Group items tagged

Statistical validation of results, as Shaffer described it, simply involves testing the null hypothesis: that the pattern you detect in your data occurs at random. If you can reject the null hypothesis—and science and medicine have settled on rejecting it when there's only a five percent or less chance that it occurred at random—then you accept that your actual finding is significant. The problem now is that we're rapidly expanding our ability to do tests. Various speakers pointed to data sources as diverse as gene expression chips and the Sloan Digital Sky Survey, which provide tens of thousands of individual data points to analyze. At the same time, the growth of computing power has meant that we can ask many questions of these large data sets at once, and each one of these tests increases the prospects than an error will occur in a study; as Shaffer put it, "every decision increases your error prospects." She pointed out that dividing data into subgroups, which can often identify susceptible subpopulations, is also a decision, and increases the chances of a spurious error. Smaller populations are also more prone to random associations. In the end, Young noted, by the time you reach 61 tests, there's a 95 percent chance that you'll get a significant result at random. And, let's face it—researchers want to see a significant result, so there's a strong, unintentional bias towards trying different tests until something pops out. Young went on to describe a study, published in JAMA, that was a multiple testing train wreck: exposures to 275 chemicals were considered, 32 health outcomes were tracked, and 10 demographic variables were used as controls. That was about 8,800 different tests, and as many as 9 million ways of looking at the data once the demographics were considered.

The "Facebook Subpoena/Search Warrant Guidelines" from the Cryptome site are dated 2008, so there's a chance they've been superseded. The document spells out how law enforcement and intelligence agenices should go about requesting information about Facebook users, and details what information is turned over. Following is what Facebook will turn over about you, taken verbatim from the guide: Types of Information Available User Neoprint The Neoprint is an expanded view of a given user profile. A request should specify that they are requesting a “Neoprint of used Id XXXXXX”. User Photoprint The Photoprint is a compilation of all photos uploaded by the user that have not been deleted, along with all photos uploaded by any user which have the requested user tagged in them. A request should specify that they are requesting a “Photoprint of user Id XXXXXX”. User Contact Info All user contact information input by the user and not subsequently deleted by the user is available, regardless of whether it is visible in their profile. This information may include the following: Name Birth date Contact e-mail address(s) Physical address City State Zip Phone Cell Work phone Screen name (usually for AOL Messenger/iChat) Website With the exception of contact e-mail and activated mobile numbers, Facebook validates none of this information. A request should specify that they are requesting "Contact information of user specified by [some other piece of contact information]". No historical data is retained. Group Contact Info Where a group is known, we will provide a list of users currently registered in a group. We will also provide a PDF of the current status of the group profile page. A request should specify that they are requesting "Contact information for group XXXXXX". No historical data is retained. IP Logs IP logs can be produced for a given user ID or IP address. A request should specify that they are requesting the "IP log of user Id XXXXXX" or "IP log of IP address xxx.xxx.xxx.xxx". The log contains the following information: * Script – script executed. For instance, a profile view of the URL http://www.facebook.com/profile.php?id=29445421 would populate script with "profile.php" * Scriptget – additional information passed to the script. In the above example, scriptget would contain "id=29445421" * Userid – The Facebook user id of the account active for the request * View time – date of execution in Pacific Time * IP – source IP address IP log data is generally retained for 90 days from present date. However, this data source is under active and major redevelopment and data may be retained for a longer or shorter period. Special Requests The Facebook Security Team may be able to retrieve specific information not addressed in the general categories above. Please contact Facebook if you have a specific investigative need prior to issuing a subpoena or warrant.

Comcast The Comcast document is labeled "Comcast Cable Law Enforcement Handbook," and is dated 2007, so there's a possibility that it, too, has been superseded. As with the other documents, it explains how law enforcement agenices can get information, and details what information is available. There's a great deal of detail in the 35-page document, which describes what Internet, phone, and television information will be turned over. For example, here's the IP information it will make available: Comcast currently maintains Internet Protocol address log files for a period of 180 days. If Comcast is asked to respond for information relating to an incident that occurred beyond this period, we will not have responsive information and can not fulfill a legal request. (Comcast can process and respond to preservation requests as outlined below in this Handbook.) As expected, Comcast will also turn over the emails, including attachments, of those who use Comcast's email service, but "In cases involving another entity’s email service or account, Comcast would not have any access to or ability to access customer email in response to a legal request." Information Comcast turns over to law enforcement agencies varies according to the request. For example, a grand jury subpoena will yield more information than a judicial summons, as you can see in the excerpt below. Comcast notes, though, that this is just a sample, and that "Each request is evaluated and reviewed on a case by case basis in light of any special procedural or legal requirements and applicable laws." So the examples "are for illustration only."

For those who worry about privacy, though, all of this information is small potatoes. The real worry is about the use of what are called pen registers or trap-and-trace devices, which essentially capture all of your Internet activity --- the Web sites you visits, the emails you send and receive, IM traffic, downloads, and so on. Here's what the document says about them: Pen Register / Trap and Trace Device Title 18 U.S.C. § 3123 provides a mechanism for authorizing and approving the installation and use of a pen register or a trap and trace device pursuant to court order. All orders must be coordinated prior to submission to Comcast. Law enforcement will be asked to agree to reimburse Comcast's reasonable costs incurred to purchase and/or install and monitor necessary equipment. See "Reimbursement," below.

Trade's effect on military conflict is one of the most important issues in international relations. The last decade has seen research and debate into the role of trade intensify; Liberals argue that trade brings peace, neo-realists and neo-Marxists reason that trade brings conflict, and classical realists contend that trade has no impact. This debate is not just academic: some key U.S. policymakers (Senator McCain and former President Clinton for instance) believe that trade brings peace, a view that contributes to their support for free trade. Economists developed bilateral trade models in isolation from models of interstate conflict, which were the work of political scientists. These two types of models handle distance between nations differently. Bilateral trade takes its cue from Isaac Newton's formula for the gravitational attraction between two objects: the larger the objects' masses and the shorter the distance between them, the larger the attraction. So the larger the trade partners' economies and the closer they are to one another, the greater their trade. However, conflict models instead incorporate shared borders by land or close distance over water (contiguity) - stressing the role of border disputes in sparking interstate conflict. Distance is included in conflict equations based on the idea that an army gets weaker the farther it strays from its base, but what point in a nation to pick for the trade and conflict equation is unclear. Often theorists use the distance between capital cities, which is problematic: wars generally happen around borders where armies are often based, and capitals have historically changed without this altering the likelihood of war between the nation and its neighbours. The authors suggest that the trade data set plugged into trade and conflict equations is critical. This type of data often contains gaps - there are a number of reasons why data from a particular nation might be unavailable, inevitably leaving researchers to make assumptions. The majority of trade and conflict studies define conflict to include all types of militarised interstate disputes (MIDs). But Keshk, Reuveny, and Pollins question the results generated when different conflict definitions are chosen. For instance, a conflict such as a threat to use nuclear weapons would not cause fatalities, but may still have some impact on trade and vice versa. In fact, by altering the data treatment and assumptions in the equation, the authors generated a variety of results, which supported several different theoretical viewpoints. The authors suggest that future research should investigate questions of missing bilateral trade data, and attempt a more subtle use of the meaning of "military conflict". Researchers might also develop distance and contiguity measures at a more sophisticated level. "Any signal that trade brings peace remains weak and inconsistent, regardless of the way proximity is modelled in the conflict equation. The signal that conflict reduces trade, in contrast, is strong and consistent," say the authors. "Any study of the effect of trade on conflict that ignores the reverse fact is practically guaranteed to produce estimates that contain simultaneity bias." Studies of the relationship between international trade and military conflict can be traced back many centuries, particularly in the works of luminaries such as de Montesquieu, Immanuel Kant, John Hobson, Vladimir Lenin, Henry Morgenthau, Kenneth Waltz, Frederic List, and Albert Hirschman. This latest study emphasises that international politics are affecting trade between nation pairs, while it is far less obvious whether trade systematically affects politics. "To our colleagues from the liberal camp we would like to say that we still believe there are limited circumstances in which more trade may help lead countries to more peaceful resolutions of their differences, particularly if they are already at peace," the authors state. "However, it is past time for academics and policymakers to look beyond the naive claim that the cultivation of trade ties will always and everywhere produce a more peaceful world."

"People tend to conform their factual beliefs to ones that are consistent with their cultural outlook, their world view," Braman says. The Cultural Cognition Project has conducted several experiments to back that up. Participants in these experiments are asked to describe their cultural beliefs. Some embrace new technology, authority and free enterprise. They are labeled the "individualistic" group. Others are suspicious of authority or of commerce and industry. Braman calls them "communitarians." In one experiment, Braman queried these subjects about something unfamiliar to them: nanotechnology — new research into tiny, molecule-sized objects that could lead to novel products. "These two groups start to polarize as soon as you start to describe some of the potential benefits and harms," Braman says. The individualists tended to like nanotechnology. The communitarians generally viewed it as dangerous. Both groups made their decisions based on the same information. "It doesn't matter whether you show them negative or positive information, they reject the information that is contrary to what they would like to believe, and they glom onto the positive information," Braman says.

"Basically the reason that people react in a close-minded way to information is that the implications of it threaten their values," says Dan Kahan, a law professor at Yale University and a member of The Cultural Cognition Project. Kahan says people test new information against their preexisting view of how the world should work. "If the implication, the outcome, can affirm your values, you think about it in a much more open-minded way," he says. And if the information doesn't, you tend to reject it. In another experiment, people read a United Nations study about the dangers of global warming. Then the researchers told the participants that the solution to global warming is to regulate industrial pollution. Many in the individualistic group then rejected the climate science. But when more nuclear power was offered as the solution, says Braman, "they said, you know, it turns out global warming is a serious problem."And for the communitarians, climate danger seemed less serious if the only solution was more nuclear power.