research – Sheila Kennedy

Filters And Lies

I know I carp constantly about the degree to which propaganda and conspiracy theories have displaced credible information, with the result that today’s Americans occupy different realities. It’s easy to blame social media for the reach of disinformation and lies–and social media does bear a significant amount of the blame–but research also illuminates the way propaganda has changed in the era of cable news and the Internet.

That research has identified two modern mechanisms for eroding social trust and constructing alternate realities. One –to quote Steve Bannon’s vulgar description–is to “flood the airways with shit.” In other words, to produce mountains of conflicting “news” along with lots of “shiny objects.” The faux “news” confuses; the shiny objects distract. Citizens don’t know what to believe, what parts of the fire hose of information, disinformation, and outright invention they can trust. They either accept a particular storyline (chosen via confirmation bias) or opt out.

But it isn’t simply the fire hose approach that has eroded our common realities. These days, when people get most of their news from partisan sources, all too often they simply don’t get news that is inconsistent with partisan biases.

A recent, widespread report illustrates that technique. As the lede put it, “The problem with Fox ‘News,’ the cable TV channel, isn’t just what it is — it’s also what it isn’t.” It was a fascinating new study in which arch-conservative Fox TV viewers were paid to watch CNN for a month.

The study, titled “The manifold effects of partisan media on viewers’ beliefs and attitudes: A field experiment with Fox News viewers,” was performed by a pair of political scientists: David Broockman, who teaches at UC-Berkeley, and Joshua Kalla, who teaches at Yale.

According to Broockman and Kalla, when these Fox viewers watched CNN, they heard about all sorts of things Fox wasn’t telling them. They processed that information. They took it in. They became more knowledgeable about what was really going on in the United States.

The individuals who took part in the experiment didn’t change their political leanings or partisan preferences, but the experience did alter their perceptions of certain key issues and political candidates.

The study authors differentiated between “traditionally emphasized forms of media influence,” like agenda setting and framing, and what they call “partisan coverage filtering”: the choice to selectively report information about selective topics, based on what’s favorable to the network’s partisan side, and ignore everything else.

The article emphasized what the author called the “real problem” with Fox : its viewers aren’t just manipulated and misinformed — they are left ignorant of much of the news covered by more reputable outlets. Fox gives them a lot of “news-like” information, but they don’t learn about things like Jared Kushner getting two billion dollars from Saudi Arabia.

That conclusion reminded me of another research project a couple of years ago. People were asked to identify their primary news sources and then quizzed on things currently in the news. Those who named Fox as their preferred news source knew less than people who didn’t watch any news from any source.

Lest you think that “filtering” of this sort is a tactic exclusive to the Right, when one of the authors of the research study was interviewed on CNN, he noted that CNN, too, filtered its reporting.

CNN’s Brian Stetler interviewed Joshua Kalla, one of the co-authors of the study, and they had the following exchange:

“You call this partisan coverage filtering,” Stelter told Kalla. “And basically, you’re proving what we’ve sensed for a while, which is that Fox viewers are in the dark about bad news for the GOP.”

Kalla confirmed the Fox News coverage model but put a stop to the victory lap: “On the flip side, CNN engages in this partisan coverage filtering as well… For example, during this time, the Abraham Accords were signed, and these were the agreements where Israel, the UAE and Bahrain signed a major peace agreement. And we see that Fox News covered this really major accomplishment about 15 times more than CNN did. So we established both networks are really engaging in this partisan coverage filtering. It’s not about one side, it’s about the media writ large.”

To be fair, CNN is apparently less culpable in this regard than Fox..

America’s ugly politics is obviously attributable to a lot more than the country’s media environment, even if you throw in the very divisive algorithms used by social media. (After all, the KKK didn’t use the Internet.) But it’s hard to avoid the conclusion that both mass media and social media have contributed disproportionately to our loss of a common reality.

As always, the questions are: what policies might make things better? And can we pass those policies once they are identified?

Comments

Language, Fact & Emotion, Oh My!!

Many thanks for all the kind comments yesterday! They were much appreciated!!

___________________________________

Computers haven’t only changed our day-to-day lives in multiple ways, their computational capacities have made it possible to conduct studies far beyond the ability of mere humans. One of my sons sent me an article published in The Proceedings of the National Academy of Scientists detailing a research project that would have been impossible to conduct prior to the availability of today’s technologies.

The researchers set out to examine the roots of what they dub–accurately, in my view– our “post-truth era.” In order to do so, they employed “massive language analysis” to document the rise of fact-free argumentation. They analyzed language used in millions of books published between 1850 to 2019–an analysis that required Google nGram data.

What they found is illuminating, to put it mildly.

After the year 1850, the use of sentiment-laden words in Google Books declined systematically, while the use of words associated with fact-based argumentation rose steadily. This pattern reversed in the 1980s, and this change accelerated around 2007, when across languages, the frequency of fact-related words dropped while emotion-laden language surged, a trend paralleled by a shift from collectivistic to individualistic language.

The researchers concluded that this “surge of post-truth political argumentation” is evidence that we are living at a time when the balance between emotion and reasoning has shifted.

To explore if this is indeed the case, we analyze language in millions of books covering the period from 1850 to 2019 represented in Google nGram data. We show that the use of words associated with rationality, such as “determine” and “conclusion,” rose systematically after 1850, while words related to human experience such as “feel” and “believe” declined. This pattern reversed over the past decades, paralleled by a shift from a collectivistic to an individualistic focus as reflected, among other things, by the ratio of singular to plural pronouns such as “I”/”we” and “he”/”they.”

The reversal occurred in both fiction and nonfiction. It wasn’t limited to books, either– they found a similar shift in media (like New York Times articles). The results of the research “suggest that over the past decades, there has been a marked shift in public interest from the collective to the individual, and from rationality toward emotion.”

The bulk of the article is a description of the methodology employed, the care taken to avoid “cherry picking” of data, and a variety of theories about the reason for the language shift. (There is definitely a “chicken and egg” aspect to the shift: did disillusionment with science and evidence drive a language shift? Or did something else prompt the change in language and thus promote an anti-science mood in the general public?)

The researchers concluded with a section they captioned “Outlook.”

It seems unlikely that we will ever be able to accurately quantify the role of different mechanisms driving language change. However, the universal and robust shift that we observe does suggest a historical rearrangement of the balance between collectivism and individualism and—inextricably linked—between the rational and the emotional or framed otherwise. As the market for books, the content of the New York Times, and Google search queries must somehow reflect interest of the public, it seems plausible that the change we find is indeed linked to a change in interest, but does this indeed correspond to a profound change in attitudes and thinking? Clearly, the surge of post-truth discourse does suggest such a shift, and our results are consistent with the interpretation that the post-truth phenomenon is linked to a historical seesaw in the balance between our two fundamental modes of thinking. If true, it may well be impossible to reverse the sea change we signal. Instead, societies may need to find a new balance, explicitly recognizing the importance of intuition and emotion, while at the same time making best use of the much needed power of rationality and science to deal with topics in their full complexity. Striking this balance right is urgent as rational, fact-based approaches may well be essential for maintaining functional democracies and addressing global challenges such as global warming, poverty, and the loss of nature.

This study is fascinating, albeit depressing.

I’ve previously suggested that our current era will be labeled (assuming there are humans and historians left to do the labeling) “the age of Unreason.” The language we use matters far more than we generally recognize; it both reflects and produces our biases.

And right now, those biases evidently elevate emotion over reason and logic. Which explains a lot….

Comments

The Cost Of Luring Jobs

Over the past decade or so, like this blog, Americans’ political discussions and debates have focused on national issues and the increasing gridlock in Washington. There are several reasons for that. The decline of local journalism has meant that local issues that might trigger local activism are increasingly less likely to be covered, while more national media highlights the growing dysfunction of the federal government. And many of the challenges we face are national–or global–in scope.

Although it’s understandable that local policies tend to fly “under the radar,” that doesn’t make those issues unimportant. For one thing, individual citizens who are powerless to change goings-on in Washington can affect many local issues.

Governing Magazine recently focused on one such issue: economic development.

The article pointed out what even casual observers have long suspected, and what the data confirms–most state and local governments approach economic development in costly and unproductive ways. The article’s subhead really sums up the conclusion: “Governments can’t seem to stop offering huge incentives to corporations, even though it’s clear they don’t have much effect on companies’ decisions. Does paying $288,000 for one job really make sense?”

The rather obvious answer to that question is no. But economic development officials are responding to the pandemic by doubling down–ignoring overwhelming evidence and instead doing more of what they know. (This situation reminds me of America’s long, counterproductive drug war. As I said in a speech some years ago, if a doctor performed a hundred identical surgeries and every single patient died, would you insist that the proper response was to have him do more of them? The logic is the same.)

Seeking to create jobs and help their local economies climb out of the pandemic recession, state and local officials are raising the ante on subsidies to big corporations. But if history is any guide, ever-increasing tax breaks and other economic development incentives will likely lead to slower — not faster — growth. Given that state and local governments have already been wasting $95 billion every year in an economic race to the bottom, more subsidies will just dig the hole deeper.

The article highlighted North Carolina’s largest-ever subsidy: $865 million for an Apple research and development center promising 3,000 new jobs. But Apple would probably have chosen North Carolina in any event–without those subsidies.

Smart companies like Apple understand that the real long-term attraction is not subsidies so much as the great economic foundation North Carolina has built: investments in top-notch research universities, a tech-ready workforce and a business-friendly environment. North Carolina is indeed a perfect place to locate a cutting-edge research center. Site Selection magazine has consistently ranked it as a top state for business climate.

Interestingly, when Apple located a facility in Austin, Texas gave the company about $10,000 per job. North Carolina promised some $288,000 per job.

Research tells us that only one in eight subsidies effects a change to a location or expansion decision, and that some 90 percent are a complete waste of money. Companies happily accept the money, but their decisions are based far more on the availability of a talented local workforce, region-specific advantages and access to supply chains and customers.

For example, Google and Fidelity Investments recently announced expansions to their existing operations in the Research Triangle — without asking North Carolina for subsidies. Both emphasized the area’s skilled workforce as the primary draw.

The consensus of academic research is that corporate handouts don’t create broad benefits for the community providing them. That’s because subsidies motivate wasteful corporate investments and create public funding trade-offs. Every dollar spent on subsidies is a dollar that can’t be used to improve infrastructure, education or public safety, or to cut taxes on smaller businesses and households.

This expensive and unnecessary fiscal competition between local units of government adds absolutely nothing to the national economy–after all, nationally, moving enterprise A from city B to city C is a zero-sum exercise. And as the article notes, paying companies to move to your state siphons off funds that could be used for things that actually make your state attractive to those companies–like a first-rate public education system that not only turns out a skilled workforce, but is an amenity valued by the management folks who would be locating in your state.

The evidence shows that one of the most persuasive “subsidies” a state can offer is an attractive quality of life.

When policymakers ignore evidence, when they make decisions on the basis of ideology–or worse, when policy decisions are simply the result of “we’ve always done it this way” or “everyone else does it this way”–the costs aren’t limited to the dollar amount of the subsidies.

Comments

About that 40%…

The problem isn’t Trump. As numerous people have recognized, Trump–despicable and dangerous and deranged as he is– is the symptom, not the disease.

I’ve previously posted about the systemic and structural fault-lines that have been exploited by Trump’s GOP supporters and fellow-travelers–but the disease, the root problem, isn’t the systems. It’s those supporters. Polls suggest that some 40% of Americans fall into that category, and the recurring, haunting question is: why? How could any sane adult look at this man and say, yep, that’s the guy I want directing my government? That’s the role model I want my kids to emulate?

Actually, I think my manicurist answered that question during a recent appointment.

She’s an adorable young woman (and very “woke” as the current terminology would have it). We were discussing the election, and she shared her distress that several family members were Trump supporters. I asked the question I always ask: why? What was her impression/ best guess about the basis of that support? She thought for a moment, then said “I hate to say this, but I think they are sort of racist, and Trump gives them permission to feel that way.”

Her anecdotal suspicions continue to be confirmed by the research, some of which I’ve referenced in prior posts. As more studies emerge, the evidence continues to grow.

The Washington Post recently reported on research into the authoritarian proclivities of Trump supporters–research that linked those tendencies to racial animus.

In “Authoritarian Nightmare,” Bob Altemeyer and John W. Dean marshal data from a previously unpublished nationwide survey showing a striking desire for strong authoritarian leadership among Republican voters.

They also find shockingly high levels of anti-democratic beliefs and prejudicial attitudes among Trump backers, especially those who support the president strongly. And regardless of what happens in 2020, the authors say, Trump supporters will be a potent pro-authoritarian voting bloc in the years to come.

The research paints a picture of people who are “submissive, fearful, and longing for a mighty leader who will protect them from life’s threats.” They are particularly prone to divide the world into friends and foes, and to believe that the foes far outnumber the friends.

Other researchers have reached similar conclusions using very different methods. Vanderbilt political scientist Larry Bartels, for instance, recently used YouGov survey data to find that many Republican voters hold strong authoritarian and anti-democratic beliefs, with racism being a key driver of those attitudes.

In the most recent study, respondents were asked whether they agreed or disagreed with the statement: “Once our government leaders and the authorities condemn the dangerous elements in our society, it will be the duty of every patriotic citizen to help stomp out the rot that is poisoning our country from within.” Roughly half of Trump supporters agreed with that statement, which–as Altemeyer and Dean point out– is “practically a Nazi cheer.”

If there has been one overarching lesson to be learned from the past few years, it is the (previously unappreciated) extent to which tribalism, racism and bigotry explain things that are otherwise inexplicable. A recent essay from New York Magazine analyzed the failure of Congress and the President to agree on a second, desperately needed stimulus package. The author’s conclusion was stunning: “bailing out” blue states would benefit ethnic minorities–something Republicans are loathe to do.

The most plausible explanation for this state of affairs is this: Most Senate Republicans face no great risk of losing their seats to a Democrat this year or any other. For them, the main threat to their power is a primary challenge. And right now, conservative media has turned opposition to fiscal aid into a cause célèbre, casting support for “blue-state bailouts” as treasonous.

How hateful do you have to be to withhold aid during a global pandemic to people you see as “Other”–even if by doing so, you and those you view as your own kind are harmed as well?

Even if there is a blue tsunami on November 3d, the people who hold these attitudes will still constitute a troubling percentage of the electorate. We can only hope that they fall far short of a majority.

And I have to wonder: What the hell is wrong with them?

Comments

Sex And The State

Indianapolis’ Pride Celebration gets bigger and better every year–this year, the parade was so crowded with people enjoying the lovely day and the multiple marchers and floats that the “usual subjects”–with their signs proclaiming the sinfulness of “homo” sex– almost escaped notice.

Those “usual subjects”–the scolds who come out of the woodwork to tell LGBTQ folks that God disapproves of them, and the “good Christians” who scream invective at women entering Planned Parenthood clinics–are reminders that Americans have always had a real problem with sex. Not just gay sex, either. Any sex.

Residents of more laid-back countries (no pun intended) have found both America’s excessive religiosity and famous prudishness puzzling, and both of those elements of our political culture are barriers to reasonable policymaking. Most of the country has finally recognized that statutes forbidding fornication, sodomy and the like didn’t prevent those behaviors, but simply allowed police who were so inclined to harass marginalized folks with what lawyers call “arbitrary and capricious” enforcement.

The gratifying disappearance of these silly statutes, however, doesn’t mean we Americans have lost our obsession with sex. The fights have simply moved to other venues, like abortion, transgender bathrooms and especially sex education policy, where “family values” warriors continue to insist that only abstinence should be taught in the classroom.

Sex education has been a controversial subject for decades as public school officials and parents have debated the best ways to help teenagers avoid unplanned pregnancies and sexually transmitted diseases. Not all states require schools to teach sex ed. But many states require sex education instructors to discuss or stress abstinence from sexual activity, with some schools offering abstinence-only programming, which urges kids to wait until marriage and often excludes information about contraceptives.

So-called “comprehensive” sex education programs teach students about abstinence, but (in a nod to hormones and reality) also teach about contraception, sexual health and how to handle unwanted sexual advances. Such curricula are gaining ground in some states.

In 2019, sex education continues to make headlines even as teen pregnancy rates continue to fall. Policymakers in Colorado, California and Alabama have pushed for big changes in the way sex education is taught there. In Colorado, a bill that would ban abstinence-only education in public schools awaits the governor’s signature. The legislation, which also requires that sex education be inclusive for students who are lesbian, gay, bisexual, transgender and questioning (LGBTQ), was “one of the most contentious battles of the 2019 legislative session,”according to the Colorado Times Recorder.

In Alabama–home of the recent law banning abortion even in cases of rape or incest– the state’s sex education law requires teachers to emphasize that “homosexuality is not a lifestyle acceptable to the general public and that homosexual conduct is a criminal offense under the laws of the state.” (A bill has been introduced that would change that requirement, but as this is written, it’s still on the books.)

As of 2016, abstinence was a required topic of instruction in states such as Arkansas, South Dakota and Texas…. 29 states, including Florida, Montana and Pennsylvania, did not require their sex education curricula to be based on medically accurate information. In some schools, teachers have been accused of inflating condom failure rates to discourage use.

I know that basing policy on evidence is out of favor in the Age of Trump, but the research is instructive: abstinence-only education results in higher teen birth rates. (And those “virginity pledges” that fundamentalist dads brag about? Researchers found that girls who took pledges were more likely to become pregnant outside of marriage when compared with girls and young women who did not take abstinence pledges.)

Facts are such inconvenient things.

I know it’s heresy, but maybe–just maybe–schools should teach kids medically-and-age appropriate information about their bodies, rather than inaccurate, incomplete or counterproductive information intended to mollify prudes and religious fundamentalists.

Comments