Group items tagged

history of voting math comes mainly in two chunks: the period of the French Revolution, when some members of France’s Academy of Sciences tried to deduce a rational way of conducting elections, and the nineteen-fifties onward, when economists and game theorists set out to show that this was impossible

The first mathematical account of vote-splitting was given by Jean-Charles de Borda, a French mathematician and a naval hero of the American Revolutionary War. Borda concocted examples in which one knows the order in which each voter would rank the candidates in an election, and then showed how easily the will of the majority could be frustrated in an ordinary vote. Borda’s main suggestion was to require voters to rank candidates, rather than just choose one favorite, so that a winner could be calculated by counting points awarded according to the rankings. The key idea was to find a way of taking lower preferences, as well as first preferences, into account.Unfortunately, this method may fail to elect the majority’s favorite—it could, in theory, elect someone who was nobody’s favorite. It is also easy to manipulate by strategic voting.

If the candidate who is your second preference is a strong challenger to your first preference, you may be able to help your favorite by putting the challenger last. Borda’s response was to say that his system was intended only for honest men.

Assistant Professor Mark Cenite of Nanyang Technological University’s Wee Kim Wee School of Communication and Information said: ‘This approach allows users to moderate themselves, and the news site is seen as being sensitive to readers’ values.’

But Mr Alex Au, who runs socio-political blog Yawning Bread, cautioned that this could lead to astroturfing. The term, derived from a brand of fake grass, refers to a fake grassroots movement in which a group wishing to push its agenda sends out manipulated and replicated online messages in support of a certain policy or issue. His suggestion: user tiers, in which comments by users with verified identities are displayed visibly and anonymous comments less conspicuously. He said: ‘This approach does not bar people from speaking up, but weighs in by signalling the path towards responsible participation.’

what is astroturfing? It is when a few people do one or both of two things: create multiple identities for each of themselves and flood a forum or topic with similar opinions, or get their friends to post boilerplate letters (expressing similar opinions of course) even if they do not totally share them to the same degree. The intent is to create an impression that a certain opinion is more widely held than is actually the case.

rather than being forced to wait for parliament to meet to air their dissent, now opposition parties are able to post pre-emptively their criticisms online, shifting the time and space of Singapore's political debate

Singapore's People's Action Party (PAP)-dominated politics are increasingly being contested online and over social media like blogs, Facebook and Twitter. Pushed by the perceived pro-PAP bias of the mainstream media, Singapore's opposition parties are using various new media to communicate with voters and express dissenting views. Alternative news websites, including The Online Citizen and Temasek Review, have won strong followings by presenting more opposition views in their news mix.

Despite its democratic veneer, Singapore rates poorly in global press freedom rankings due to a deeply entrenched culture of self-censorship and a pro-state bias in the mainstream media. Reporters Without Borders, a France-based press freedom advocacy group, recently ranked Singapore 136th in its global press freedom rankings, scoring below repressive countries like Iraq and Zimbabwe. The country's main media publishing house, Singapore Press Holdings, is owned by the state and its board of directors is made up largely of PAP members or other government-linked executives. Senior newspaper editors, including at the Straits Times, must be vetted and approved by the PAP-led government.

being both a working scientist and a science writer gives me a unique perspective on science, scientific publications, and the significance of scientific work. The final disclosure should be that I have never published in any of the top rank physics journals or in Science, Nature, or PNAS. I don't believe I have an axe to grind about that, but I am also sure that you can ascribe some of my opinions to PNAS envy.

If you asked most scientists what their goals were, the answer would boil down to the generation of new knowledge. But, at some point, science and scientists have to interact with money and administrators, which has significant consequences for science. For instance, when trying to employ someone to do a job, you try to objectively decide if the skills set of the prospective employee matches that required to do the job. In science, the same question has to be asked—instead of being asked once per job interview, however, this question gets asked all the time.

Because science requires funding, and no one gets a lifetime dollop-o-cash to explore their favorite corner of the universe. So, the question gets broken down to "how competent is the scientist?" "Is the question they want to answer interesting?" "Do they have the resources to do what they say they will?" We will ignore the last question and focus on the first two.

As a verbal artifact, farmed content exhibits neither style nor substance.

The insultingly vacuous and frankly bizarre prose of the content farms — it seems ripped from Wikipedia and translated from the Romanian — cheapens all online information.

These prose-widgets are not hammered out by robots, surprisingly. But they are written by writers who work like robots. As recent accounts of life in these words-are-money mills make clear, some content-farm writers have deadlines as frequently as every 25 minutes. Others are expected to turn around reported pieces, containing interviews with several experts, in an hour. Some compose, edit, format and publish 10 articles in a single shift. Many with decades of experience in journalism work 70-hour weeks for salaries of $40,000 with no vacation time. The content farms have taken journalism hackwork to a whole new level.

In the realm of public policy, we live in an age of numbers.

do wehold an outsize belief in our ability to gauge complex phenomena, measure outcomes and come up with compelling numerical evidence? A well-known quotation usually attributed to Einstein is “Not everything that can be counted counts, and not everything that counts can be counted.” I’d amend it to a less eloquent, more prosaic statement: Unless we know how things are counted, we don’t know if it’s wise to count on the numbers.

The problem isn’t with statistical tests themselves but with what we do before and after we run them.

in the last several months, one name turned up, with uncanny regularity, in the No. 1 spot for each and every term: J. C. Penney. The company bested millions of sites — and not just in searches for dresses, bedding and area rugs. For months, it was consistently at or near the top in searches for “skinny jeans,” “home decor,” “comforter sets,” “furniture” and dozens of other words and phrases, from the blandly generic (“tablecloths”) to the strangely specific (“grommet top curtains”).

J. C. Penney even beat out the sites of manufacturers in searches for the products of those manufacturers. Type in “Samsonite carry on luggage,” for instance, and Penney for months was first on the list, ahead of Samsonite.com.

the digital age’s most mundane act, the Google search, often represents layer upon layer of intrigue. And the intrigue starts in the sprawling, subterranean world of “black hat” optimization, the dark art of raising the profile of a Web site with methods that Google considers tantamount to cheating.

Here’s another hypothesis, this one for the conspiracy-minded. Last year, Advertising Age obtained a Google document that listed some of its largest advertisers, including AT&T, eBay and yes, J. C. Penney. The company, this document said, spent $2.46 million a month on paid Google search ads — the kind you see next to organic results.

Is it possible that Google was willing to countenance an extensive black-hat campaign because it helped one of its larger advertisers? It’s the sort of question that European Union officials are now studying in an investigation of possible antitrust abuses by Google.

Investigators have been asking advertisers in Europe questions like this: “Please explain whether and, if yes, to what extent your advertising spending with Google has ever had an influence on your ranking in Google’s natural search.” And: “Has Google ever mentioned to you that increasing your advertising spending could improve your ranking in Google’s natural search?”

Whether individual Likert items can be considered as interval-level data, or whether they should be considered merely ordered-categorical data is the subject of disagreement. Many regard such items only as ordinal data, because, especially when using only five levels, one cannot assume that respondents perceive all pairs of adjacent levels as equidistant. On the other hand, often (as in the example above) the wording of response levels clearly implies a symmetry of response levels about a middle category; at the very least, such an item would fall between ordinal- and interval-level measurement; to treat it as merely ordinal would lose information. Further, if the item is accompanied by a visual analog scale, where equal spacing of response levels is clearly indicated, the argument for treating it as interval-level data is even stronger.

When treated as ordinal data, Likert responses can be collated into bar charts, central tendency summarised by the median or the mode (but some would say not the mean), dispersion summarised by the range across quartiles (but some would say not the standard deviation), or analyzed using non-parametric tests, e.g. chi-square test, Mann–Whitney test, Wilcoxon signed-rank test, or Kruskal–Wallis test.[4] Parametric analysis of ordinary averages of Likert scale data is also justifiable by the Central Limit Theorem, although some would disagree that ordinary averages should be used for Likert scale data.

The Straits Times Forum editor Yap Koon Hong has told opposition politician Chee Soon Juan that he would be denied space in the Forum pages until he withdrew his “serious and unfounded aspersions” on the integrity of the newspaper. He was responding to previous reports on the SDP website which implied that The Straits Times had unfairly edited certain portions of Chee’s letters before publishing them.

Chee took issue with the fact that the title of his second letter had been changed to "PAP just as confrontational, replies Chee", which implied that he was indeed confrontational. In fact, newspapers rarely use the headlines suggested by writers and most readers know that headlines, together with picture selection and captions, are the work of sub-editors and not writers.

The SDP website shows that almost half of Chee's letter was deleted before publication. The original text was 635 words long. Forum page letters are generally less than 400 words long. The published version of Chee's letter was 348 words long.

This was not a statistically controlled study: the subjects took a free test online and of their own accord.  They were by definition connected to the internet and interested in testing their English; they will also be younger and more urban than the population at large.

But Philip Hult, the boss of EF, says that his sample shows results similar to a more scientifically controlled but smaller study by the British Council.

Wealthy countries do better overall. But smaller wealthy countries do better still: the larger the number of speakers of a country’s main language, the worse that country tends to be at English. This is one reason Scandinavians do so well: what use is Swedish outside Sweden?  It may also explain why Spain was the worst performer in western Europe, and why Latin America was the worst-performing region: Spanish’s role as an international language in a big region dampens incentives to learn English.

Mr Murdoch's close henchman Les Hinton assured MPs that the affair had been dealt with and when, two years later, Mr Coulson – by now director of communications for David Cameron – appeared before a renewed parliamentary inquiry he seemed confident of being fireproof. "We did not use subterfuge of any kind unless there was a clear public interest in doing so," he told MPs. When Scotland Yard concluded that, despite more allegations of hacking, there was nothing new to investigate, Wapping and Mr Coulson must again have concluded the affair was over.

But after an election campaign in which the Conservatives were roundly supported by Mr Murdoch's papers, a succession of further claimants against the News of the World has come forward. Sienna Miller, among others, seems determined to take her case to court, compelling Mulcaire to reveal his handlers and naming in court documents Ian Edmondson, once one of Coulson's executives. Mr Edmondson is now suspended. But the story is unlikely to end there

When Rupert Murdoch came to England last October to deliver a lecture, there were some in the audience who raised eyebrows when the media mogul broke off from a paean to Baroness Thatcher to say of his journalists: "We will vigorously pursue the truth – and we will not tolerate wrongdoing." The latter comment seemed to refer to the long-running phone-hacking scandal involving the News of the World, the tabloid he has owned for 41 years. Mr Murdoch's executives at his British headquarters in Wapping, east London, tried to draw a veil over the paper's own dirty secrets in 2007 and had no doubt assured him that the matter was history. Yet here was the boss, four years later, having to vouch for his organisation's honesty. Related articles

The cost of a college education has risen, in real dollars, by 250 to 300 percent over the past three decades, far above the rate of inflation. Elite private colleges can cost more than $200,000 over four years. Total student-loan debt, at nearly $830 billion, recently surpassed total national credit card debt. Meanwhile, university presidents, who can make upward of $1 million annually, gravely intone that the $50,000 price tag doesn’t even cover the full cost of a year’s education.

Then your daughter reports that her history prof is a part-time adjunct, who might be making $1,500 for a semester’s work. There’s something wrong with this picture.

The higher-ed jeremiads of the last generation came mainly from the right. But this time, it’s the tenured radicals — or at least the tenured liberals — who are leading the charge. Hacker is a longtime contributor to The New York Review of Books and the author of the acclaimed study “Two Nations: Black and White, Separate, Hostile, Unequal,”

We have carried out a three-part study comparing the research performance of Indian institutions with that of other international institutions. In the first part, the publication profiles of various Indian institutions were examined and ranked based on the h-index and p-index. We found that the institutions of national importance contributed the highest in terms of publications and citations per institution. In the second part of the study, we looked at the publication profiles of various Indian institutions in the high-impact journals and compared these profiles against that of the top Asian and US universities. We found that the number of papers in these journals from India was miniscule compared to the US universities. Recognizing that the publication profiles of various institutions depend on the field/departments, we studied [in Part III] the publication profiles of many science and engineering departments at the Indian Institute of Science (IISc), Bangalore, the Indian Institutes of Technology, as well as top Indian universities. Because the number of faculty in each department varies widely, we have computed the publications and citations per faculty per year for each department. We have also compared this with other departments in various Asian and US universities. We found that the top Indian institution based on various parameters in various disciplines was IISc, but overall even the top Indian institutions do not compare favourably with the top US or Asian universities.

The comparison groups of institutions include MIT, UMinn, Purdue, PSU, MSU, OSU, Caltech, UCB, UTexas (all from the US), National University of Singapore, Tsing Hua Univerrsity (China), Seoul National University (South Korea), National Taiwan University (Taiwan), Kyushu University (Japan) and Chinese Academy of Sciences.

... [T]he number of papers in these [high impact] journals from India was miniscule compared to [that from] the US universities. ... [O]verall even the top Indian institutions do not compare favourably with the top US or Asian universities.

A computer has beaten a human at shogi, otherwise known as Japanese chess, for the first time.

computers have been beating humans at western chess for years, and when IBM's Deep Blue beat Gary Kasparov in 1997, it was greeted in some quarters as if computers were about to overthrow humanity. That hasn't happened yet, but after all, western chess is a relatively simple game, with only about 10123 possible games existing that can be played out. Shogi is a bit more complex, though, offering about 10224 possible games.

Japan's national broadcaster, NHK, reported that Akara "aggressively pursued Shimizu from the beginning". It's the first time a computer has beaten a professional human player.

Dan Sarewitz tossed some red meat out on the table in the form of an essay in Slate on the apparent paucity of Republicans among the US scientific establishment.  Sarewitz suggests that it is in the interests f the scientific community both to understand this situation and to seek greater diversity in its ranks, explaining that "the issue here is legitimacy, not literacy."

The issue that Sarewitz raises is one of legitimacy.  All of us evaluate knowledge claims outside our own expertise (and actually very few people are in fact experts) based not on a careful consideration of facts and evidence, but by other factors, such as who we trust and how their values jibe with our own.  Thus if expert institutions are going to sustain and function in a democratic society they must attend to their legitimacy.  Scientific institutions that come to be associated with one political party risk their legitimacy among those who are not sympathetic to that party's views.

Of course, we don't just evaluate knowledge claims simply based on individuals, but usually through institutions, like scientific journals, national academies, professional associations, universities and so on. Sarewitz's Slate article did not get into a discussion of these institutions, but I think that it is essential to fully understand his argument.