A study in which Facebook manipulated the news feeds to more than 600,000 users sent social media users into a cyber-swoon this week and spilled over into the mainstream media: “Facebook Tinkers With Users’ Emotions,” began the headline on the New York Times website.
But the controversy over what these researchers did may be overshadowing other important discussions, specifically conversations about what they really found—not much, actually—and the right and wrong way to think about and report findings based on statistical analyses of big data. (We’ll get to the ethics of their experiment in a moment.)
Because they are so large, studies based on supersized samples can produce results that are statistically significant but at the same time are substantively trivial. It’s simple math: The larger the sample size, the smaller any differences need to be to be statistically significant—that is, highly likely to be truly different from each other. (In this study, the differences examined were between those who saw more and those who saw fewer emotion-laden posts compared with a control group whose news feeds were not manipulated.) Read More →
Category: Social Studies
Of the thousands of unaccompanied children apprehended at the U.S. border in recent months, many can be attributed to poverty and regional violence in three Central American countries, a new U.S. Department of Homeland Security document finds. The document says the reasons driving the migration are different for each country, attributing it to local conditions.
“For example, many Guatemalan children come from rural areas, indicating they are probably seeking economic opportunities in the U.S. Salvadoran and Honduran children, on the other hand, come from extremely violent regions where they probably perceive the risk of traveling alone to the U.S. preferable to remaining at home.” Read More →
The earnings gap in the nation’s workforce has widened in recent years as the pay of high-wage workers has risen and the pay of low-wage workers has fallen. But while this double-edged phenomenon affects all racial and ethnic groups, Hispanics may be feeling the impact more acutely than others, according to a Pew Research Center analysis.
The median weekly earnings of Hispanics at the low end of the wage scale fell by 9.4%, from $278 to $252 (in fourth quarter 2013 dollars) between 2007 and 2013, years that span the Great Recession and the fitful economic recovery. Meanwhile, the median weekly earnings of Hispanics in the highest wage bracket increased 4.4%, from $1,604 to $1,675.
The earnings gap also stretched among other groups of workers, but to a lesser extent.
Among blacks, the median weekly earnings of those in the lowest wage bracket of the U.S. income distribution decreased 7.7% from the fourth quarter of 2007 to the fourth quarter of 2013 while the earnings of those in the highest wage bracket increased 1.9%. Over the same period, the pay of white low-wage workers fell 5.6% and the pay of white high-wage workers rose 3.4%. Among Asians, the earnings of low-wage workers were unchanged but the earning of high-wage workers increased 6.8%. Read More →
American political parties have long been vehicles to represent ever-shifting coalitions of particular interests — economic, regional, social and ideological. For example, Great Plains farmers who often voted Democratic a century ago now solidly favor Republicans; African Americans who were loyal GOP voters for decades after the Civil War began shifting during the New Deal and now are overwhelmingly Democratic. Political observers can still spend hours of pre-election time ruminating over whether the “big-city ethnic vote” or the “farm vote” will prevail.
But the parties also are coalitions of distinct groups of voters, whose shared attitudes and values unite them — and shape the parties they incline toward — at least as much as more impersonal economic and societal forces. The Pew Research Center’s mammoth new political typology report offers a different way to think of the two major parties’ component pieces. (For the purposes of this post, we limited our analysis to registered voters and combined self-identified Republicans or Democrats with independents who lean toward one or the other party.) Read More →
So far, the growing crisis in Iraq has not drawn strong interest from the American public. As Sunni militants extend their control of large swaths of Iraq, 25% say they are paying very close attention to the growing violence and political instability in Iraq.
By comparison, 28% of Americans surveyed June 26-29 say they followed news about problems with care at veterans’ hospitals very closely and 21% paid very close attention to news about the IRS losing employee emails. Read More →
The Supreme Court today expanded the scope of religious-liberty rights and dealt a blow to part of the Obama administration’s health care overhaul when it ruled 5-4 that some for-profit businesses have religious rights and, as a result, can opt out of the law’s contraception mandate. But the ruling, while important, is limited because it applies only to smaller, closely held businesses and specifically warns against companies using religious-liberty claims to discriminate. Read More →
The Pew Research Center’s studies about libraries and where they fit in the lives of their communities and patrons have uncovered some surprising facts about what Americans think of libraries and the way they use them. As librarians around the world are gathered in Las Vegas for the American Library Association’s annual conference, here are findings that stand out from our research, our typology of public library engagement and the quiz we just released that people can take to see where they compare with our national survey findings: What kind of library user are you?
1Each time we ask about library use, we find that those ages 65 and older are less likely to have visited a library in the past 12 months than those under that age. Equally as interesting is the fact that younger Americans (those ages 16-29) are just as likely to be library users as those who are older.
2Although 10% of Americans have never used a library, they think libraries are good for their communities. We’ve identified this group of library users as “Distant Admirers,” and they are the majority of the nearly 15% of Americans ages 16 and older who have never been to a library. Despite their lack of personal use of libraries, their positive views of libraries might stem from the fact that 40% of Distant Admirers report that someone else in their household is a library user. About two-thirds of them or more say libraries are important because they promote literacy and reading, that they play an important role in giving everyone a chance to succeed and they improve the quality of life in a community. Finally, 55% say the loss of the local library would be a blow to the community.
The Pew Research Center released on Thursday our 2014 political typology, including the newest version of our political typology quiz, in which users can see where they fit in the current political landscape by answering 23 questions about core political values. Within the first 24 hours, more than 90,000 people had taken the quiz, and the last quiz, launched in 2011, had over 1.5 million users over three-plus years.
One of the strongest reactions we have received from some quiz-takers is frustration over the either-or choices each question offers. Many of the comments we’ve received say that the questions feel like choosing “between two extremes,” or that the “right answer is somewhere in between,” or that the options “aren’t really opposites.”
These are all legitimate concerns, and many of our own early users here on the Pew Research Center staff expressed the same frustrations. But there is a reason the questions are asked the way they are: The intent is not to put people “in a box” but rather to understand how their values across multiple political dimensions are related to each other.
Take a question like the one shown here. For many Americans, the two options present a very stark way of thinking about poverty, and it is a fair guess that many people have some level of agreement with both points of view. At the same time, this “forced choice” format reflects a fundamental debate in American politics right now, and most people at least lean one way or the other on this potentially difficult tradeoff. In fact, the American public as a whole splits almost exactly in two on this question – 44% select the first statement, 47% pick the second – and their choice is strongly related to their specific policy preferences and priorities in this realm.
Ramadan, which begins this weekend, is widely observed by Muslims around the world with fasting from sunrise until sundown. That creates a potential dilemma for Muslim players at the World Cup whose teams have advanced to the knockout stage of the tournament, as the ninth month of the Islamic calendar and the World Cup overlap for the first time since 1986.
Although the exact religious makeup of the teams’ rosters is unclear, Muslims comprise 11% of the collective population of the 16 countries that advanced out of the tournament’s group stage, according to 2010 data from the Pew Research Center’s Global Religious Landscape report. Several prominent players on teams that are still alive – including Germany’s Mesut Ozil, France’s Karim Benzema and Switzerland’s Xherdan Shaqiri – are practicing Muslims. The national team of Algeria, a country comprised of 98% Muslims, advanced on a goal by Islam Slimani and will face Germany on Monday.
While some Muslim athletes choose to fast while competing, others see room for interpretation in the Quran, which instructs “whosoever of you is present, let him fast the month, and whosoever of you is sick or on a journey, a number of other days.” Some Islamic leaders endorse fasting at a later date, especially if it is “a once-in-a-lifetime opportunity for an athlete,” a London-based imam told the soccer website Goal.com in 2011. The High Egyptian Islamic Council even issued a fatwa before the 2012 London Olympics, allowing athletes to eat and drink during Ramadan. Read More →
Anyone (well, almost anyone) could make a map, or even a series of maps, about party control of Congress over time. But turning that data into an interactive that’s in its own way as addictive as Minecraft or Fruit Ninja is something else again, which is where MapStory comes in.
MapStory is a nonprofit that’s created a online, open-source mapping platform to enable people to create and share data visualizations. Mapmakers have created timeline maps to illustrate everything from charter-school expansion in Minnesota to the proliferation of wind farms across the United States. But this interactive, uploaded to MapStory earlier this month by Jonathan Davis of Arizona State University, stands out both for the sheer amount of information it conveys and how darned fun it is to play with.
Starting in 1918, you can simply play the animation forward to 2012 and watch the shifting patterns of party control of the House of Representatives. Davis’ map also lets you freeze on specific years and zoom in to see individual districts; it also shows the handful of minor-party representatives who’ve gotten themselves elected to Congress over the decades. (That’s catnip for political-trivia buffs: Who knew that the Prohibition Party held a congressional seat in California in the years leading up to the 18th Amendment?)
Comparing past and present, a couple of related points become clear. First, districts have become more and more jigsaw-puzzle-like, as sophisticated mapping software and detailed demographic data have combined to make gerrymandering a fine art. And partly as a consequence, Democratic and Republican seats were more geographically intermixed in 2012 than in 1918, when virtually all seats in the South and Southwest states were held by Democrats and most Northern states were dominated by the GOP.
Category: Chart of the Week