Group items matching

in title, tags, annotations or url

University students, like many in society, demonstrate bias against women and particularly women from non-English speaking backgrounds. That's the take home message from a new and comprehensive analysis of student experience surveys. The study examined a large dataset consisting of more than 500,000 student responses collected over 2010 to 2016. It involved more than 3,000 teachers and 2,000 courses across five faculties at the University of New South Wales (UNSW), Sydney.

Podcast from UTS Think:Digital Futures (that I briefly appear on). Sometimes we get caught up thinking data and science are the be all and end all - it can give us a lot of answers sure, but the devil is in the detail. How are we interpreting data wrong? And why do we have trouble communicating it? Presenters/Producers: Cheyne Anderson & Ellen Leabeater Speakers: Jon Wardle - Faculty of Health, University of Technology Sydney Simon Knight - Lecturer, Connected Intelligence Centre, UTS Mark Moritz - Associate Professor, Department of Anthropology, Ohio State University Think: Digital Futures is supported by 2SER and the University of Technology Sydney. http://2ser.com/shows/think-digital-futures/

This week, Hannah Devlin speaks with some of the statistical vigilantes who are scouring datasets to identify cases of fraud and poor scientific practice. These include the consultant anaesthetist John Carlisle, from Torbay Hospital in Devon, who details his role in the Fujii scandal. Hannah also speaks to a PhD student from Tilburg University in the Netherlands, Michèle Nuijten, about software she has helped develop to "spell-check" statistics found in psychology papers. And finally, we hear from the University of Cambridge's Winton professor for the public understanding of risk, David Spiegelhalter, who is also president of the Royal Statistical Society, about the dangers of statistical malpractice.

The New York Times has run a couple of pieces comparing university tuition fees in America with the US, UK, and other systems around the world. These pieces look at the advantages and disadvantages of the different systems (with Australia coming out as having some clear benefits)

Discusses a recent paper and press-release (a short summary circulated to the media by the people involved. In research, this is generally a summary of the paper often published by the university or organisation that has conducted the research). Do you think starting to eat nuts is likely to decrease your risk of heart disease? What other behaviours or characteristics do you think nut-eating might be associated with? Would any of them also be related to health factors?

"we pushed back against a misleading news release from Yale University that ran with the headline, "Yale researchers discover underlying cause of myeloma." The release claimed that the researchers identified "what causes a third of all myelomas," describing a faulty immune system response to compounds known as lysolipids as the culprit. But that characterization was wrong. First off, the research only demonstrated an association between lysolipids and this cancer - they didn't prove that one caused the other. And far from applying to a third of myeloma patients, the findings applied only to a tiny group of patients "

Interesting discussion of data on vote recounts and using electronic or hand counting methods (in America where they use electronic voting machines quite commonly). These numbers represent three main kinds of disputes, Foley told me. First, candidates (and their lawyers) argue over what ballots should be counted and which should be thrown out as ineligible. Then, they argue over which candidate specific ballots should count for. Finally, they argue over whether all the eligible votes were counted correctly - the actual recount. Humans are much better than machines at making decisions around the first two kinds of ambiguous disputes, Stewart said, but evidence suggests that the computers are better at counting. Michael Byrne, a psychology professor at Rice University who studies human-computer interaction, agreed. "That's kind of what they're for," he said.

In August 2016, the Australian government released an "anonymised" data set comprising the medical billing records, including every prescription and surgery, of 2.9 million people. Names and other identifying features were removed from the records in an effort to protect individuals' privacy, but a research team from the University of Melbourne soon discovered that it was simple to re-identify people, and learn about their entire medical history without their consent, by comparing the dataset to other publicly available information, such as reports of celebrities having babies or athletes having surgeries.

For decades, researchers and physicians tended to think about pregnancies as either planned or unplanned. But new data reveals that for a significant group of women, their feelings don't neatly fit into one category or another. As many as one-fifth of women who become pregnant aren't sure whether they want a baby. This fact may reshape how doctors and policymakers think about family planning. For women who are unsure, it doesn't seem enough for physicians to counsel them on pregnancy prevention or prenatal care. "In the past we thought of it as binary, you want to be pregnant or not, so you need contraception or a prenatal vitamin," said Maria Isabel Rodriguez, an obstetrician-gynecologist at Oregon Health and Science University whose research focuses on family planning and contraceptive policy. "But it's more of a continuum." The new data comes from a recent change in the Centers for Disease Control and Prevention's big survey of new mothers, now allowing them to answer a question about their pregnancy desires by saying "I wasn't sure." It shows that some women want to avoid making a decision about becoming pregnant, or have strong but mixed feelings about it.

Most of the poor people in the world are women. In no country on earth are women economically equal to men, and Australia is no exception. Research from Acoss and the University of New South Wales last year showed that a higher share of people living in poverty in Australia are women. The experience of living below the breadline in our very wealthy nation is a gendered one, for reasons that are complex and intertwined. As women progress through life, they encounter a series of barriers and setbacks that simply do not encumber men in the same way. The cause of gendered poverty is structural. It is entrenched in our workplace settings, and embedded in our personal relationships. It is at play at every stage of a woman's life, from childhood to the grave, making its mark on our education, our employment, our homes, our familial responsibilities and our retirement options. At its heart is the simple fact that women do the lion's share of caring for others. Caring is women's work, and our society does not value women's work.

Scientists have typically justified excluding female animals from experiments - even when studying conditions that are more likely to affect women - on the basis that fluctuating hormones would render the results uninterpretable. However, according to Rebecca Shansky, a neuroscientist at Northeastern University, in Boston, it is entirely unjustified by scientific evidence, which shows that, if anything, the hormones and behaviour of male rodents are less stable than those of females. Shansky is calling for stricter requirements to include animals of both sexes in research, saying the failure to do so has led to the development of drugs that work less well in women.

Interesting discussion of a contentious issue related to the idea of a 'basic income' (but here, in a development context). Martin Ravallion (former Chief Economist of the World Bank, now at CGD) published a useful paper this week asking exactly this question. As he says, there's no simple answer - which is why the question is so interesting. Both 'the right to work' and 'the right to income' aim to secure a more fundamental right: freedom from poverty. Workfare has a long history, notably in India, where the National Rural Employment Guarantee Act (NREGA) guarantees (in theory) up to 100 days work per year, paid at a minimum wage, to anyone who requests it. Cash transfers (often with conditions) have expanded enormously in recent years, while the hot topic of Universal Basic Income (UBI) has advocates across the political spectrum. Which of these approaches is most cost-effective? Ravallion sets out the arguments clearly.

Numbers are often used to persuade rather than inform, statistical literacy needs to be improved, and so surely we need more statistics courses in schools and universities? Well, yes, but this should not mean more of the same. After years of researching and teaching statistical methods, I am not alone in concluding that the way in which we teach statistics can be counterproductive, with an overemphasis on mathematical foundations through probability theory, long lists of tests and formulae to apply, and toy problems involving, say, calculating the standard deviation of the weights of cod. The American Statistical Association's Guidelines for Assessment and Instruction in Statistics Education (2016) strongly recommended changing the pedagogy of statistics into one based on problemsolving, real-world examples, and with an emphasis on communication.

In this case, because pancreatic cancer is relatively rare, the impressive-sounding 50% reduction doesn't amount to very much. According to the American Cancer Society, a 60-year-old man has a 0.41% (1 in 241) chance of developing pancreatic cancer during the next ten years. (Risk varies greatly with age and is much lower at younger ages.) So cutting that risk in half might bring it down to about 0.2% (1 in 480). It's a 50% drop, sure, but the risk was already very small to begin with. In this case, it's more helpful to news and health care consumers to describe it as a 0.2 percentage point reduction. And then there are the harms of regular aspirin use; the Yale news release that the tweet links to doesn't mention any. But taking aspirin regularly isn't a harmless intervention - far from it. It's well known that taking aspirin every day can cause serious bleeding in the gastrointestinal system and, less frequently, in the brain. That's why guidelines for aspirin use in cardiovascular disease prevention don't recommend it for people at low risk of a heart attack. The potential benefits may be outweighed by the risks of a serious bleed.