2006.14140

Analyses of urban scaling laws assume that observations in different cities are independent of the existence of nearby cities. Here we introduce generative models and data-analysis methods that overcome this limitation by modelling explicitly the effect of interactions between individuals at different locations. Parameters that describe the scaling law and the spatial interactions are inferred from data simultaneously, allowing for rigorous (Bayesian) model comparison and overcoming the problem of defining the boundaries of urban regions. Results in five different datasets show that including spatial interactions typically leads to better models and a change in the exponent of the scaling law. Data and codes are provided in Ref. [1].

Spatial interactions in urban scaling laws

2006.14493

We develop a principled methodology to infer assortative communities in networks based on a nonparametric Bayesian formulation of the planted partition model. We show that this approach succeeds in finding statistically significant assortative modules in networks, unlike alternatives such as modularity maximization, which systematically overfits both in artificial as well as in empirical examples. In addition, we show that our method is not subject to a resolution limit, and can uncover an arbitrarily large number of communities, as long as there is statistical evidence for them. Our formulation is amenable to model selection procedures, which allow us to compare it to more general approaches based on the stochastic block model, and in this way reveal whether assortativity is in fact the dominating large-scale mixing pattern. We perform this comparison with several empirical networks, and identify numerous cases where the network's assortativity is exaggerated by traditional community detection methods, and we show how a more faithful degree of assortativity can be identified.

Statistical inference of assortative community structures

2006.14496

Communication networks, power grids, and transportation networks are all examples of networks whose performance depends on reliable connectivity of their underlying network components even in the presence of usual network dynamics due to mobility, node or edge failures, and varying traffic loads. Percolation theory quantifies the threshold value of a local control parameter such as a node occupation (resp., deletion) probability or an edge activation (resp., removal) probability above (resp., below) which there exists a giant connected component (GCC), a connected component comprising of a number of occupied nodes and active edges whose size is proportional to the size of the network itself. Any pair of occupied nodes in the GCC is connected via at least one path comprised of active edges and occupied nodes. The mere existence of the GCC itself does not guarantee that the long-range connectivity would be robust, e.g., to random link or node failures due to network dynamics. In this paper, we explore new percolation thresholds that guarantee not only spanning network connectivity, but also robustness. We define and analyze four measures of robust network connectivity, explore their interrelationships, and numerically evaluate the respective robust percolation thresholds for the 2D square lattice.

Percolation Thresholds for Robust Network Connectivity

The estimate comes from a nationwide look at antibody tests.

CDC says COVID-19 cases in U.S. may be 10 times higher than reported

With social distancing and shelter-in-place mandates in effect worldwide, the COVID-19 pandemic is necessitating large-scale behavior change and taking a significant psychological toll. How can we use insights from the behavioral and social sciences to align human behavior with the recommendations of epidemiologists and public health experts, and guide decision-making? Northwestern University professor of political science Dr. Jamie Druckman and University of Cambridge social psychology professor, Dr. Sander van der Linden address these questions and more, drawing on key insights from a just-released report they coauthored with more than two dozen social scientists from universities around the world.

A New Normal: How Social and Behavioral Science Can Help Us Cope With COVID-19

Jay Van Bavel shares the presentation he made to the World Health Organization (WHO). Dr. Van Bavel is an associate professor of Psychology & Neural Science at New York University. He completed his PhD at the University of Toronto and a postdoctoral fellowship at The Ohio State University. Jay has published 75 academic papers and written research essays in The New York Times, Scientific American, Wall Street Journal, Quartz, and the Washington Post. He also writes a mentoring column, entitled Letters to Young Scientists, for Science Magazine. Dr. Van Bavel is the recipient of the American Psychological Foundation’s 2015 Visionary Grant and F. J. McGuigan Early Career Investigator Research Grant on Understanding the Human Mind.

Using Social and Behavioral Science to Support COVID 19 Pandemic Response with Dr. Jay Van Bavel

On Thursday May 21, at 11:00 am The Global Policy Institute and Bay Atlantic University held an event via Zoom titled: The Impact of COVID-19 on Global Politics and International Institutions.

Webinar: The Impact of COVID-19 on Global Politics and International Institutions

It is commonly thought that morality applies universally to all human beings as moral targets, and our general moral obligations to people will not, as a rule, be affected by their views. I propose and explore a radical, alternative normative moral theory, ‘Designer Ethics’, according to which our views are pro tanto crucial determinants of how, morally, we ought to be treated. For example, since utilitarians are more sympathetic to the idea that human beings may be sacrificed for the greater good, perhaps it is permissible (or, even under certain conditions, obligatory) to give them ‘priority’ as potential victims. This odd idea has manifold drawbacks but I claim that it also has substantial advantages, that it has some affinities to more commonly accepted moral positions, and that it should be given a significant role in our ethical thinking.

10.1093/pq/pqaa003

Should We Sacrifice the Utilitarians First?

Loneliness may be a fundamental part of the human condition, but scientists have only recently begun exploring its causes, consequences, and potential interventions. A special section in Perspectives on Psychological Science, a journal of the Association for Psychological Science, aims to bring these strands of inquiry together, presenting a series of articles that review the current state of scientific research on loneliness. The section, edited by psychological scientist David Sbarra of the University of Arizona, investigates loneliness across multiple levels, from evolutionary theory to genetics to social epidemiology. “As a group, these articles set the bar high for future research on loneliness,” Sbarra writes in his introduction to the special section. “At the same time, they also contain ‘something for everyone’ — they are accessible, thought-provoking ideas that can be tackled from many different perspectives.”

Understanding Loneliness Through Science

Have courage. Be kind. Keep trying. We can do this — we have to do this — for them. END

4) In the end I think of the lesson from the bunnies. We have the keys to make this a better world. Those keys are logic, science, reason, and data. But the more important ones are: kindness, communication, open hearts and courage.

… but I can’t look at my students, or my kids, or the young people I mentor, and feel anything but an obligation to do what I can for them. No matter how tired or cynical I feel.

3) The next generation needs us. They are amazing and hopeful and inspirational and trying so hard to make something of the shit sandwich they’ve been given. But they can’t do it alone. I’m old and cynical and exhausted…

2) The world seems to be getting harder. There’s no denying that. Hard things suck, but they can be cleansing too. I have had *such* a hard time lately, and it’s still hard. But it’s made me better: I know who I am, and I know how to be that person.

1) There IS no science/art divide. Both are so important. Math, coding, informed data analysis, proper scientific method — these are keys to understanding. But art — stories, like the bunnies — that is what gives it all heart and meaning.

… I like to think *we* learned to rely on each other and to do the same. Even online, I felt like we were all pulling each other through this difficult time. And, looking back, I see a lot of lessons for myself that I want to remember.

And through all these terrible months, these students (and the bunny story) kept me going. Just as the bunnies learned to rely on each other and use data and science and reasoning to solve their problems…

… careers in data science, in analysis, in psychology, in all kinds of social science. I’ve told them I’m so proud of them but I don’t think they grasp HOW proud of them I am and how completely impressive they all are.

And in this mastery, they’ve found confidence. Some have realised that they actually DO like math and coding. Some have realised that they can do really hard things. Many now have the foundations for all sorts of careers…

THEY STEPPED UP, ya’ll. These students, who three months ago knew no stats and no coding, somehow in the middle of a pandemic managed to master this stuff at a level that will open major doors for them in the future.

… for an inability to differentiate students. I had to make the exam difficult for this reason, and I was afraid they’d hate me or not be able to do it. We’re still in the middle of marking it but neither of those things seem to have happened.

Some students struggled, and I have so much respect for them, because they KEPT TRYING. Most of them ended up getting it, or at least getting most of it. On the first assignment, which was not easy, the grades were so good I worried about getting in trouble…

Some students went so far beyond expectations: finding other datasets on the web and playing with them for fun; asking highly technical questions about the central limit theorem and the foundations of statistics; asking for more math, more resources.

And yet, OMG, they persevered. They believed me (or at least tried to believe me) when I told them they could do it. They didn’t let their fear stop them from learning more about R and stats in a few months than many people had told me they’d be capable of.

Plus they were taking my subject — a subject most of them feared and thought they’d loathe. A subject that is really quite difficult.

As hard as this semester was for me, it was also really hard for my students, if not harder. Many didn’t know what country they’d be living in. Many couldn’t support themselves anymore, or had terrible living situations, or no childcare.

I’ll tell you. The hope is the students. The hope is this next generation we have coming up. They are spectacular, y’all, and we fucking OWE it to them to step up and fix this shitshow of a world we’re leaving to them.

Where’s the hope? You may be asking at this point. So far this seems terrible. Where’s the grace?

My subject was better adapted to online learning than many — we already did flipped classroom, for instance — but it was still a huge source of stress and uncertainty and extra work for myself, the tutors, and the students.

Then Covid hit and it went from difficult to feeling at times impossible. We have two small children, ages 4 and 7, and while my partner has been fantastic at taking the lion’s share of the unexpected child-rearing, it still hit me too.

(And of course in addition to this subject I also had a lab to run, honours/PhD/etc students to support, grants to administer and write, and colleagues who depended on me to work on joint projects.)

The bunny story added to the workload even more but I couldn’t not do it. Anyway, all of that is to say that even before Covid, I knew it was going to be a tough semester: SO much work on top of the substantial personal stuff I’ve been dealing with.

.. and over the semester they would learn to use data and come together to solve their problems and find a way forward. I was afraid the students would find it a bit cringey, but at the same time I thought it sounded so fun I had to give it a try.

While planning the subject I came up with the ridiculous idea of wrapping the entire semester around an ongoing story centred around my children’s stuffed animals. It would be a tale of bunnies facing food shortages, terror of outsiders, fear of the unknown…

Still, I asked for this. I love teaching SO much, and I love teaching coding and math more than any other kind. Coding and math are my happy place — meditative, soothing — and I love the challenge of getting people who fear and loathe it to see some of its beauty.

In March I started teaching an #rstats class at the intro level to almost 700 psych undergrads. It was my first time teaching it, which meant spending months on an insane sprint creating 25 lectures, weekly tutorials, assessments, and answering zillions of emails.

I’ve been having a difficult time lately — partly because of [insert frantic gesturing at the state of the world], partly personal — but one thing has been a real bright light for me in the last few months. I think it has some broader lessons that might give some hope, so THREAD

Coronavirus cases rising in Europe following eased lockdowns, says WHO

Covid-19 news: Cases rising in Europe following eased lockdowns

In court, the jury is told that DNA found at the crime scene is very likely to have come from the suspect. How good is this as evidence? A couple have two children, both of whom die of cot death. Is this such an unlikely coincidence that there must be some other explanation? In many countries, people are routinely tested for certain cancers, including bowel, breast, prostate and cervical cancers. But does a positive test mean you definitely have cancer? Whether evidence is used in a court of law or in a diagnostic test, or in a variety of other everyday situations, we need to understand how probability and statistics can help us to evaluate that evidence. This video clip, featuring Professor Philip Dawid, is part of a set of free online mathematics resources designed to help school students become informed citizens, who can understand that unless we ask the right questions, we won't do the right things with the numbers.

Misuse of Probability

You may think that statistical causal inference is about inferring causation. You may think that it can not be tackled with standard statistical tools, but requires additional structure, such as counterfactual reasoning, potential responses or graphical representations.  I shall try to disabuse you of such woolly misconceptions by locating statistical causality firmly within the scope of traditional statistical decision theory.  From this viewpoint, the enterprise of "statistical causality" could fruitfully be rebranded as "assisted decision making".

Causal inference isn't what you think it is

The reduction of the 2-metre rule in England tells people we’re out of the woods, and they’ll behave accordingly

The way Boris Johnson has eased lockdown sends all the wrong messages

“It wasn’t diverse enough, maybe it wasn’t nimble enough” Sir Paul Nurse tells @Peston that SAGE needs to be truly multidisciplinary to perform its duties. #Peston

As case counts surge, we look at regional and state-level numbers to find out which recent jumps in COVID-19 case counts are likely to be explained by increased testing, and which are not. For the states with the worst recent numbers, the news is not good.

It's Not Just Testing

In the midst of a global pandemic, @BU_Tweets has just threatened to pull the medical insurance of any graduate student unwilling to teach in the classroom this fall.

1805.08838

Unsupervised learning is widely recognized as one of the most important challenges facing machine learning nowa- days. However, in spite of hundreds of papers on the topic being published every year, current theoretical understanding and practical implementations of such tasks, in particular of clustering, is very rudimentary. This note focuses on clustering. I claim that the most signif- icant challenge for clustering is model selection. In contrast with other common computational tasks, for clustering, dif- ferent algorithms often yield drastically different outcomes. Therefore, the choice of a clustering algorithm, and their pa- rameters (like the number of clusters) may play a crucial role in the usefulness of an output clustering solution. However, currently there exists no methodical guidance for clustering tool-selection for a given clustering task. Practitioners pick the algorithms they use without awareness to the implications of their choices and the vast majority of theory of clustering papers focus on providing savings to the resources needed to solve optimization problems that arise from picking some concrete clustering objective. Saving that pale in com- parison to the costs of mismatch between those objectives and the intended use of clustering results. I argue the severity of this problem and describe some recent proposals aiming to address this crucial lacuna.

Clustering - What Both Theoreticians and Practitioners are Doing Wrong

In the current system of pre-publication peer review, which papers are scrutinized most thoroughly? In this blog post, I argue that scrutiny is mis-allocated: the average number of pre-publication reviews (and reviewer effort) a manuscript has accumulated by the time it is published is lower for papers published very visibly (e.g., in high-impact journals).

Mis-allocated scrutiny

classical economics, like philosophy, is a theoretical field. It considers behavior under abstract, idealized conditions. Those conditions, and even the behavior being theorized about, need never have been shown to exist.   In this way, classical economics differs from neuroscience and other experimental disciplines. In neuro-economics, a branch of neuroscience, we model how the brain makes decisions. Models are evaluated by their ability to predict real behavior, as measured through experiments, rather than behavior that should be true because it makes sense.

Policy shouldn't rely on economic theory, but on data about actual human behavior

A second wave of coronavirus infections in the UK is a “real risk” and all political parties should work together to ensure the country is ready for it, warned a group of health leaders including presidents of the Royal College of Physicians, Surgeons, GPs and Nursing and the chair of the British Medical Association. In a letter addressed to leaders of UK political parties published on the British Medical Journal’s website, they say, “the available evidence indicates that local flare-ups are increasingly likely and a second wave a real risk. Many elements of the infrastructure needed to contain the virus are beginning to be put in place, but substantial challenges remain.”

Covid-19 news: UK health leaders warn of 'real risk' of a second wave

Since the start of the new coronavirus outbreak, our researchers have been working hard to understand the epidemic. From research on the spread of the virus and number of people infected to the impact of measures taken to prevent the spread of the disease, the task is huge. Scientists from around the world are working together to better understand the virus and its spread. Epidemiologists, immunologists, virologists, phylogeneticists and healthcare professionals are all working at lightning speed on different bits of the puzzle. How do we bring it all together to understand what we’re dealing with and what we should do about it? Mathematical modelling helps to understand patterns in data and inform the outbreak response. But how does it work? How has our understanding of COVID-19 developed? How are we using what we’ve found to inform big decisions? Dr Katy Gaythorpe, Dr Juliette Unwin and Dr Lilith Wittles will be answering your questions about our research and how their work has changed since the outbreak started. This is also an opportunity for them to understand your thoughts on their work.

Q&A: Modelling COVID-19

10.31234/osf.io/epsqt

Meta-analysis and structural equation modeling (SEM) are two popular statistical models in the social, behavioral, and management sciences. Meta-analysis summarizes research findings to provide an estimate of the average effect and its heterogeneity. Moderators may be used to explain the heterogeneity in the data. On the other hand, SEM includes several special cases, including the general linear model, path model, and confirmatory factor analytic model. SEM allows researchers to test hypothetical models with empirical data. This article introduces meta-analytic structural equation modeling (MASEM) that combines the advantages of both meta-analysis and SEM. There are usually two stages in the analyses. In the first stage of analysis, a pool of correlation matrices is combined to form an average correlation matrix. In the second stage of analysis, proposed structural equation models are tested against the average correlation matrix. MASEM enables researchers to synthesize researching findings using SEM as the research tool in primary studies. Several popular approaches to conduct MASEM are introduced. These include, for instance, the univariate-r, generalized least squares, two-stage SEM, one-stage MASEM. A real example is used to illustrate how to apply MASEM in empirical research. Extensions of MASEM to other applications are briefly discussed.

Meta-Analytic Structural Equation Modeling

The idea behind Open Science is to allow scientific information, data and outputs to be more widely accessible (Open Access) and more reliably harnessed (Open Data) with the active engagement of all the stakeholders (Open to Society). By encouraging science to be more connected to societal needs and by promoting equal opportunities for all (scientists, policy-makers and citizens), Open Science can be a true game changer in bridging the science, technology and innovation gaps between and within countries and fulfilling the human right to science.

Open Science

10.1080/15534510.2020.1783758

Editorial

To further enrich the platform that social influence provides, we will introduce two types of registered reports. In the classical registered report, we seek contributions from scholars who want to ensure the veracity of the findings, irrespective of whether the results ultimately support (or fail to support) their initial hypotheses (Chambers, 2019 Chambers, C. (2019). The registered reports revolution. Lessons in cultural reform. Significance , 16(4), 23–27. https://doi.org/10.1111/j.1740-9713.2019.01299.x  [Crossref], [Google Scholar]). In this classical registered report, we will follow the traditional guidelines where researchers need to explain the relation between expected effect size, statistical power, and needed sample size. In the proof of concept registered report, we seek contributions from scholars who want to demonstrate the feasibility of a research idea and convince fellow researchers of the importance to collaboratively collect data in the future.

10.31234/osf.io/zvs9f

Increases in the incidence of psychological distress and alcohol use during the COVID-19 pandemic have been predicted. Environmental reward and self-medication theories suggest that increased distress and greater social/environmental constraints during COVID-19 could result in increases in depression and drinking to cope with negative affect. The current study had two goals: (1) to clarify the presence and direction of changes in alcohol use and related outcomes after the introduction of COVID-19 social distancing requirements, and; (2) to test hypothesized mediation models to explain individual differences in alcohol use during the early weeks of the COVID-19 pandemic. Participants (n = 1127) were U.S. residents recruited for participation in an online survey. The survey included questions assessing environmental reward, depression, COVID-19-related distress, drinking motives, and alcohol use outcomes (alcohol use; drinking motives; alcohol demand, and solitary drinking). Outcomes were assessed for two timeframes: the 30 days prior to state-mandated social distancing (‘pre-social-distancing’), and the 30 days after the start of state-mandated social distancing (‘post-social-distancing’). Depression severity, coping motives, and frequency of solitary drinking were significantly greater post-social-distancing relative to pre-social-distancing. Conversely, environmental reward and other drinking motives (social, enhancement, and conformity) were significantly lower post-social distancing compared to pre-social-distancing. Time spent drinking and frequency of binge drinking were greater post-social-distancing compared to pre-social-distancing, whereas typical alcohol quantity/frequency were not significantly different between timeframes. Indices of alcohol demand were variable with regard to change. Mediation analyses suggested a significant indirect effects of reduced environmental reward with drinking quantity/frequency via increased depressive symptoms and coping motives, and a significant indirect effect of COVID-related distress with alcohol quantity/frequency via coping motives for drinking. Results provide early evidence regarding the relation of psychological distress with alcohol consumption and coping motives during the early weeks of the COVID-19 pandemic. Moreover, results largely converged with predictions from self-medication and environmental reinforcement theories. Future research will be needed to study prospective associations among these outcomes.

Depression, Environmental Reward, Coping Motives and Alcohol Consumption During the COVID-19 Pandemic

Session 3, speaker 4

James Evans - Designing Diversity for Collective Advance (ACM CI'20)

Session 3, speaker 3

Jeff Howe - Crowdsourcing and the Crisis: Collective Intelligence in the Age of Covid-19 (ACM CI'20)

Session 2, speaker 3

Thomas W. Malone - COVID-19 and Collective Intelligence (ACM CI'20)

COVID-19 has an astonishing association with age. Over 40 years, and the risks of critical illness and death rise progressively. But under 40 years, serious disease is rare. The government should, my friend suggested, encourage everyone under 40 years to go out, get back to work, and, if at all possible, do all they could to become infected. Infection would bring with it immunity (he agreed this was one area of uncertainty, although we have no reason to believe that immunity would not follow infection). Immunologist friends had advised him that natural immunity would be stronger than the immunity induced by vaccination. Once immune, this cohort of under-40-year-olds would no longer be at risk of infection and they would no longer be a risk to others.

10.1016/S0140-6736(20)31416-1

Offline: A novel solution to live with coronavirus

10.31234/osf.io/u4cns

Meta-analysis is an important quantitative tool for cumulative science, but its application is frustrated by publication bias. In order to test and adjust for publication bias, we extend model-averaged Bayesian meta-analysis with selection models. The resulting Robust Bayesian Meta-analysis (RoBMA) methodology does not require all-or-none decisions about the presence of publication bias, can quantify evidence in favor of the absence of publication bias, and performs well under high heterogeneity. By model-averaging over a set of 12 models, RoBMA is relatively robust to model misspecification and simulations show that it outperforms existing methods. We demonstrate that RoBMA finds evidence for the absence of publication bias in Registered Replication Reports and reliably avoids false positives. We provide an implementation in R and JASP so that researchers can easily apply the new methodology to their data.

Robust Bayesian Meta-Analysis: Addressing Publication Bias with Model-Averaging

India is emerging from the world's largest lockdown in its fight against COVID-19. What challenges lie ahead for the country's 1.3 billion residents? How effective has the response from India's government been and what are the short and longer-term impacts of COVID-19 for India's citizens and its economy?Join us on Thursday, June 25, 2020, for a King Center on Global Development special virtual event with Stanford Seed. You will hear from experts as they analyze the Indian government’s initial response to COVID-19, as well as how the virus is impacting public programs, businesses and workers, and the health care sector. Yamini Aiyar, president and chief executive of the Centre for Policy Research, Naushad Forbes, co-chairman of Forbes Marshall, and Stephen Luby, professor of medicine (infectious diseases), will share their expertise in a conversation moderated by Saumitra Jha, associate professor of political economy.

COVID-19: The Impact in India

Despite various levels of preventive measures, in 2020 many countries have suffered severely from the coronavirus 2019 (COVID-19) pandemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus. We show that population heterogeneity can significantly impact disease-induced immunity as the proportion infected in groups with the highest contact rates is greater than in groups with low contact rates. We estimate that if R0 = 2.5 in an age-structured community with mixing rates fitted to social activity then the disease-induced herd immunity level can be around 43%, which is substantially less than the classical herd immunity level of 60% obtained through homogeneous immunization of the population. Our estimates should be interpreted as an illustration of how population heterogeneity affects herd immunity, rather than an exact value or even a best estimate.

10.1126/science.abc6810

A mathematical model reveals the influence of population heterogeneity on herd immunity to SARS-CoV-2

The mitigation measures enacted as part of the response to the unfolding COVID-19 pandemic are unprecedented in their breadth and societal burden. A major challenge in this situation is to quantitatively assess the impact of non-pharmaceutical interventions like mobility restrictions and social distancing, to better understand the ensuing reduction of mobility flows, individual mobility changes, and impact on contact patterns. Here, we report preliminary results on tackling the above challenges by using de-identified, large-scale data from a location intelligence company, Cuebiq, that has instrumented smartphone apps with high-accuracy location-data collection software. Italy has been severely affected by the COVID-19 pandemic. Following the identification of the first infections, on February 21, 2020, national authorities have put in place an increasing number of restrictions aimed at containing the outbreak and delaying the epidemic peak. On March 12, the government imposed a national lockdown. On May 4, the lockdown has been lifted and since May 18 most of business acitivities, especially in the retail sector, have restarted. Since June 3, restrictions on the inter-regional mobility have been completely removed, thus restoring the freedom of movement across the country in full. Here, we analyze how mobility and proximity patterns in Italy have changed during the week of June 6-12, at the sub-national scale.

Mobility in Italy after the complete removal of travel restrictions

Conditions like chronic fatigue syndrome have been linked to viral infections, so it’s possible that the covid-19 virus may go on to trigger similar conditions

Could the coronavirus trigger post-viral fatigue syndromes?