providers? Which mechanisms are most effective in assessing the quality of innovative, nontraditional providers and the quality of all providers? What roles can federal agencies play in encouraging innovative pathways and innovative providers while ensuring integrity and quality among the providers? How can federal agen-
cies encourage stronger partnerships between traditional 2-year and 4-year colleges and the new, alternative providers?
2. Professional skills or “employability” skills. What are the types of nontechnical skills that institutions need to develop more fully in their students to prepare them for career success? How are these skills taught and assessed, both by educational institutions and by employers? Where are institutions doing a good job—and a poor job—of preparing students for career success with respect to problem-solving, creativity, teamwork, and communication? How can agencies foster the development of skills in these key “non-cog” areas as part of regular undergraduate coursework, lab experiences, and internships/fellowships? What new and innovative assessment tools need to be developed to account for the integration of multiple knowledge, skills, and attributes that graduates need for career success? How can a more multidisciplinary approach to grant making foster student development in all of these areas?
3. STEM career awareness and participation. How can K-12 schools and postsecondary institutions provide greater awareness of STEM career options and career pathways for students, and ensure their participation in STEM academic programs and careers—especially among low-income and underrepresented minority students? How can federal agencies support career awareness efforts among schools and institutions? How can each stakeholder promote the attainment of advanced degrees in STEM fields, especially for students from underrepresented groups? How can agencies do a better job of providing information and awareness about education and training options to the more than 24 million adult workers who need “upskilling” in our workforce?
4. Faculty development. Particularly in rapidly changing STEM fields, what can be done to overcome “faculty calcification” in postsecondary institutions? By what means can federal agencies incentivize more project- and problem-based learning and critical thinking at the undergraduate and graduate levels? How can faculty training support student preparation for nonacademic career pathways for STEM graduate students? What are examples of new and innovative strategies for giving faculty an opportunity to experience the STEM workplace and to bring industry scientists to campus to teach and work directly with students with support from federal agencies? Are faculty able to remain current both in their particular fields of study and in broader, multidisciplinary areas that reflect the increasing influence of “convergence” on STEM careers?
5. Broadening participation and diversity. Much has been studied and written over the past 50 years about diversity, access, and equity in STEM, but how much has changed? Where are there time-tested examples of successful interventions at the K-12 and higher education levels that have increased the participation and success of women, underrepresented
minorities, persons with disabilities, and veterans in STEM? What works and does not work in bringing those models to scale? How can business and industry be more directly involved in recruitment and retention of women and minorities in STEM careers? How can the new INCLUDES (Inclusion across the Nation of Communities of Learners that have been Underrepresented for Diversity in Engineering and Science) initiative at the National Science Foundation (NSF) leverage significant reform in this area?
6. Emerging priority content areas. What are the content areas that demand more significant focus for preparing the STEM workforce of the future? Are these areas considered particularly important: math/quantitative literacy, computational skills, cyberinfrastructure, analytics, data science, and convergence? What are other areas that need special attention? What mechanisms need to be put into place to create a strong “employer voice,” as postsecondary institutions provide experiences that ensure students have broad and deep exposure to emerging content areas? Within these new and emerging content areas, how can NSF and other agencies support skills development, curriculum development, and faculty professional development?
The report for Group 1 was made by four individuals. Dale Allen, from Quinsigamond Community College in Worcester, Massachusetts, said one topic of discussion was how to integrate into federal funding the ability to monitor contiguous labor markets and develop regional approaches to respond to changes in those labor markets. One suggestion from this group was for NSF to consider using its funding pools to incentivize K-16 partnerships for stackable programming to be delivered in concert with state agencies that focus on regional economic development. As Allen reported, the discussion in this group raised the point that there are likely to be common competencies and curricula that could be shared across a state or region, or even nationwide to some degree. One question in this regard, said Allen, was how to bundle NSF funding in a way that would build capacity so that local successes could be developed regionally. He added that NSF has begun to work together with other agencies and to move in that direction to some extent.
Katherine McClelland, from the National Manufacturing Institute, said another topic of discussion in Group 1 was how to make sure students, teachers, and parents understand the career opportunities available in their region and to create mechanisms by which employers can signal what openings will be available and to identify the job skills students will need to fill those openings. She noted, too, that it is important to understand
how to translate the skills students are developing to fit employer needs. Oftentimes, said McClelland, the words employers use to describe skills are not the same as schools are using, but the skills are the same. This group, said McClelland, discussed the importance of using labor market analyses and data to identify the jobs that are available today and those that will be available over the next few years and to then provide that information to parents so that they will be motivated to encourage their children to stay in STEM courses despite challenges they might have with algebra and other classes that often discourage students from sticking with STEM pathways.
Mary Alice McCarthy, from the New America Foundation, then reported that one of the big issues discussed in Group 1 was how to think about workforce as an outcome of workforce development. “If we think about what that outcome is, that could change how programs and pathways are constructed because we are thinking about preparing people for the workforce as a major goal,” explained McCarthy. If workforce development is the desired outcome, she wondered what the effect would be on how NSF structures its programs and measures their success. “Are there outcomes we can articulate that can better reflect the workforce mission?” she asked. “We need to think about what those outcomes are and embed them in program structures.”
Michelle Van Noy, from Rutgers University, concluded the Group 1 report by noting two additional points that were raised during the group’s discussions. One point had to do with the alternative approaches to skills acquisition, such as massive open online courses, boot camps, gaming, extracurricular activities, and others, and the role NSF might play in incorporating these pathways into STEM education and STEM careers. NSF, she said, funds research, pilot programs, and evaluation of different approaches, and it was suggested during the group’s discussions that these alternative approaches might be a space in which NSF could become involved in evaluation, particularly regarding access and equity. “It seems there is a great role for NSF to help this space mature and to see what works and what does not work in the K-12 area and in the postsecondary school area,” said Van Noy. The second point the group raised was that NSF might develop some programs to help people who enter STEM through alternative pathways understand how their skills and knowledge translate across different domains and how to sell those skills to employers without the traditional credentials that come through more traditional pathways. The group said that NSF should not get involved in credential development, but rather in the area of skills translation, which it is already doing with some programs for military veterans. “This might help more of these outside spaces become inside spaces,” said Van Noy.
Rebecca Dernberger, from Manpower, reported Group 2’s discussions, which began, she said, by noting that students need to be at the center of whatever approach is used to develop employability skills. “Whatever the desired outcome, it cannot be accomplished without involving the students because they will be the ones who will be motivated to get the job, get the higher income, and be successful,” she said. This group also discussed the importance of developing a common language so that students, parents, and employers truly understand what employability skills are, and group members noted that active listening is one skill students need to develop along with the ability to change their behavior based on what they hear from employers. Other aspects of common language the group noted were project management, entrepreneurship, leadership, conflict resolution, and communication. Regarding communication, the group discussion raised the point that students need to be able to communicate with a wide range of people in the workplace. “You may be a scientist talking to a public relations person and you have to have the ability to make that connection,” Dernberger said by way of example.
One challenge raised during the discussion was the need to keep up with ever-changing workforce demands. Industry wants results fast, said Dernberger, and so the speed with which the system can produce workers with needed skills will be a key part of how well the STEM educational enterprise can meet the nation’s workforce needs. Group 2 also discussed data and assessment, and the suggestion was made that digital badging may be one solution. The group noted that certification may be a path for some industry skill sets, but credentialing efforts will only be impactful if led by industry. The discussion also raised the point that industry’s needs can present a moving target: when unemployment is high, companies look for the “perfect fit,” while when unemployment is low, companies become more adaptable and are willing to train new employees.
This group noted that the use of data is important for predicting demand and that social and behavioral scientists may be able to help with their tools for predictive behaviors and for assessing student soft skills. Educational and behavioral psychologists could also help standardize the assessment tools industry uses to judge a student’s suitability for employment. Other points of discussion included the need for career centers and universities to develop strong connections with alumni and the suggestion that schools can provide useful data on the types of jobs their alumni hold, the career pathways they take, and perhaps going forward, on income and income growth.
DeRionne Pollard, from Montgomery College, and Russell Rumberger, from the University of California, Santa Barbara, served as the rapporteurs for Group 3. Pollard reported the group spent time talking about the fact there is no simple way to share information about model programs that are occurring across the country for STEM awareness. “Each of us could talk about various programs that were happening in our area, but how do we gather information to share best practices so they can be scaled and replicated across the country?” she asked. The group also discussed the need for communities to assess the pool of local human capital and their capacity to form alliances between local business and educational institutions at all levels. One suggestion from the group was to think about what kinds of policies could motivate those types of evidence-driven partnerships and enable better information sharing across regions and the nation.
This group also discussed planning for career pathways and how such activities have become more deliberate. The group talked about the need to introduce students early to the idea of what it means to be STEM ready so that they can prepare thoughtfully for different options, both including and beyond academia. One way to broaden student perspective about STEM is through early work-based experiences through partnerships with industry that would expose students to some of these options. The group’s discussions also noted the importance of situating education so that is has more of a purpose for students and helps them better understand what they have to do to prepare themselves for these opportunities.
Group 3 also discussed the need for STEM education programs to do a better job of branding to shift the perception of STEM from being something for the elite class to an avenue for economic development and the ability to secure rewarding careers for middle-skill STEM workers. The question was raised as to how the STEM community can partner with a wide range of community groups, including social services organizations, unions, and places of worship to get this message out to the broader community. “There needs to be an intentional campaign to do this kind of work,” said Pollard. She also reported the group had a robust conversation about the value of the teacher (at all levels—middle school, high school, and college) in getting this message across to students and the need to develop programs to give teachers industry experience so that they can expose students to careers they may not have thought possible.
In his report, Rumberger noted the need to start early with any efforts to raise awareness of the wide range of STEM career opportunities as a major topic of discussion. Various members of the group pointed out there is a large body of information available about STEM careers and STEM career pathways, but the question is how the STEM community can do
better getting that information to students and teachers. One question that arose during the discussions was whether math and physics teachers know about the jobs that may be related to the subjects they are teaching, which led to the idea that there needs to be a mechanism for collecting and synthesizing information across the nation about successful approaches to raising awareness about STEM careers.
Rumberger mentioned a program he has been involved with that is being used with some success in California. This program starts with having ninth graders think about what they might want to do when they first enter the job market given their general interests and how those interests could translate into career pathways. The students then learn about the availability of different kinds of jobs they might like, how they would need to prepare themselves for those jobs, what kind of income they might realize in those jobs, and how that income translates into the ability to buy a house and car and support a desired lifestyle. The students then pick potential careers given their interests and desires and map an educational pathway to those careers. He also noted work being done in the field of behavioral economics, where information is shared using mobile phones and social media to get students to fill out Free Application for Federal Student Aid forms. “Perhaps we could take that kind of approach to build career awareness over time,” said Rumberger. Along those lines, there was discussion within this group about using such methods to enlighten teachers about career options for their students.
There was some discussion about gender differences related to experiences in STEM classes and how to overcome gender stereotypes about different STEM careers. The group also noted the challenge of raising STEM career awareness in a fluid labor market. “We talk about mapping students to careers that exist now, but the real challenge is preparing them for jobs that do not yet exist,” said Rumberger. “What kind of skills-based education do we provide them in that context?”
Debra Stewart, former president of the Council of Graduate Schools, provided a recap of Group 4’s discussions on how the federal government could best invest in faculty development in ways that would help the nation’s STEM workforce development efforts. From these discussions, the group identified three priority areas for investment. The first area had to do with how to encourage faculty at all levels to take seriously career development of their students, but particularly among college and research university faculty. The discussion raised the idea that NSF and other agencies should emphasize programs that explicitly recognize that many STEM students, and STEM graduate students in particular, pursue career pathways
outside of academia and need to receive encouragement to engage in professional development activities that will lead to success on those pathways. NSF training grants provide a straightforward mechanism for enabling a range of professional development activities, but as 80 percent of NSF funding goes directly to faculty, other mechanisms will need to be developed. Stewart noted there was strong interest within this group in building student career development into the broader impact statement associated with federal funding. “Faculty could demonstrate broader impact within the context of a grant by paying attention to the ultimate career outcomes of the students who benefit from a grant,” Stewart said. Along the same lines, the group’s discussions highlighted the need to encourage programs to track the career outcomes of their graduates and acknowledged the professional disciplinary societies are leading the way in this regard. Having said that, the discussions noted the importance of having NSF and other federal agencies encourage universities to collect data on student employment and career outcomes and then share those data with faculty.
The second priority area for investment to come from the group’s discussions concerned mentorship development for faculty as part of broader professional development activities. One suggestion from the group was to create opportunities to include mentor training as part of NSF grants to faculty who will be supervising students. It was noted during the discussion that NSF grants already include the requirement to mentor postdoctoral fellows, but the group also acknowledged that adding graduate student mentoring obligations to grants could be seen as an unfunded mandate, so NSF would need to consider how to fund this kind of activity. Stewart said that Group 4 had an interesting conversation about the need to increase the faculty community’s understanding and use of the proxy measures for teaching efficacy. “While there is evidence-based research on the efficacy of different strategies, it is not adequately disseminated, and the funding agencies need to begin to think about strategies for incentivizing the active utilization of what we now know about what works in teaching in the evaluation and promotion process for faculty,” said Stewart.
The third priority area for NSF Group 4 discussed was to advance and improve faculty members’ capability to more effectively utilize what is known about how students learn and embed that pedagogical knowledge into curricula. The group’s discussions noted that NSF, with the help of disciplinary societies, should find ways of disseminating new effective ways of teaching and encouraging faculty to adopt those methods. There was some concern voiced during the discussions that pedagogy may differ among the various STEM fields, which might then reinforce what is happening today at universities, which is that the chemistry department teaches one way and the physics department teaches a different way. This discussion led to a suggestion that NSF should reflect on how to take advantage of what many dis-
ciplines are learning and doing with respect to implementing new pedagogy and embedding it in curriculum while doing it in a way that simultaneously spreads this new understanding across STEM disciplines within a university.
Holly Zanville, from the Lumina Foundation, and Shelley Westman, from IBM, were the rapporteurs for Group 5. Zanville reported that one key point raised during the discussions was that it will take a 10- to 20-year commitment to programs aimed at broadening the diversity of the STEM workforce to have a major and long-lasting impact, but grants do not go on for this long. She noted that foundation representatives who were present during this conversation suggested program guidelines could accommodate requirements for sustainability and the corresponding development of sustainability plans. Another point the group raised was that meeting the imperative to change the face of science so that it looks like the rest of society will require getting rid of the stigma of exclusion that science and technology have. It will also require getting out the message that STEM careers can include people with a broad range of education and training.
The discussions noted, too, that multiple approaches are needed to address diversity gaps, and funding will have to be spread out across many areas, including social media campaigns and joint efforts with industry, and across all levels of education, including adult education, to make real progress. Efforts to spread the evidence about the importance of developing STEM capabilities across all segments of our society will take tremendous attention from the entire STEM community, Zanville said. The group pointed out the important role that industry will have to play as part of the solution.
This group’s discussions also raised the point that there are many innovative programs for mentorship, early engagement in research, and guided pathways that have demonstrated the ability to broaden diversity among students in STEM. The challenge now is to scale these programs to increase their impact. Regarding research, the discussions noted it is important to move beyond just counting heads and delve into what makes for a successful program. “We’re going to have to get a little deeper into the weeds as students move through these programs to understand where people are falling out,” said Zanville. Along the same lines, it is important to learn from programs that achieve near-zero rates of attrition and develop new pedagogical approaches for reducing attrition.
Other topics discussed included the need to better understand how to use social media more effectively and to get diversity models into the media, the need to use nontraditional and broadened pathways to broaden diversity, and the need to involve families more in efforts to encourage students to pursue STEM career pathways.
Westman added it is important to get across the message that diversity affects everybody, and particularly in STEM. “If you are a man and have daughters, diversity affects you,” she said. She also reported there was discussion in the group about why some successful programs do not scale, and one reason suggested was there may not be the same level of involvement, passion, or sponsorship when programs are introduced in new places.
The rapporteurs for Group 6 were Daniel Atkins, from the University of Michigan, and Gregory Camilli, from Rutgers University. Atkins started his report by noting that the participants in this group’s discussions were largely from academia, and they rejected the sweeping assumption that higher education was irrelevant to the future of STEM career pathways. “Higher education has and will continue to engage in transformative change in the nature of disciplines and related educational experiences,” said Atkins. Participants in this group noted that new fields and new content demands emerge from research activities in academia as well as industry, and that public universities in particular are the largest source of STEM workers with college degrees. It was pointed out that the interdisciplinary activities taking place at the nation’s large research universities have and continue to produce new fields and create new industries. One point made during this group’s discussions was the difficulty NSF review committees have dealing with interdisciplinary proposals.
Another point of discussion was that the emergence of new fields and the demands for new content are intrinsically an organic process that can and needs to be nurtured, but cannot be prescribed or built. Along those lines, Atkins said an important insight to come from the group’s discussions was that it is not NSF’s role to specify the exact content of these new areas but to nurture the processes from which they evolve and are propagated and disseminated. It was pointed out that NSF works to encourage community engagement at workshops for visioning and developing new solicitations, with the findings from those activities serving as the basis for new solicitations. New fields then emerge from the intersections and white spaces of existing fields, and this group said that NSF needs to continue supporting those white spaces. The discussions identified a number of emerging areas that merit continued NSF support, including adaptive manufacturing, neurotechnology, autonomous systems, and “x-analytics,” or big data, and also raised the point that when NSF identifies a new area of interest, it gets the attention of provosts and deans as well as other private funding agencies that also start investing in these areas. “The role of NSF is to support propagation and adoption of requirements for new
content and new educational experience that emerge from the research agendas that they help nurture,” said Atkins in summarizing the group’s discussions. “NSF encourages the engagement of researchers in educational innovation that promotes meaningful broader impact. At various times, this latter requirement gets taken more or less seriously, and lately NSF has been placing more emphasis on this.”
It was noted during the discussions that the federal government could do more to engage the research community in curriculum development and equipment sharing with K-12 systems, but that the U.S. Department of Education is prohibited from developing curricula. NSF, it was noted, has fewer constraints on curriculum development, but its budget for doing so is limited. Some of the participants in Group 6 suggested NSF needs to be more flexible in its assessment of education and learning research, particularly for projects proposed by groups outside of the formal education area who want minimal funding to pilot a new idea without the need for extensive and expensive evaluation. At the same time, it was noted NSF has played a meaningful role in developing and propagating new content relevant to contemporary science and engineering, and it should study some of these programs with the goal of replicating those successes. As examples, Atkins pointed to the key role NSF played in nurturing integrated manufacturing, the digital library movement that led to Google, the concepts of distributed collaboration, and most recently data science.
Atkins recounted that the group’s discussions did note there are some meta issues about how all of this will get done. “If you look at real innovations—open courseware at the Massachusetts Institute of Technology, the School of Information at the University of Michigan, the work in learning analytics coming out of the physics department at the University of Michigan—all of those trace back to leadership, people who stepped forward and took risks, people who were nurtured and supported by their management and who got venture funding,” said Atkins. The group also pointed out there will always be questions in academia about incentives, mitigation of risk, faculty not being able to step outside of the norms of publish or perish, and the need for flexibility in terms of broader notions of impact. Another challenge with new areas is how to characterize the skills that they require in students. For example, today, individuals with training in user experience and interface design are in high demand, but it was only a couple of years ago that very few people knew what those were and what skills they required.
Camilli said another topic discussed was employer concerns for high-demand skills, which he said is different from employer needs in high-demand fields. One subject not broached, he said, was globalization, which he characterized as the elephant in the room, and he thought it might be interesting to compare the National Science Board reports from 2012 and 2014 and the
different pictures they portray of off-shoring of research and on-shoring of production. “If we don’t take these issues into consideration, we’re not preparing the workforce we think we’re preparing,” said Camilli. He also voiced his opinion that NSF and the Academies should have American interests at heart, and those are not always the interests of industry. “I believe industry should be concerned about the return on investment for shareholders—that is industry’s moral imperative—but our moral imperative is to think about how globalization affects the kinds of skills and demand for those skills,” said Camilli as a final comment.
Jodi Wesemann, from the American Chemical Society, highlighted that throughout the breakout groups that she participated in there was a compelling need for data collection to feed back to the research questions and inform ideas coming forward. Atkins added he heard an appeal for applying big data to the topic of how to approach issues such as diversity and meeting the demand for skills. “This suggests we need to think about what kind of infrastructure we could put in place that would allow for the aggregation and sharing of data,” said Atkins. He noted that a coalition of Big Ten schools is cooperating on a system for sharing the output of learning management systems to inform learning research, and he suggested NSF might consider taking an inventory of projects such as this and making a modest investment to nurture cooperation, interoperability, and sharing between these projects. Along those lines, he noted NSF has started a regional data hub initiative that could grow and accommodate sharing of data relevant to learning research. “We have to not assume the infrastructure for data sharing actually exists and be intentional about developing that infrastructure,” said Atkins. Thomas Rudin, from the Academies, suggested there are lessons to be learned from the health care and manufacturing industries, which have built systems to share data in a manageable and concise manner.
Earnestine Easter, from NSF, commented that the University of Michigan, with NSF support, has developed a new institute that is building an infrastructure to enable a variety of longitudinal studies dealing with workforce development issues. This institute, she said, is integrating existing data from a number of sources to enable studies that will include examining the economic impact of investment in STEM workforce development. She added that the University of Michigan will be making this infrastructure available to all investigators who want to do research using these data.
As a final comment, Aprille Ericsson, from NASA Goddard Space Flight Center, remarked that when talking about innovation, it is important to resist not moving ahead and not taking risks in the absence of data. Science and technology advance, she said, because people take risks and try new things in the absence of data.