Science Policy For All

Because science policy affects everyone.

Science Policy Around the Web – March 4, 2016

leave a comment »

By: Valerie Miller, Ph.D.

License: (license) Photo credit: DSC_7102 via photopin (license)

Election Policy

How different polling locations subconsciously influence voters

In his final State of the Union address, President Obama discussed the need to revamp the voting process, stating, “We’ve got to make it easier to vote, not harder. We need to modernize it for the way we live now.” Now that we are fully entrenched in election season, many are questioning the fairness of voter ID laws, which can disproportionately affect minority voters. However, there is research demonstrating another phenomenon: that the location where you vote can influence the choices you make in the voting booth. Priming, in which the identification of ideas and objects by our subconscious memory can manipulate thoughts or behaviors, may be responsible for the way in which voting patterns are influenced by outside stimuli. To help curtail priming, most states prohibit campaign materials within 100 feet of polling locations, and some forbid wearing campaign paraphernalia while voting, yet a number of studies have shown that polling locations can themselves prime voters toward specific behaviors. For example, one study found that voting in a church or other religious location can prime attitudes associated with conservative values, such as negative attitudes toward the LGBT community or same-sex marriage. Social scientists also question voting in schools, which has been shown to increase voting in favor of education-related measures on the ballot. In all, there are six published studies that examine whether voting behaviors can be influenced by polling location, and each one has concluded that polling place priming is real. As an alternative, it has been suggested that polling locations should be eliminated all together, in favor of adopting a ballot-by-mail system. While some argue against all-mail voting, because of tradition or potential voter fraud, Colorado, Oregon and Washington have already implemented a voting system that is done entirely by mail. All three states have seen significant increases in voter turnout. (Ben Pryor, Scientific American)

Global Health

In 2050, half the world will be nearsighted

In a new study published in the journal Ophthalmology, scientists have predicted that nearly 5 billion people will suffer from myopia (nearsightedness) by the year 2050, which corresponds to half the projected global population. Researchers also found that almost 10 percent (nearly 1 billion people) could develop high myopia, in which severe nearsightedness leads to increased risk for cataracts, glaucoma, retinal detachment and macular degeneration. In contrast, in the year 2010, just 28.3 percent (2 billion people) had myopia, with 4 percent (277 million people) experiencing high myopia, corresponding to a 140 percent increase by 2050.

The researchers attribute the rise in myopia to be driven by lifestyle and environmental factors, likely decreased time spent outdoors and increased time spent doing near-work activities on screens. Indeed, there are regional differences in the incidence of nearsightedness, with more cases in high-income countries in North America and Asia. In the United States, myopia has been increasing steadily, with one study finding a 66 percent increase between the early 1970s and the early 2000s. The researchers conclude that if current trends continue, with more time on screens and less time outdoors, their projections for 2050 will likely hold true, and the number of people experiencing vision loss from high myopia is likely to increase 7-fold from the year 2000, becoming a leading cause of blindness. (Julie Beck, The Atlantic)

Research Misconduct

Many surveys, about one in five, may contain fraudulent data

Scientists are often reminded of data fabrication and research misconduct in the laboratory, but what about in social sciences and survey research? Last year, a high-profile case of research misconduct brought survey fraud to the forefront, when a researcher was found to have fabricated data demonstrating that short conversations could change people’s minds on same-sex marriage. But an important question remains: how wide-spread is the problem? Two researchers, Noble Kuriakose, a research scientist at SurveyMonkey, and Michael Robbins, a researcher affiliated with the University of Michigan and Princeton University, sought to address this issue by developing a statistical test designed to detect when a survey may contain fabricated data. The test is based on determining the likelihood that two independent respondents will give highly similar answers to survey questions, and is meant to examine large-scale opinion surveys that cover broad topics and are designed to identify community differences.

Robbins, who is also the director of Arab Barometer, a project measuring opinions in the Middle East, notes that in developing countries, conducting survey research often requires in-person interviews, which can be dangerous and time-consuming. Thus, one problem that arises is that interviewers will sometimes invent survey responses to avoid risk and save time. Indeed, when Robbins and Kuriakose applied their test to 1000 sets of public data they found that nearly one in five of the surveys failed their test, indicating they were likely to contain significant portions of fabricated data. While only 5 percent of surveys conducted in westernized nations failed, 26 percent of studies conducted in developing nations were flagged by the test.

However, not all groups agree with Kuriakose and Robbins’ findings. Pew Research Center, which has performed hundreds of international surveys, is disputing the claims and have suggested that Kuriakose and Robbins should retract their findings. Courtney Kennedy, director of survey research at Pew, stated that after using the test on their own data, some were flagged, but after digging deeper, found that only a few surveys had serious questions. Kennedy believes the test is likely to find false positives, and doesn’t account for number of survey questions, number of respondents or other factors that could skew results. Pew has since posted a rebuttal online. (John Bohannon, Science Magazine)

Have an interesting science policy link?  Share it in the comments!


Written by sciencepolicyforall

March 4, 2016 at 9:00 am

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: