Computing Research News

April 2015     Vol. 27/No. 4

Back to Front Page

CRA Best Practices Memo on Evaluating Scholarship in Hiring, Tenure, and Promotion

By CRA Staff

The CRA Board of Directors has recently released its latest Best Practices Memo, “Incentivizing Quality and Impact: Evaluating Scholarship in Hiring, Tenure, and Promotion.” Distinguishing between quality and quantity is key to promoting the future growth of the computing and information field. The memo advocates adjustments to hiring, promotion, and tenure practices as well as to the publication culture. Below is a summary of the reports main points. Click here to download the full memo.

The recommendations in the report were developed over an 18-month period by the CRA Committee on Best Practices for Hiring, Promotion, and Scholarship, led by Fred B. Schneider (Chair) and Batya Friedman (Co-Chair). The committee conducted interviews in autumn 2013 with more than 75 academic and industry computing and information unit heads to understand the issues and gain insights from practice. Preliminary recommendations were vetted with department chairs and CRA Deans at the Snowbird Conference in July 2014.

“Above all, quality and impact need to be incentivized over quantity…What ultimately should matter when it comes to hiring, tenure, and promotion is the quality of the research.”

The memo provides recommendations for both hiring and tenure and promotion cases.

Hiring Recommendation. Evaluate candidates on the basis of the contributions in their top one or two publications, in concert with the research statement and the other standard material(e.g., letters of recommendation, full CV, teaching statement) generally read by hiring committees in determining whom to invite to campus for an interview and, ultimately, whom to hire. Candidates should identify publications where they have played a significant role.

Tenure and Promotion Recommendation. Evaluate candidates for tenure and promotion on the basis of the contributions in their most important three to five publications (where systems and other artifacts may be included).[1] Tenure and promotion committees should invite external reviewers to comment on impact, depth, and scholarship of these publications or artifacts as well as the standard material (e.g., full CV, research statement, teaching statement). Some institutions might ask a candidate to suggest which publications or artifacts be considered, other institutions might leave that determination to the external reviewers. Per standard practice, tenure and promotion committees should read the external letters and the standard material in determining tenure and promotion decisions.

Implementing these recommendations will require attention to the transition for young researchers. Annual or reappointment reviews (which often occur after three years of hiring) should reflect the emphasis on quality—not quantity—and should recognize that high caliber research activities may take two or three years to come to fruition (e.g., publication or artifact deployment) and even longer for the impact to become apparent. A corollary follows: Evaluation of senior faculty similarly should emphasize quality over quantity, with incentives for pursuing greater risk-taking in research activities.

Publication Culture. Systemic changes throughout the publication culture would help to support better scholarship. For example, publishers could remove page limits for reference lists and could allow appendices for data, methods, and proofs. Editors, as appropriate, could consider longer submissions with the understanding that, in such cases, a longer review period would be likely. In addition to conferences with published proceedings, other professional gatherings (that do not publish proceedings) might be held where work-in-progress could be presented.

“The field benefits when researchers build on each other’s work…Certain publication formats and review processes, however, encourage practices inconsistent with these elements of good scholarship.”

CRA Committee on Best Practices for Hiring, Promotion, and Scholarship

Members include: Lorenzo Alvisi (University of Texas, Austin), David Culler (University of California, Berkeley), Batya Friedman [Co-chair] (University of Washington), Eric Grimson (Massachusetts Institute of Technology), Mark D. Hill (University of Wisconsin), Julia Hirschberg (Columbia University), Benjamin Kuipers (University of Michigan), Keith Marzullo (National Science Foundation and University of California, San Diego), Tamer Ozsu (University of Waterloo), Frank Pfenning (Carnegie Mellon University), Jennifer Preece (University of Maryland), Fred B. Schneider [Chair] (Cornell University), Eva Tardos (Cornell University), Jennifer Widom (Stanford University), Jeannette Wing (Microsoft Research), and Ellen Zegura (Georgia Tech).

 “There is of course a long road ahead to shifting the culture and traditions of an academic field, especially because here it would mean changes to the publication infrastructure as well as reforming the value system.  But our’s is a field that is used to quick response and there is every reason for optimism."


[1]For some disciplines represented in Information Schools (e.g., philosophy), the publication outcome is a book, with the expectation that one book would be in press or published at the time of evaluation for tenure.

2015 CRA Board Election Results

By CRA Staff

CRA members have elected seven new members to its Board of Directors: Joel Emer, Stephanie Forrest, Michael Franklin, Greg Hager*, Farnam Jahanian and Vivek Sarkar. Five current board members were re-elected to the CRA Board: Sarita Adve, H.V. Jagadish, Margaret Martonosi, Greg Morrisett, and Kathryn McKinley. Their terms run from July 1, 2015 through June 30, 2018. Retiring from the Board as of June 30, 2015 are Corinna Cortes, Jeanne Ferrante, Lance Fortnow, Eric Grimson, and J Moore. CRA thanks them all for contributions during their service on the board.

* Hager is currently holding a non-elected position on the Board.

Joel Emer

Joel EmerDr. Joel S. Emer is a Senior Distinguished Research Scientist in Nvidia's Architecture Research group. He is responsible for exploration of future architectures as well as modeling and analysis methodologies. In his spare time, he is a Professor of the Practice at MIT, where he teaches computer architecture and supervises graduate students. Prior to joining Nvidia he worked at Intel where he was an Intel Fellow and Director of Microarchitecture Research. Even earlier, he worked at Compaq and Digital Equipment Corporation.

Dr. Emer has held various research and advanced development positions investigating processor microarchitecture and developing performance modeling and evaluation techniques. He has made architectural contributions to a number of VAX, Alpha and X86 processors and is recognized as one of the developers of the widely employed quantitative approach to processor performance evaluation. More recently, he has been recognized for his contributions in the advancement of simultaneous multithreading technology, processor reliability analysis, cache organization and spatial architectures.

Dr. Emer received a bachelor's degree with highest honors in electrical engineering in 1974, and his master's degree in 1975 -- both from Purdue University. He earned a doctorate in electrical engineering from the University of Illinois in 1979. He has received numerous public recognitions, including being named a Fellow of both the ACM and IEEE, and he was the 2009 recipient of the Eckert-Mauchly award for lifetime contributions in computer architecture.

Stephanie Forrest

Stephanie ForrestStephanie Forrest is Regents Distinguished Professor of Computer Science at the University of New Mexico in Albuquerque, and a member of the Santa Fe Institute External Faculty. Her interdisciplinary research studies adaptive systems and includes biological modeling (immunology and evolutionary processes), computer security, and software engineering. Professor Forrest received M.S. and Ph.D. degrees in Computer and Communication Sciences from the University of Michigan and a B.A. from St. John's College. At UNM, she served as Dept. Chair 2006-2011, and at SFI she has served as Interim Vice President for Academic Affairs and Co-Chair of the Science Board. She has received several awards and honors, including: the Stanislaw Ulam Memorial Lectureship (2013), the ACM/AAAI Allen Newell Award (2011), and the Presidential Young Investigator Award (1991). She is a Fellow of the IEEE.

Michael Franklin

Michael FranklinMichael Franklin is the Thomas M. Siebel Professor of Computer Science and Chair of Computer Science at UC Berkeley where he also serves as Director of the Algorithms, Machines and People Lab (AMPLab). The Berkeley AMPLab is integrating machine learning, scalable computing, and human computation to develop a next generation Big Data analytics platform. Components of this platform, including the Spark and Shark analytics frameworks and the Mesos virtualization layer have become key parts of the emerging Big Data ecosystem. AMPLab is supported by more than two dozen leading companies including founding sponsors Amazon Web Services, Google, and SAP and received an NSF Expeditions in Computing award, which was announced by the White House in 2012. Franklin was founder and CTO of Truviso, a real-time data analytics company acquired by Cisco Systems. He is an ACM Fellow and two-time winner of the ACM SIGMOD Test of Time Award.

Farnam Jahanian

Farnam JahanianFarnam Jahanian is the Vice President for Research at Carnegie Mellon University (CMU) where he is responsible for nurturing excellence in research, scholarship and creative activities. Recently appointed as CMU’s Provost, he will begin this position in June 2015. Prior to CMU, Jahanian led the National Science Foundation Directorate for the Computer and Information Science and Engineering (CISE) from 2011 to 2014. He guided CISE, with a budget of almost $900 million, in its mission to advance scientific discovery and engineering innovation through its support of fundamental research and transformative advances in cyberinfrastructure. Previously, Jahanian was the Edward S. Davidson Collegiate Professor at the University of Michigan where he served as Chair for Computer Science and Engineering from 2007 to 2011 and as Director of the Software Systems Laboratory from 1997 to 2000. His research on Internet infrastructure security formed the basis for the Internet security company Arbor Networks, which he co-founded in 2001 and where he served as Chairman until its acquisition in 2010. He has testified before Congress on a broad range of topics, including cybersecurity, next generation computing, and big data. Jahanian received his M.S. and Ph.D. in Computer Science from the University of Texas at Austin. He is a Fellow of the Association for Computing Machinery (ACM), the Institute of Electrical and Electronic Engineers (IEEE), and the American Association for the Advancement of Science (AAAS).

Vivek Sarkar

Vivek SarkarVivek Sarkar is Professor and Chair of Computer Science at Rice University. He conducts research in multiple aspects of parallel software including programming languages, program analysis, compiler optimizations and runtimes for parallel and high performance computer systems. He currently leads the Habanero Extreme Scale Software Research Laboratory at Rice University, and serves as Associate Director of the NSF Expeditions Center for Domain-Specific Computing. Prior to joining Rice in July 2007, Vivek was Senior Manager of Programming Technologies at IBM Research. His responsibilities at IBM included leading IBM’s research efforts in programming model, tools, and productivity in the PERCS project during 2002- 2007 as part of the DARPA High Productivity Computing System program. His prior research projects include the X10 programming language, the Jikes Research Virtual Machine for the Java language, the ASTI optimizer used in IBM’s XL Fortran product compilers, the PTRAN automatic parallelization system, and profile-directed partitioning and scheduling of Sisal programs. In 1997, he was on sabbatical as a visiting associate professor at MIT, where he was a founding member of the MIT Raw multicore project. Vivek became a member of the IBM Academy of Technology in 1995, the E.D. Butcher Chair in Engineering at Rice University in 2007, and was inducted as an ACM Fellow in 2008. He holds a B.Tech. degree from the Indian Institute of Technology, Kanpur, an M.S. degree from University of Wisconsin-Madison, and a Ph.D. from Stanford University. Vivek has been serving as a member of the US Department of Energy’s Advanced Scientific Computing Advisory Committee (ASCAC) since 2009.

President’s Budget for Science Mixed, But Computing Research Does OK

By Peter Harsha, CRA Director of Government Affairs

The President’s FY 2016 Federal budget request, released in early February, would present a bit of a mixed bag for Federal science agencies. While agencies like the National Science Foundation and Department of Energy would see some increases for their research investments — including investments in computing research — other agencies like the Department of Defense, NASA, and the Department of Homeland Security would endure cuts to their research budgets under the President’s plan.

The President’s request represents the first step in the annual appropriations process — a process this year, with Congress now completely in the hands of a Republican majority, likely to get reshaped around Republican priorities. But this does not mean that the President’s request is “dead on arrival” in Congress. While it’s likely that there will be significant differences of opinion over appropriate funding levels for the many programs in the budget, with their majority not “veto-proof,”[1] congressional Republicans will ultimately have to pass appropriations bills that the President will sign, or risk yet another government shutdown — something both parties are looking to avoid. The veto gives the President his leverage, and his budget represents his “marker” on the table in the negotiations.

The President’s budget courts controversy in one of its central assumptions: that the sequestration regime enacted into law by the Budget Control Act (BCA) of 2011 and designed to enforce strict caps on discretionary spending is harmful to the Nation and ought to be abandoned. Designed to curb deficit spending and reduce the national debt, the BCA enacted a 10-year plan to cut nearly $1.2 trillion from Federal spending by capping defense and non-defense discretionary accounts —that is, money that Congress appropriates every year, as opposed to non-discretionary spending like that on Social Security, Medicare, Medicaid and interest on the National Debt. Should Congress exceed those caps in any year, automatic cuts would trigger, lopping off an equal percentage of money from every discretionary account in the budget to get the budget beneath the cap. While both Republicans and Democrats have both agreed that this is a pretty poor way to run a government, the sequestration regime has actually managed to curb discretionary spending. In fact, discretionary spending is now nearly $200 billion less, in inflation adjusted dollars, than it was in FY 2010.

In his budget, the President proposes revising those caps. The BCA differentiates between defense and non-defense discretionary spending. The President’s budget would require exceeding the defense caps by $38 billion in FY 2016, and the non-defense caps by $33 billion. Among congressional Republicans, there’s little support for busting the non-defense spending caps by $33 billion. However, as this goes to press, the House Republican leadership is moving a budget resolution that would exceed the President’s requested increase for defense discretionary spending. In fact, House Republicans would like to see the $38 billion increase grow to a $90 billion increase above the cap, though it appears they would fund that increase through the use of the “Overseas Contingency Operations” (OCO) account — money outside the normal budget accounting used to pay for U.S. war operations.

But on the non-defense discretionary side, there’s little room for growth, and this will likely make for a tough year for Federal science agencies like NSF, National Institute of Standards and Technology, and the Department of Energy’s Office of Science, when the appropriations process concludes at the end of the calendar year (hopefully).

The President has requested increases for all three of those agencies. For NSF, the President would like to see the agency grown by 5.2 percent in FY 2016, to $7.7 billion. Included in that increase is an increase to the Computing and Information Science and Engineering directorate of about 3.5 percent, or about $33 million more than FY 2015 funding. CISE would see increases across all of its divisions of about 3.8 percent under the President’s plan, and CISE would play a role in nearly all of the agency’s Foundation-wide initiatives including:

  • $28.5 million in the Understanding the Brain initiative;
  • $13.5 million in Innovations at the Nexus of Food, Energy, and Water Systems (INFEWS);
  • $8 million in Risk and Resilience;
  • $1.8 million in the new Inclusion across the Nation of Communities of Learners that have been Underrepresented for Diversity in Engineering and Science (INCLUDES) initiative;
  • $94 million in Cyber-Enabled Materials, Manufacturing and Smart Systems (CEMMSS);
  • $84 million in Cyberinfrastructure Framework for 21st Century Science, Engineering, and Education (CIF21);
  • $70 million for Secure and Trustworthy Cyberspace (SaTC);
  • and $11.7 million for Innovation-Corps.

The Department of Energy’s Office of Science (SCI) would also grow at a similar rate to NSF under the President’s plan. SCI would see an increase of 5.4 percent, or $27 million, to $5.34 billion for the programs across the office. However, the Advanced Scientific Computing Research (ASCR) would see a disproportionate amount of growth under the President’s budget. ASCR would increase $80 million, to $621 million, in FY 2016 — an increase of nearly 15 percent. The primary driver behind the increase is the agency’s priority on its exascale computing efforts, for which it would spend $87 million in FY 2016. ASCR’s Mathematical, Computational and CS Research account would grow a more modest $2.5 million under the President’s plan, and the Computational Science Graduate Fellowship, a program for which CRA joined with the Society of Industry and Applied Mathematics (SIAM) in efforts to reverse cuts to the program proposed by the Administration in previous years, would be “fully-funded” at $10 million to fund a new cohort of fellows.

NIST would grow by nearly 30 percent in the President’s budget, in large part due to a focus on advanced manufacturing programs. The Science and Technical Research Service of NIST, where most of NIST’s core research efforts reside, would also grow by nearly 12 percent in FY 2016 under the President’s plan. Included are research efforts focused on advanced communications, cybersecurity, urban cyber physical systems, and quantum information science.

One area of the research portfolio that does not fare particularly well under the President’s plan and bears watching is funding for basic research (6.1) at the Department of Defense. The Administration would cut the basic research budget by 8.3 percent in this budget, a cut of $189 million to $2.09 billion in FY 2016. Applied research (6.2) and Advanced Technology Development (6.3) would grow — by 1.4 percent and 2.6 percent respectively — and DARPA would see an overall increase of about 3 percent to $3 billion in FY 2016. But the cut to the basic research account has some in the science advocacy community, including those of us at CRA, concerned. In part, this is perhaps a bit of gamesmanship from the Administration. The Pentagon knows that the Congress is favorably disposed to defense spending and the basic research account in particular, and so it’s often the case that they will propose cuts in one area like this, having some confidence that Congress will reverse the cut later, in order to “pay for” increases elsewhere in the agency that the Congress may be less inclined to support. But while it appears there is support for mitigating the cut to basic research in Congress again this year, the magnitude of the cut proposed in this budget may make it difficult to reverse completely. We will continue to watch and advocate for basic research and have more detail on this as the Defense Appropriations process moves forward later this year.

What is also not particularly clear at the moment to science funding advocates is how this will all play out in appropriations this year. What is clear is that it likely will not get done in “regular order” — having each of the 12 annual appropriations bills necessary to fund government passed by the September 30th end of the current fiscal year. Congressional appropriations staff are already talking openly about a “continuing resolution” strategy: passing a stop-gap funding bill to give Congress time after the September 30th deadline to finish up appropriations.

Before then, given the current climate, it is likely that the Republican Congress may force the President to use his veto on these spending bills, if only for political posturing and symbolism. But in the end, the bills have to pass. Because they have to pass, many believe that the endgame will feature yet another mammoth “omnibus” appropriations bill that will bundle many, or all, of the unfinished appropriations bills into one, must-pass measure that will likely be full of compromises — compromises that would likely be unpalatable for either side if the bills were free-standing. But doing them as a bundle decreases somewhat the amount of attention paid to each one and perhaps paves the way to getting appropriations done.

However it works out, we will have all the details for you on the Computing Research Policy blog! (http://cra.org/blog).

 


[1] The GOP lacks a large enough majority to override a presidential veto on a party-line vote.

That Was Then, This Is Now: 25 Years Ago in CRN

By Betsy Bizot, CRA Director of Statistics and Evaluation

A look through the back issues of CRN provides some interesting contrasts in what’s changed and what hasn’t. This is the second of a series of occasional articles looking back at the hot topics in CRN 25 years ago.

In Winter and Spring 1990, as reported in CRN:

The subcommittee on Investigations and Oversight of the House of Representatives Committee on Science, Space, and Technology held hearings to address problems of software system safety, reliability, and quality. The precipitating incident was several deaths that had occurred due to software errors in the control of a radiation therapy device. Notably missing from the list of concerns, by our current view, is security. But the Internet age was just dawning. Another article in the same issue of CRN referred to “the current NSF-sponsored networks (commonly called the Internet).”

The Taulbee Survey reported that in 1988-89, 807 PhDs were awarded in Computer Science and Computer Engineering. Thirteen percent went to women and one percent went to underrepresented minorities. Four percent of tenured and tenure-track faculty were female, and two percent were underrepresented minorities. In 2013-2014, 1,940 PhDs were awarded in Computer Science, Computer Engineering, and Information. Eighteen percent went to women and three percent to underrepresented minorities. Seventeen percent of tenured and tenure-track faculty are female and four percent are underrepresented minorities.

The Canadian government announced a large research funding program, Networks of Centres of Excellence, which included three groups working in computing research: the Institute for Robotic and Intelligent Systems (IRIS), the Institute for Telecommunication Research, and the Ultra Large Scale Integration network. Both then and now, CRA’s charter is all of North America.

The January 1990 CRN included 25 job ads, all for North American academic positions. The January 2015 CRN included 164 job ads. Twelve of the ads were for non-North American positions (primarily academic, some with overseas campuses of US institutions); three were for industry or government research labs.

An analysis by the Computing Research Board (CRA’s predecessor organization) reported that 60% of federal R&D expenditures in computer science and engineering came from the defense sector; the three largest sources of funds were, in decreasing order of funding, DARPA, NSF, and ONR. Most of the DOD funding supported applied research and development, but a significant fraction supported basic research. The landscape for research funding has shifted considerably. In the most recent proposed FY16 budget, defense (DARPA, the service labs, and DOE nuclear stockpile stewardship) supports about 29% of all federal IT R&D (definitions of “IT R&D” can differ across agencies). NSF is expected to support about 89% of all fundamental CS research at universities.*

The article “Is Computing Research Isolated from Science?” emphasized the need for computer scientists to engage in interdisciplinary research. Research funding was likely to support work directed at solving societal problems, and computing research was expected to be the enabling technology for advances in many of these areas, but computing researchers needed to be directly involved in working with other areas. However, many researchers were more interested in, and more rewarded by their communities for achievements in, the core of computing. Interdisciplinary research continues to be vital to solving societal problems. Computing researchers participate in interdisciplinary projects in health, education, transportation, and business. CRA's Computing Community Consortium helps organize workshops that bring together researchers from various disciplines to discuss how advances in computing could address societal issues such as Aging in Place, Brain research, and human computation.

*Thanks to CRA’s Director of Government Affairs Peter Harsha for providing the current figures for government funding.

April 2015 CERP Infographic

By Jane Stout, CERP Director

Deaf and hard of hearing (DHH) students are more likely to have a mentor at specialty institutions versus conventional institutions

Note: Thirty-five deaf and hard of hearing (DHH) and 20 hearing undergraduate computing majors reported who they went to most often for career advice and assistance. Seventeen DHH students were enrolled at institutions that specialize in providing support services for DHH students; 18 DHH and 20 hearing students were enrolled at conventional institutions. DHH students at specialized institutions were just as likely to have a mentor within their institution as hearing students. However, DHH students at conventional institutions were significantly less likely to  have a mentor at their institution compared to (a) their DHH student counterparts who were at specialized institutions, p < .05, and (b) hearing students, p < .05. These data suggest that institutions with accessibility built into their institutional identity tend to also foster access to mentors for DHH students. Importantly, mentors provide information and guidance for successful career development.

This infographic is brought to you by the CRA’s Center for Evaluating the Research Pipeline (CERP). CERP provides social science research and comparative evaluation for the computing community. To learn more about CERP, visit our website at http://cra.org/cerp/.

Solving the Equation: The Variables for Women’s Success in Engineering and Computing

By Christianne Corbett

More than ever before in history, girls are studying and excelling in science and mathematics. Yet the dramatic increase in girls’ educational achievements in scientific and mathematical subjects has not been matched by similar increases in the representation of women working as engineers and computing professionals. Women made up just 26 percent of computing professionals in 2013, a substantially smaller portion than the 35 percent women comprised in 1990 and about the same percentage as in 1960. In engineering, women are even less well represented, making up just 12 percent of working engineers in 2013.

With funding from the National Science Foundation, the American Association of University Women (AAUW) recently released Solving the Equation: The Variables for Women's Success in Engineering and Computing, which highlights recent research on the factors underlying the underrepresentation of women in these fields, including stereotypes and biases, college curriculum, and workplace environment, and makes evidence-based recommendations for change. Many of the recommendations are targeted toward employers as they play an especially important role in creating workplace environments that support women. Some specific recommendations include holding managers accountable for their hiring and promotion decisions so they're less likely to rely on stereotypes; removing gender information from job applications and evaluation scenarios when possible; making clear that a workplace with more technical women is a priority and a desired goal for an organization; and emphasizing the societal benefit of engineering and computing work.

Combating Stereotypes and Biases

We all hold gender biases, shaped by cultural stereotypes in the wider culture, that affect how we evaluate and treat one another. Several recent research findings shed light on the effects of stereotypes and gender bias as they relate to women in engineering and computing.

Many of you are likely familiar with the recent study by Dr. Corinne Moss-Racusin and colleagues which found that scientists were more likely to choose a male candidate over an identical female candidate for a hypothetical job opening at a lab. Both female and male scientists also offered a higher salary to the male candidate. Another recent study by Dr. Ernesto Reuben and colleagues found that potential employers systematically underestimated the mathematical performance of women compared with men, resulting in the hiring of lower-performing men over higher-performing women for mathematical work. Once objective past-performance information was introduced, however, the employers made less biased hiring decisions. Bias is prevalent, but its effects can be diminished with more comprehensive information.

Hundreds of studies have documented the power of stereotypes to influence performance through a phenomenon known as “stereotype threat” in many domains, including academic performance among black students, memory in older adults, girls’ chess performance, and women’s athletic performance. In every case even subtle reminders of negative stereotypes can have an impact on performance, sometimes in dramatic ways. For example, according to a recent meta-analysis, stereotype threat results in an underestimation of the intellectual ability of black and Latino students by approximately 40 points on the SAT math and reading tests.

Stereotype threat occurs when individuals fear that they will confirm a negative stereotype about a group to which they belong. One such group is “women.” When negative stereotypes about women’s mathematical abilities are brought to test-takers’ attention during tests, women’s performance drops. Stereotype threat has been theorized not only to influence women’s mathematical performance but also to contribute to disengagement from fields in which women are negatively stereotyped, such as engineering and computing.

Much research has been done on how stereotype threat can affect academic performance, but researchers are only recently beginning to examine how stereotype threat affects women in the workplace. One finding in this area, from a study conducted by Dr. Shannon Holleran, Dr. Toni Schmader, and their colleagues, showed that the more often female STEM faculty had research-related conversations with their male colleagues, the less engaged they felt with their work. In contrast, the more social conversations female STEM faculty had with their male colleagues, the more engaged they reported being with their work. One possible explanation for this finding is that research-related conversations with male colleagues may generate stereotype threat for female scientists. Social conversations with male colleagues, on the other hand, may lessen the threat by increasing a feeling of belonging in their work environment. Research suggests that stereotypes are activated for women more frequently when few women work in an organization. The presence of women at all levels of an organization has the potential to create environments that are less threatening for women.

Gender biases affect not only how we view and treat others but also how we view ourselves and what actions we take as a result. As early as first grade, children have already developed implicit biases associating math with boys. Studies suggest that girls who more strongly associate math with boys and men are less likely to perceive themselves as being interested in or skilled at math and less likely to spend time studying or engaging with math concepts.

A recent study conducted by Dr. Frederick Smyth, Dr. Brian Nosek, and Dr. Anthony Greenwald, to be published in a forthcoming issue of Frontiers in Psychology finds that most men who major in engineering and computing have relatively strong implicit biases associating men with science, whereas their female counterparts tend to have relatively weak science-male implicit biases. Engineering and computing workplaces have a wider gap in gender-science bias among female and male employees relative to other fields. Exactly how this gender difference affects the work environment in engineering and computing is an area ripe for future research.

Emphasizing Social Relevance

Another factor that may contribute to girls and women choosing to pursue fields other than engineering and computing is the small but well-documented gender difference in desire to work with and help other people. Although communal goals are widely valued by both women and men, research conducted by Dr. Amanda Diekman and colleagues finds that women are more likely than men to prioritize helping and working with other people over other career goals. Engineering and computing jobs clearly can provide opportunities for fulfilling communal goals, but jobs in these fields are not generally viewed that way. Rather, engineering and computing are often thought of as solitary occupations that offer few opportunities for social contribution. The perception and, in some cases, the reality that engineering and computing occupations lack opportunities to work with and help others may in part explain the underrepresentation of women in these fields. Incorporating communal aspects—both in messaging and in substance—into engineering and computing work will likely increase the appeal of these fields to communally oriented people, many of whom are women.

Cultivating a Sense of Belonging

Perhaps because of this combination of stereotypes, biases, and values, women often report that they don’t feel as if they belong in engineering and computing fields. A recent study by Dr. Erin Cech and colleagues found that female engineering students were less likely than their male counterparts to feel a strong sense of fit with the idea of “being an engineer” as early as their first year in college. This more tenuous sense of fit with the professional role of an engineer was found to be associated with a greater likelihood of leaving the field. By emphasizing the wide variety of expertise necessary to be a successful engineer or computing professional—including less stereotypically masculine skills such as writing, communicating, and organizing—college engineering and computing programs can help young women see engineering and computing as fields in which they belong.

Changing the Environment

Past decades have shown that simply trying to recruit girls and women into existing engineering and computing educational programs and workplaces has had limited success. Changing the environment in college and the workplace appears to be a prerequisite for fully integrating women into these fields.

College

Harvey Mudd College is a prime example of how changing structures and environments can result in a dramatic increase in women’s representation in computing. With leadership from the college president Marie Klawe, and college-wide support, Harvey Mudd increased the percentage of women graduating from its computing program from 12 percent to approximately 40 percent in five years. This dramatic increase was accomplished through three major changes: revising the introductory computing course and splitting it into two levels divided by experience, providing research opportunities for undergraduates after their first year in college, and taking female students to the Grace Hopper Celebration of Women in Computing conference. These changes can be modified and applied at other colleges and universities. Taken together, they provide a roadmap for reversing the downward trend in women’s representation among bachelor’s degree recipients in computing.

The Workplace

Finally, while many studies have focused on factors contributing to women entering STEM occupations, far fewer have looked at the arguably equally important question of why women leave these fields, often after years of preparation, and what factors support them in staying. Recent research by Dr. Nadya Fouad and colleagues sheds light on why some women leave the engineering workforce and why others stay. Women who leave engineering are very similar to women who stay in engineering. The differences the researchers found were not in the women themselves but in their workplace environments.

Women who left engineering were less likely to have opportunities for training and development, support from co-workers or supervisors, and support for balancing work and non-work roles than were women who stayed in the profession. Female engineers who were most satisfied with their jobs, in contrast, worked for organizations that provided clear paths for advancement, gave employees challenging assignments that helped develop and strengthen new skills, and valued and recognized employees’ contributions.

Stereotypes and biases lie at the core of the challenges facing women in engineering and computing. Educational and workplace environments are dissuading women who might otherwise succeed in these fields. Expanding women’s representation in engineering and computing will require concerted effort by employers, educational institutions, policy makers, and individuals to create environments that are truly welcoming for women.

Christianne Corbett is a senior researcher at the American Association of University Women. Patty Lopez served as an advisor for Solving the Equation.

Theoretical Foundations for Social Computing

By CCC

The CCC Visioning Workshop Theoretical Foundations for Social Computing will be held in Washington, DC on June 29-30th.

Social computing encompasses the mechanisms through which people interact with computational systems---for instance, crowdsourcing platforms, ranking and recommendation systems, online prediction markets, or collaboratively edited wikis. It is blossoming into a rich research area of its own, with contributions from diverse disciplines spanning computer science, economics, sociology, systems research, and HCI, to name just a few.

Foundational theoretical research has great potential to influence and shape the future of social computing. However, while there is a small amount of literature that uses theoretical models to analyze and propose design recommendations for social computing systems, there are several barriers that must be overcome and questions that must be answered before theory can have the same degree of impact on social computing that it has had in other fields:

  • What are the fundamental social computing problems? Is it possible to identify problems that are general enough to capture the core challenges of social computing across a wide range of applications, yet capture real issues?
  • What are the right models? A growing body of literature suggests that human behavior in many online settings often deviates from standard economic models of agent behavior, and that these deviations can have significant effects on how to optimally design social computing systems. Defining appropriate models requires a dialog between theory and experimental and empirical research.
  • How should we measure success? When evaluating the progress of the field as a whole, the criteria for success are vague, especially given that the capabilities and uses of social computing systems are changing all the time.

This visioning workshop has three major goals:

  • Identify core problems that the community believes are important to focus on in order to establish theoretical foundations of social computing.
  • Identify ways in which the theory community can learn from existing, ongoing, and future empirical and experimental work.
  • Identify effective ways for the theory community to have impact on social computing in practice.

For more information, please see the Theoretical Foundations for Social Computing website or contact Ann Drobnis.

Privacy by Design - Privacy Enabling Design

By CCC

The second in a series of four CCC Visioning workshops on Privacy by Design, Privacy Enabling Design will be held in Atlanta, Georgia on May 7-8th.

Building on the first workshop, this workshop will explore in depth privacy design practice. The goals are to survey current research on privacy tools and motivations and consider the effect this research should have on real-world problems, regulatory frameworks, and design practices in the public and private sector.

Examples of topics to be explored include:

  • How do designers and privacy professionals engage with privacy challenges in industry and research? What works today? What are the obstacles?
  • How do designers and implementers of products make decisions about tradeoffs with respect to privacy? How explicitly are various criteria considered when decisions are made?
  • What if any education, research, methods, policy, or organizational changes need to happen to empower designers?
  • How do we know when a product design or research framework is effective in respecting users’ privacy? What methods if any have proven themselves useful for privacy design or for evaluating privacy frameworks?
  • How is designing for privacy similar to or different from designing for security? How does trust factor into this process?
  • What privacy design challenges are posed by non-traditional technologies and interfaces, such as mobile devices, wearable computing, or the Internet of Things?
  • What can academia do to support work by practitioners and policymakers?
  • Does academic privacy research have any bearing on how privacy design is currently being addressed in industry? How can we improve the transfer of academic research into industrial practice?

For more information, please see the Privacy by Design website or contact Ann Drobnis.

Capitol Hill Presentation on Deconstructing Precision Agriculture

By CCC

Contributions to this post were made by Shashi ShekharComputing Community Consortium (CCC) Council member and Distinguished University Professor at the University of Minnesota.

The Task Force on American Innovation held a Capitol Hill reception titled “Deconstructing Precision Agriculture” on Wednesday, March 4. The Computing Research Association was a co-sponsor of the event. It showcased U.S. farmers, leading agriculture technology companies, and scientists including Computing Community Consortium (CCC) Council member and University of Minnesota distinguished university professor Shashi Shekhar.

The event exhibited three essential technologies of precision agriculture that originated from a broad spectrum of federally funded science: Guidance Systems and GPS, Data & Mapping with GIS, and Sensors & Robotics.

Rajiv Khosla, Professor of Precision Agriculture at Colorado State University opened the reception by saying that “precision agriculture is not rocket science, but we use rocket science to do precision agriculture.”

Shashi Shekhar explained that the geographic information system (GIS) based soil maps help farmers see that soil properties and fertilizer needs vary across locations in a large farm. GIS and complementary spatial computing technologies help farmers apply the right amount of fertilizer at each location within a large farm to increase yield while reducing waste and runoffs. If you apply the same amount of fertilizer everywhere you are over fertilizing in some places (which leads to increased runoff) and under fertilizing in others (which reduces yield). To relate the need for this technology, William R. Raun, Professor at Oklahoma State University pointed out, you don’t go to a gas station and put fifteen gallons of gas in a ten gallon tank, watching five gallons spill. GIS soil maps help tractor-based spatial decision support systems control for this and prevent excess runoff and conserve water, while increasing yield by location-aware rate of fertilization. Farmer Del Unger of Carlisle, Indiana, said that precision agriculture has dramatically increased yields and farm profitability. Farmer Rod Weimer, of Fagerberg Produce in Eaton, Colorado said it himself when he remarked that “Technologies like this makes farming more fun.”

Shekhar added that agriculture is not only a compelling use case of but also the inspiration for many transformative spatial computing discoveries and inventions. Positioning methods, e.g. modern GPS, started with land surveying by Egyptian civilization to reestablish farm ownership boundaries periodically erased by Nile floods. Spatial statistics traces its roots to concern of agriculture census that agricultural samples violated central assumptions underlying sampling theory. Many new basic research challenges and opportunities including many in Computer Science, are arising from the societal need of dramatically increasing farm yields without degrading environment to address food security (as well as the nexus of food, energy and water security) in face of growing population, increasing urbanization in developing world, and climate change.

This topic is of great interest right now. The National Science Foundation (NSF) released a Dear Colleague Letter (DCL) called SEES: Interactions of Food Systems with Water and Energy Systems that encompasses precision agriculture and more.

NSF established the Science, Engineering, and Education for Sustainability (SEES) investment area in 2010 to lay the research foundation for decision capabilities and technologies aimed at mitigating and adapting to environmental changes that threaten sustainability.

In this context, the importance of understanding the interconnected and interdependent systems involving food, energy, and water (FEW) has emerged. Through this Dear Colleague Letter (DCL), the NSF aims to accelerate fundamental understanding and stimulate basic research on systems that extend beyond the interests of the SEES Water Sustainability and Climate (WSC) program to include couplings to energy and food systems where the NSF already has established presence.


Shashi Sheker

The NSF requests innovative proposals for (1) supplements, to build upon existing NSF-funded research activities; or (2) workshops of typically 30-80 attendees that stimulate debate, discussion, visioning and collaboration across research communities, and enable a higher appreciation, visualization and understanding of food systems and their couplings to energy and water systems.

This is a huge opportunity with tremendous impact potential for the computer science community. These workshops are to prepare for the transition to the Innovations at the Nexus of Food, Energy and Water Systems (INFEWS) Program under NSF’s new FY 16 budget request. If you are interested, workshop proposals and supplement requests must be submitted by March 30, 2015 for consideration. For more information, please see the DCL.

CCC Logo

CERP Logo CRA-E Logo CRA-W Logo

1828 L STREET, NW SUITE 800, WASHINGTON, DC 20036 | P: 202-234-2111 | F: 202-667-1066