Data Science newsletter – May 24, 2020

GROUP CURATION: N/A

Amid enormous uncertainty about the future of the Covid-19 pandemic, epidemiologic models are critical planning tools for policymakers, clinicians, and public health practitioners. Some models with apparently conflicting conclusions have received substantial press coverage, giving the impression that mathematical models are in general unreliable or inherently flawed. But infectious disease modeling is an expansive field with a long history, encompassing a range of methods and assumptions that are not necessarily directly comparable, or even designed for the same purpose (see box).

Michelle O’Malley and Matthew Helgeson, chemical engineering faculty at the University of California, Santa Barbara, are still adjusting to life under the coronavirus pandemic.

For the married professors, this means running their respective research groups from home while also taking care of their almost-3-year-old son. Elliott, like most preschoolers, needs constant attention. Except during nap time. That 2 h stretch about every other day is the only time that they can count on getting work done. “That’s essentially where we are, as sad as that is,” O’Malley says with a rueful laugh. “We pack Zoom meetings into nap time and hope for the best.”

In between virtual group meetings and checking in with students, they try to analyze data or edit manuscripts, which are now their only options to move research forward. With lab experiments on hold indefinitely and grant managers asking how they’re using the time, O’Malley is anxious about the possibility of losing grants that support trainees. “That keeps me up at night a lot.”

In response to the COVID-19 pandemic, researchers from the Massachusetts Institute of Technology have published a pair of studies in a COVID-19 special issue of the Harvard Data Science Review, freely available via open access, describing new methods for accelerating drug approvals during pandemics and for providing more accurate measures of the probabilities of success for clinical trials of vaccines and other anti-infective therapies.

Aggregated COVID-19 information matters to public health, but it’s also a big economic opportunity. Now a Trump-aligned company gets that data daily from all 50 states.

With infectious-disease experts forecasting recurring waves of Covid-19 contagion, a number of colleges are coalescing around a plan to send students home by Thanksgiving this fall.

Hundreds of institutions have pledged to return to in-person classes in August, after the coronavirus forced them to move instruction online in the spring. While some of those colleges intend to return to normal operations, others have configured their calendars with earlier start and end dates.

Rice University, a private institution in Houston with about 7,000 students, will have an abbreviated fall semester. While students will report to campus as scheduled, the semester will end in November — about a month earlier than usual.

“We decided to make that decision more quickly than other schools in order to make sure we have the maximum time to prepare,” said David W. Leebron, Rice’s president. He added that the decision was made in late April.

The coronavirus has underscored inequities in American society, including among college students. When campuses first began emptying in early March, low-income and first-generation students were more likely to have lost critical sources of food and shelter. When courses moved online, a strong Wi-Fi connection and quiet place to focus were harder to come by.

Now, in the next phase of the pandemic, new inequities are poised to arise, this time among both who gets into college, and among who completes their degrees. So argues H. Holden Thorp in the latest issue of the journal Science, of which he’s editor in chief. He has two ideas for leveling the playing field: Suspend both the U.S. News & World Report college rankings, and the use of standardized tests in admissions.

The people who design and build cities have a saying, or they used to: The mistakes of the planners are inherited by the health department. When basic city functions fail, people get sick.

So it makes sense, in a syllogistic sort of way, that the converse might also seem true: If people get sick in a city, the planners must somehow be at fault. When 193,000 people test positive for Covid-19 and nearly 16,000 die in New York City, the densest major urban concentration in the United States, maybe the closely woven fabric of the city itself is to blame. Both Governor Andrew Cuomo and Mayor Bill De Blasio said it straight-out: New York’s density makes it especially vulnerable to a respiratory disease pandemic—all those people crowding into subways, skyscrapers, studio apartments, Brooklyn coffee shops, and presumably Greenwich Village hep-cat jazz boîtes, asymptomatically exhaling the virus on each other and causing a catastrophe that could play out again in city after city without some posthaste suburbanization. “There are mechanistic reasons we would expect there to be more transmission in places where the population density was higher,” says Linsey Marr, a Virginia Tech researcher who studies airborne virus transmission. “I think there are more opportunities for transmission.”

The attrition has occurred at even the top part-time programs: Between 2016 and 2020, the University of Chicago Booth School of Business, currently the No. 1-ranked program by U.S. News, shrank from 1,366 students to 1,245, an 8.9% decline. UCLA Anderson School of Management, ranked No. 5, saw a more significant loss, from 986 students to 857 (13.1%). No. 4 NYU Stern School of Business saw enrollment decrease by a whopping 242 students, 1,381 to 1,139, equaling 17.5%. No. 6 University of Michigan Ross School of Business went from 430 to 406 (-5.6%), and No. 9 Indiana University Kelley School of Business fell from 294 to 278 (-5.4%). The biggest drop-off for a prominent school was at No. 8 Carnegie Mellon University Tepper School of Business, which dropped from 143 to 74, a decline of nearly half (48.3%). … Enrollment at No. 3 Northwestern Kellogg, however, has declined only slightly, from 804 to 795 (1.1%). Why? Emily Haydon says engagement and innovation are, as always, prime factors. But there’s more. As part-time programs have struggled to remain viable in a sea of new options — online MBAs, specialized master’s offerings, MOOCs — Kellogg’s E&W program has grown its offerings, improved engagement with employers, and invested in better marketing of the advantages of the format. It’s also super flexible, with students having the option to learn more on weekends, more in the evenings, or some combination of the two.

“Most of the ones we test are the individuals who die at home,” says Gary Watts, the coroner in Richland County, S.C., who is president of the International Association of Coroners and Medical Examiners.

If family or friends say the person had symptoms consistent with COVID-19, the coroner’s office will typically do a nasal swab to test for the virus, he says. If the test is positive and the office can determine the cause of death without an autopsy, one will generally not be performed.

Coroners and medical examiners have similar responsibilities but their backgrounds are often different. Coroners are typically elected officials who may or may not have a medical degree. Medical examiners are typically medical doctors and may have a specialty in forensic pathology.

Welcome to the annual Milliman Medical Index (MMI), which gauges the yearly price of an employer-sponsored preferred-provider organization (PPO) health insurance plan for a hypothetical American family and an N of 1 employee.

That is a 4.1% increase from the 2019 estimate, about twice the rate of U.S. gross domestic product growth, Milliman points out in its report.

The coronavirus pandemic could push as many as 60 million people into extreme poverty, the World Bank said on Tuesday.

The warning suggests deepening pessimism among economists about the scale and duration of the fallout from what the bank described as an “unprecedented crisis.”

The World Bank, which provides loans and grants to the governments of poorer countries, predicted a month ago that this year would mark a historic step back for inequality, with the pandemic “likely to cause the first increase in global poverty since 1998.”

If you’ve spent any time stoking your curiosity with the UC Berkeley Library’s new online Digital Collections website, you’ve likely discovered all types of treasures digitized from the Library’s collections. The Library has already scanned and made available a virtual mountain of materials, from a photo of folk icon Joan Baez singing in front of Sproul Hall in 1964, to (almost) the entire run of the Daily Californian student newspaper.

The effort is part of the Library’s moonshot goal of wanting to make its estimated 200 million items from its special collections (rare books, manuscripts, photographs, archives, and ephemera) available online for the world to discover and use. But there’s a catch: Before institutions can reproduce materials and publish them online for worldwide access, they have to sort out complicated legal and ethical questions — ones that often stop libraries and other cultural heritage organizations from being able to move forward in setting these treasures free.

The good news? It just got easier to navigate these challenges, thanks to newly released responsible access workflows developed by the Library, which stand to benefit not only UC Berkeley’s digitization efforts, but also those of cultural heritage institutions such as museums, archives, and libraries throughout the nation.

The class of 2020 has become known as the class of COVID-19. Because of the coronavirus pandemic, these students have been forced to cut their college careers short, give up traditional graduation ceremonies and begin their professional careers during the most hostile labor market since the Great Depression.

Many have also had their first full-time job offers revoked.

“This is much worse than the Great Recession. Over the entire Great Recession I think maybe 8.5, 9 million jobs were lost over the course of a 5-year period. Between February and April, the United States lost 21.5 million payroll jobs,” says Gary Burtless, senior fellow at the Brookings Institution. “And so now, people graduating this spring are going to face the worst job market in the entire post-depression history.”

The institute, which will be part of the Tagliatela College of Engineering, will foster collaboration and interdisciplinary research across the University, effectively incorporating critical knowledge in cybersecurity, computing, data science, and AI across all programs.

There are still significant partisan and demographic divides, but they’re not so pronounced as to leave the two parties diametrically opposed. Rather, the gap is a matter of degrees, with a broad majority of Democrats and a more modest majority of Republicans offering support for masks and rejecting the idea that wearing them is a pointless practice or a sign of weakness.

Americans say, 62% to 29%, that deciding to wear a face mask is more a matter of public health than a matter of personal choice, the survey finds.

Online July 5-10. “Organizers of the 58th annual meeting of the Association for Computational Linguistics (ACL) on Sunday announced the list of accepted papers for the world-leading natural language processing (NLP) conference. A total of 779 papers (571 long papers and 208 short papers) made it out of 3,088 submissions, for an acceptance rate of 25.2 percent this year.”

Online June 19, starting at 8:30 a.m. PDT. “HILDA brings together researchers and practitioners to exchange ideas and results on human-data interaction. It explores how data management and analysis can be made more effective when taking into account the people who design and build these processes as well as those who are impacted by their results.” [save the date]

If you are newly working or going to school from home, the remote approach can be a big shift in how to get things done. Firefox has a number of browser extensions that might help make the #WFH transition a little easier and more productive.

Roughly a month after first announcing their unprecedented collaboration, Apple and Google have updated their devices’ operating systems today with the first component of their contact tracing API.

Referred to by the companies as “Exposure Notifications,” the technology aims to help public health agencies deploy apps that tell individuals when they may have been exposed to another person with COVID-19.

Device owners must opt in to enable the functionality, which according to the companies does not collect location data.

What’s so great about working groups? For one, when they work well, they can have a transformative effect on science. They can open up new research fields while shifting energy away from older ones. They can show connections among subdisciplines and establish high-level consensus among many individual studies. Working groups can also have lasting effects on participants, creating long-term collaborative networks and providing researchers a window into how others approach science. The average working group paper accumulates citations 2-5x faster than non-working group papers (Wei et al. 2020). The catering is also usually pretty solid.

So what’s the downside? A recent analysis of NSERC-funded researchers showed that, in terms of career progression, the benefits of working groups are not equally distributed among all participants in ways you can probably predict.

Biomedical researchers are racing to understand the virus that causes COVID-19, to evaluate its spread, and to develop tests, treatments and vaccines.

Physicists are volunteering to assist in these efforts, using their skills in data analytics, machine learning, simulation, software, computing, hardware development and project management. And an organization called Science Responds is helping to match them with projects that need their support.