The OECD's 2011-2015, 33 country, 215,942-person study of computer skills paints a deceptively grim picture of the average level of computer proficiency around the world -- deceptive because it excludes over-65s, who research shows to be, on average, less proficient than the 16-65 cohort sampled.
Read the rest
Princeton University psych prof Susan Fiske published an open letter denouncing the practice of using social media to call out statistical errors in psychology research, describing the people who do this as "terrorists" and arguing that this was toxic because of the structure of social science scholarship, having an outsized effect on careers.
Read the rest
It's been nearly 20 years since the publication of Bellwether, Connie Willis's comic novel about scientists caught in the turmoil of bureaucratic fads. I had very fond memories of this book, though I hadn't read it in more than a decade, so I gave the DRM-free audiobook a whirl, and fell in love with it all over again.
Read the rest
Amara's Law states, "We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run" -- Robert Charles Wilson's gripping conspiracy novel The Affinities
brings the experience of that process to life.
People who claim to be real vampires and must drink others' blood for energy are hesitant to talk to psychological counselors and other helping professionals about their lives because they fear being "judged as being wicked or evil or viewed as being psychotic, delusional or having a psychological problem," according to a new scientific study. Read the rest
In an outstanding lecture at the London School of Economics, Macarthur "genius award" recipient Sendhil Mullainathan explains his research on the psychology of scarcity, a subject that he's also written an excellent book about.
Read the rest
Ben Ho is a behavioral economist who studies apologies. He presents an economic theory of apologies that predicts when apologies will change the outcome of disputes, and proposes policies to increase the frequency and sincerity of apologies. The best evidence for economics-driven apology policies are the laws that make doctors' apologies inadmissible in court; Ho cites research that claims that this leads to more physicians' apologies, which reduces patient grief and anger, and cuts down on malpractice suits.
Read the rest
Part of the problem with the Chelsea Manning situation is that it's spawned a lot of not-terribly-well-informed discussion about the roles and experiences of transgendered people in the military. There's a risk of this one big anecdote coming to represent the whole. Enter the Kinsey Institute — America's favorite source of sexuality science — which just got a grant to do actual research on the lives of transgender service members. Read the rest
Human emotions and social interaction have a lot to do with body language — how our faces express what we're thinking and feeling, how our gestures are read by other people, and how we invade (or retreat from) each other's personal space. In fact, those movements and behaviors are so important that, if you map them onto an otherwise completely non-human, non-animal form, we'll start interpreting it as engaging with us — even if that form is nothing more than a moving stick.
This video, clips from a study that was published in 2011 by computer scientists at the University of Calgary, shows what test subjects did and said when they were left alone in a room with a stick-like robot, and asked to just think out loud and interact with the robot in whatever ways felt natural. Some people made friends. Others tried to fight it. And a few tried to talk it out of wanting to fight them.
Video Link Read the rest
There's a fantastic long read up at Aeon Magazine about the science of child development and the ethics of running scientific experiments on vulnerable populations. Virginia Hughes goes to Romania to follow a long-term study comparing children placed in orphanages with children placed in foster homes. The catch: Scientists already know that foster homes are better for kids than institutions
. But that fact isn't well-known or accepted in Romania. So scientists had to ask — is it ethical to run an experiment involving kids when you already know the answer if there's a chance that it might help other kids in the future? Read the rest
The Prisoner's Dilemma is a basic part of game theory. Two prisoners are given the choice between informing on the other, or staying silent. They can't communicate with each other. The choices they make determine how many years in prison they both get.
This analogy/brain game is often used to demonstrate the ways that different people can work with or against each other in economic and social situations. Now, for the first time, scientists have done a study based on The Prisoner's Dilemma that used real prisoners. Instead of time off their sentences, they were given the choice of competing or cooperating to earn goodies like coffee and cigarettes.
And here's the surprise: Compared to college students, the prisoners actually cooperated with each other much more often. Read the rest
Really interesting little bit of social science
at Slate where Lisa Wade looks at the "OMG COLLEGE STUDENTS ARE JUST INDISCRIMINATELY BANGING EACH OTHER WHILE DRUNK ALL THE TIME" scare story, and finds a very different picture of what's happening in reality. The catch: In order to understand why both the pop narrative and the reality can co-exist, you have to break college students down by demographics. Some students really are engaging in what's come to be called "hook-up culture", but they tend to be the most privileged students — the ones whose wealth, race, and social status can better protect them from the consequences of mistakes, and who think about their college life (and future goals) in very different ways compared to less-privileged peers. Read the rest
Behind the unveiling of J.K. Rowling
as the author of a pseudonymously published detective novel lies a messy and not-terribly-precise science called "forensic stylistics". Using specially designed software — or, less often, just a trained eye — experts in the field try to match writing styles and discern the true authorship of disputed texts. But, even when they turn out to be right, as in the Rowling case, their findings are less than exact
. Linguist Ben Zimmer explores the field and its usefulness at The Wall Street Journal. Read the rest
At Download the Universe, i09 editor Annalee Newitz critiques a new e-book about epigenetics — the science of how environmental factors can influence genetic expression — and violence
. The book makes some pretty terrible (and non-scientific) insinuations about the idea of an inherent propensity towards violence and Newitz does a good job of both taking down the specific book and explaining the nuance behind a complicated topic. Read the rest
Last week, after giving myself an initial overview of the scientific research on how gun ownership and gun laws affect violent crime, I told you that it seems like there's not a solid consensus on this issue. At least not in the United States. Different studies, of different laws, in different places seem to produce a wide variety of results.
On the one hand, this is kind of to be expected with social science. People are hard to pin down. Harder, often, than the Higgs Boson particle. And you can't just do a clean, controlled laboratory study of these issues. Instead, you're left trying to compare specific places, laws, and enforcement techniques that may not be easily comparable, in an attempt to draw a broad conclusion. That's hard.
But, it seems, the National Rifle Association has gone out of its way to make this work even more difficult than it would otherwise be. Since the early 1990s, NRA-backed politicians have attacked firearms research they believe is biased against guns. Alex Seitz-Wald at Salon.com wrote a piece on this back in July, after an earlier mass shooting. He describes how a vaguely worded clause has lead researchers to avoid doing firearms studies at all, for fear of losing their funding.
Read the rest
The Centers for Disease Control funds research into the causes of death in the United States, including firearms — or at least it used to. In 1996, after various studies funded by the agency found that guns can be dangerous, the gun lobby mobilized to punish the agency.
Did you know that there was a major American election on Tuesday? Great. Let us all never speak of it again. At least for the next 3.5 years.
But before we send the parts of our brains that care about politics off to recuperate at a nice imaginary spa, take a quick look at a page of election maps put together by University of Michigan physics professor Mark Newman. He studies complex systems, including the networks of human relationships and decision-making that go into election results. His page of maps shows several different ways to visualize the same 2012 presidential election data — methods which provide different pieces of context that you don't normally see in the simple state-by-state map.
The basic map — the one you see on TV and in the newspaper — doesn't really tell you the whole story. It gives you no idea of population density (a factor that obviously matters a lot in tallying the popular vote), and it only shows the winning party in each state. In reality, the vote is seldom all-Democrat or all-Republican. There's a gradient, no matter where you live.
The map above takes both those factors into account — distorting the country to make the more populous parts larger, and showing split turnouts in shades of purple.
See all Mark Newman's maps at his website
And here's his FAQ
Thanks, Rick Musser! Read the rest
The Mary Sue's Becky Chambers rounds up the coverage and analysis of an anti-trolling/griefing experiment in League of Legends, a massively multiplayer online RPG battle arena. League's management hired a team of social scientists who designed a system of peer-rewards that allowed users to hand each other publicly visible points for positive, friendly interactions (there was already a system of reporting bad behavior and meting out punishments, but it wasn't working very well). Unlike previous attempts to use public reward to improve behavior, this one was not yet turned into a back-scratching system where friends just vote one another up, and has reportedly resulted in massive improvements in the quality of group interactions.
Read the rest
Ten days after Honor went live, an update from Dr. Lyte appeared on the official LoL blog, detailing the global changes they’d noticed in reported bad behavior:
Negative Attitude reports: -29% in normals and -11% in ranked
Offensive Language reports: -35% in normals and -20% in ranked
Verbal Abuse reports: -41% in normals -17% in ranked
Check that out. Ten days of a voluntary system that grants nothing more than a tiny perk for being amiable, and folks were already cleaning up their acts. Of course, these stats only show a decline of reported incidents, which, while encouraging, is could be different than how things look down in the trenches. As LoL is not part of my repertoire, I took to Twitter earlier this week to get the word on the street. Lo and behold, players are indeed noticing a difference.