Week of 11/15: Regulating Facebook
Summary: Right-wing messaging that paints Facebook whistleblower Frances Haugen as a Democratic operative was effective at reducing Republican support for stricter regulation of Facebook. Pro-regulation messages using conservative values or a Republican Senator as messenger were effective at reversing that effect.
This research was conducted in partnership with Strategic Victory Fund.
Right-wingers have been spinning up conspiracy theories attacking the credibility of Facebook whistleblower Frances Haugen.
Haugen’s testimony before Congress highlighted that Facebook knew their products were harmful to children, and that they’d withheld this information from Congress in their own hearings. Both Democratic and Republican elected officials reported finding Haugen’s testimony compelling, and expressed an openness to government intervention.
That said, government officials having a belief doesn’t necessarily mean the public is on board.
We had three goals for this week’s research:
- understand people’s current opinions, beliefs, and opinions about Facebook and what kind of regulation it ought to face,
- explore the impact of the right-wing conspiracy theory message on support for regulating Facebook, and
- measure the effects of messages designed to counter the right-wing message.
Audience Understanding Survey
On November 16, we conducted an Audience Understanding Survey, asking 782 people the following question:
Which of the following is closest to your view, even if none is exactly right?
- Facebook is harmful to American society and should face much heavier government regulation, including being broken up into smaller, less powerful companies
- Facebook is harmful to American society and should face heavier government regulation, but shouldn’t be broken up
- Facebook is harmful to American society but shouldn’t face heavier government regulation
- Facebook is not harmful to American society
- Don’t know
We grouped people according to their response to that question, and followed up with questions about their values, feelings, and demographics to build up an understanding of each group.
People largely believe Facebook is harmful; split on need for government regulation
About 44% of people in our sample favored more government regulation of Facebook, while 41% opposed. There was essentially no partisan polarization on this question, in contrast to every other issue we’ve tested this fall.
In the open-ended questions, people expressed a wide range of different reasons for finding Facebook harmful. Some identified selling or leaking personal data as the big issue. Others flagged the mental health harms of using Facebook products. Others were upset about Facebook censoring conservatives, or censoring liberals. There wasn’t a unified theory of the case against Facebook.
Among people who opposed more regulation of Facebook, some commented that the government should only get involved when “punishable behavior and crime is present on Facebook” or that the government should only be involved “to regulate unlawful things.”
In every group, even those who said Facebook should face “much heavier government regulation, including being broken up,” people expressed skepticism about the government. In some cases, people described overcoming that skepticism because Facebook is so bad, but in others, skepticism of government seemed to drive their opposition to regulation.
Rapid Message Test
For this test, we first showed respondents a message that opposed further regulation of Facebook (referred to as the “oppo” message throughout this summary). That message was from right-wing commentator Ben Shapiro, attacking Haugen as “a partisan activist,” with “a record of donations to far-left Democrats,” among other things. See the appendix to read all the messages in full.
In response, we drafted four messages by making use of respondents’ open-ended answers from the Audience Understanding Survey, varying the values our messages were based in, and selecting both governmental and non-governmental messengers.
Two messages shared messages from elected officials, one from Republican Senator Roger Wicker, and one that quoted both Democratic and Republican electeds. Though both messages described harms to children, each used a slightly different persuasion tactic. The Republican Senator message used an “unlikely messenger” in an anti–regulation Republican official, while the Bipartisan message was meant to convey the consensus among both parties’ representatives.
We looked for opportunities to reflect respondents’ language back to them, using the word “crime” in the Bipartisan message, and crafting an entire message about Facebook’s violations of the law. Lastly, we wrote a message based on protecting the vulnerable that focuses on harm to children, from the perspective of a concerned parent.
Respondents were randomly sorted into one of six groups:
- Those who saw only a placebo message
- Those who saw the Ben Shapiro message and then one of our four messages
- Those who saw the Ben Shapiro message and then a placebo message so that we could measure the effectiveness of the Shapiro message alone.
After seeing their message or messages, respondents were asked the same survey question as in the Audience Understanding Survey. In the results, we report the percentage of people who selected either of the responses that calls for greater regulation of Facebook.
The message attacking Haugen and opposing regulation of Facebook drove support for regulation down by 10 points compared to the pure placebo group — in the graph below, compare the gray bar on the far right to the green bar on the far left.
The pro-regulation messages were generally effective at reversing that effect, though. People who saw the Ben Shapiro message and the “Rule of law” message or the “Bipartisan” message actually ended up slightly more supportive of regulation than those in the placebo group.
Support for regulation among people who saw the “Oppo” message and a placebo was 10 points lower than the pure placebo
Looking at subgroups of the population, we can see that the anti-regulation message had no real effect among people who voted for Biden, but reduced support for regulation by 20 points among those who voted for Trump.
Trump voters were persuaded by the oppo message while Biden voters were not
Among those who voted for Trump, the messages that made use of conservative values or a Republican messenger were effective. “Rule of law” was based in authority and order, while “Republican Senator” used the words of Mississippi Republican Senator Roger Wicker. Each of those entirely erased the drop in support that the anti-regulation message caused, and may have even slightly increased support for greater Facebook regulation among Trump voters.
In the end, we have to acknowledge that the conspiracy-minded message spread by right-wing “provocateurs” was effective, if only with Republican partisans. But its effect—to reduce Republican support for regulating Facebook—wasn’t durable, as it could be erased with a pro-regulation message, especially if that message was tailored to Republicans through messenger or values.
The Daily Wire reports:
“The Facebook whistleblower, Frances Haugen, is a partisan activist, who has a record of donations to far-left Democrats, according to official records. She also has a history of raising issues about purported bias while at previous employers.
She is working with Democrat operatives to roll out her complaint and has the same lawyers as the anonymous Ukraine ‘whistleblower’ whose allegations led to Donald Trump’s impeachment, but who reportedly turned out to be then-Vice President Joe Biden’s top advisor on the country.”
Congress should reject Haugen’s call to impose burdensome government regulations on Facebook.
Rule of law
Facebook has no respect for the rule of law. They violate the law with their anti-competitive practices. They violate the law when they hand over Americans’ private information to unauthorized parties. Both the Trump and Biden administrations have pursued legal action against Facebook, but have not yet been able to rein the company in.
This is a country of law and order, and we must insist on stricter regulation and oversight that will force Facebook to comply with American laws.
Protect our children
I’ve enjoyed using Facebook and Instagram to keep up with friends and family. But my girls were telling me they can’t stand these apps—all they see is an endless feed of perfect bodies and faces and it makes them feel bad about themselves.
Then I learned that these companies’ own researchers proved that they’re causing mental health issues for kids, and they haven’t done anything to fix it! Even Facebook should know we have to protect our children.
I’m going to quit using these apps, but I know nothing will really change until the government forces them to change their ways.
Senator Roger Wicker, R-Mississippi:
“The Wall Street Journal exposed company research showing that Instagram, Facebook’s photo-sharing app, has negative effects on children. Facebook researchers found that Instagram makes body image problems worse for teenage girls and makes teenagers feel worse about themselves. Facebook also found that teens blame Instagram for higher rates of anxiety and depression.
Although Facebook has known about these harms, it has actively withheld the truth from the public.
Congress does not agree on everything regarding Big Tech, but there is a growing bipartisan consensus that more must be done to defend the interests of children and the broader public against these powerful companies.”
Representative Bill Johnson, R-Ohio
“Big tech is essentially handing our children a lit cigarette and hoping they stay addicted for life.”
Representative Ken Buck, R-Colorado
“I think that people [at Facebook] who realized that there was an increase in teen suicide rates, and that there was a relationship between their product and that increase—and they continued doing what they did—should be held criminally liable.”
Senator Ed Markey, D-Massachusetts
“Facebook won’t protect young people. It’s obvious that Facebook only cares about children to the extent to which they are of monetary value.”
Key experiment details
- Audience: All adults, balanced on age, race, and gender
- Geography: the United States
- Sample size (raw / weighted): 1,199 / 1,199
- Dates in field: Wednesday, November 17, 2021 to Thursday, November 18, 2021
- Weighting factors: age, race, gender, education, and party