Public opinion polls can be valuable snapshots of what people are thinking at a particular point in time. But, in our view, election polls are too often used by media organizations as a way of saying who’s “winning” – or who’s going to win, no matter how far away the election is. That, in turn, encourages us to treat a campaign like it’s a sports season, instead of focusing on the issues at stake in an election. Here’s how we’ll handle polling:

• We will not publish or air our own or partner stories about who is leading whom in whatever race, according to XYZ poll. That means you will not see a headline or hear a story that only reports on poll results showing who’s “ahead” in the race, like, “Candidate X closes gap on candidate Y, new poll finds.” We believe the drumbeat of those kinds of stories diverts attention away from issues.
• We may use polling information in other kinds of stories to provide context, or to provide  information that can help you assess a poll’s credibility. For example, in 2022, two polls came out almost at the same time showing very different numbers in the Pa. Senate race. Anticipating that people would question why they were different, we published a story headlined “How do polls account for undecided voters in Pa.’s Senate race?”
• We may use polling information as part of our reporting on stories that arise from listening to people in our community. For example, if a group we’re talking with raises a certain issue as important, we may look at polling to see whether that issue is resonating with Pennsylvanians statewide.
• When we do use polling information in a story, we’ll include language explaining the limitations of polls generally, and the specific poll’s methodology and margin of error; and if possible, we’ll cite more than one poll to give you a more balanced look at what polls are saying. Our goal will be to give you knowledge you can use to assess polls. 

Generally speaking, we will resist saying someone “lied” when reporting on a public comment, in large part because it is difficult to prove, and may require evidence such as a recording or written statement.

An exception: We will use the term “election-fraud lie” to describe the systematic, organized campaign led by former President Trump to try to invalidate millions of legally cast votes in his effort to stay in power. And, because the evidence is clear that Trump knows his claims are false, yet continues to repeat them, we may use the word “lies” when referring to his comments about the 2020 election.

Otherwise, when a public official makes a statement that’s false according to evidence, we will frame the story around the facts, not around the false statement. Other acceptable references include “falsely stated,” “claimed without evidence,” “said incorrectly,” “spread misinformation about” and so on. But use of those terms must be accompanied by the relevant facts.

Importantly, when possible, we should avoid amplifying a politician’s false statements. Part of their strategy is repetition, which helps bad information take root. We want to repeat facts, instead.

In January 2021, we began holding accountable the state and federal lawmakers who took specific actions in support of the 2020 election-fraud lie. In any story that quoted one of those lawmakers, we added a line that noted the action the lawmaker took, and that the election-fraud lie led to the attack on the U.S. Capitol, which was an attempt to overthrow representative democracy in America. (We used a shorter version of that on air.) We created the accountability policy because we believe it is important to keep the facts public and visible regarding who supported an effort that shook the foundations of America’s electoral system.  

 We eventually re-focused the use of the policy, and now apply the language mainly in stories that have to do with elections, voting or democracy. However, we may also use the language when it is appropriate as context in a news story – for example, when a legislator who supported the election-fraud lie announces they aren’t running again. 

Here is what we’ve written about that policy:

• WITF is connecting these lawmakers to their actions on the election-fraud lie. Here’s why (Jan. 28, 2021)
• Countering the big lie: WITF’s newsroom will connect lawmakers with their election-fraud actions (Jan. 28, 2021)
• WITF focuses election-fraud lie accountability policy on lawmakers involving key areas of democracy (June 20, 2023)

The American system of democracy includes core tenets such as everyone who’s legally able to vote should have access to the ballot; all legal votes will be counted; and power will be transferred peacefully. An annual report from The Economist Intelligence Unit, a division of The Economist magazine, rates the U.S. as a “flawed democracy,” with its lowest ratings in “functioning of government” (how well a government acts on behalf of people) and “political culture” (how well electoral losers accept the voters’ decision and carry out a peaceful transfer of power).

Our coverage will explain how our democracy works, how it could improve, and when people are attempting to break the system. 

We believe that news organizations cannot cover elections as though Jan. 6 was a policy disagreement. The 2020 election-fraud lie, the Capitol attack and the continued support of those attacks on democracy threaten the very system Americans use to settle our policy differences at the ballot box. We believe that using democracy framing can elevate facts over falsehoods, and can arm people with information they can use to protect and improve our democratic system.

At WITF, democracy framing means that in stories relating to democracy, voting and elections, we:

• Lay out the stakes: How does a story fit in to the broader threat to democracy, including attempts to disrupt or overturn future elections?
• Establish context clearly and early in a story: How does this news connect to elected officials’ actions to try to invalidate legal, certified votes in the 2020 presidential election.
• Repeat facts to counter the repetition of lies and misinformation.
• Anchor the story in the moment, connect it to the broader picture with relevant context, and avoid signaling that this issue is politics as usual.
•  Work to address potential solutions and provide information that can help us work toward strengthening democracy.

Journalists are trained to be independent of the people and organizations they report on. It doesn’t mean we don’t have opinions on things – we’re human beings, just like you are. But when we’re at work, we have to put our personal feelings aside and focus on providing the best possible information for listeners and readers, without favoritism.

Strong newsrooms work to catch and correct when a reporter or editor is guided by personal views rather than being open-minded, curious and independent. At WITF, we work hard to make sure we’re including various perspectives in news stories.

But staying true to that core value is a challenge, made harder by Donald Trump’s inaccurate descriptions of legitimate news organizations and journalists. And, unfortunately, some Republicans at state and local levels of government, as well as many people who support Trump and/or the Republican Party, believe and amplify his messaging.

So, when we report facts – for example, that Trump tried to remain in power despite losing, and that he continues to tell lies about election fraud in 2020 – a good chunk of people who live in central Pennsylvania view that as partisan coverage.

But our coverage will never support, or oppose, a candidate or party. Our mission is to provide factual information, with context, that people can use as they go about their daily lives or make decisions on which political candidates to support. That mission is what leads us directly to using language that, for example, explicitly states when an elected official says something that contradicts known facts. Party affiliation and party platform just don’t matter. 

As special projects editor Tim Lambert and I have often said when people criticize our 2020  election accountability policy for being partisan: Lawmakers are on the accountability list not because of anything they said, but because they took one or more of four actions: Signed a state House or state Senate letter urging Congress to delay or vote against certifying Joe Biden’s victory; signed on to Texas’ Supreme Court lawsuit that would have invalidated legal Pennsylvania votes; or voted against certifying Biden’s win. If any Democratic politician had done any of those, they’d be on the list. To do otherwise would be to abandon our mission to report facts and context, without partisanship.

It’s important to note that when we say we stand for facts, that means following the facts wherever they lead – not just if they support a certain narrative – and correcting factual mistakes when we make them. 

This one seems like it should be easy, but it’s hard to stay out of the weeds because there are so many layers to it – and because labels can convey information, but they can also be overused to the point that they become essentially meaningless. Here is our thinking:

• Avoid labels when possible, and convey information instead. Examples:
  • Instead of “far-right Congressman Jim Jordan,” say “Jordan has spread disinformation that the 2020 election was stolen.”
  • Instead of “far-left Senator Bernie Sanders,” say “Sanders is a Democratic Socialist, whose party platform includes gradually reducing police budgets to zero and free(ing) ‘all people from involuntary confinement.’”
• If someone self-describes with a label, it’s OK to use that description. Best practice would be to attribute the label to the person so it is clear to a listener or reader where it’s coming from.
• Avoid labels for events or movements, instead using information: An event was “a rally to support Trump’s effort to overturn the 2020 election” or an event was “a rally to support Biden’s re-election.” Avoid using campaign slogans.
• We don’t use the word “lie” lightly. But we have examined the facts, including those laid out in news reports, court documents and decisions, and testimony to the Jan. 6 committee. We’ve decided that we are comfortable using the term “election-fraud lie” because, based on evidence, it describes an organized strategy with identifiable actors and actions spread across the country as part of Trump’s attempt to stay in power despite his loss.
  • There is no evidence that there was fraud in the 2020 election, according to sources including a voter data expert hired by Trump, multiple federal agencies and court cases, a study of Trump’s statistical claims, a comprehensive fact-check by The Associated Press, a Brookings Institution study, a Republican study into post-election lawsuits, and a Trump pollster’s analysis that explains why he lost. and which does not mention the possibility of fraud or voting irregularities.

We’ve heard from people – mainly right-leaning readers/listeners – that we’re quick to ID politicians as Republicans if the story is perceived as “negative” or controversial, and don’t always do that with Democrats; also, we sometimes hear that we ID politicians by party in stories where political affiliation is or appears to be irrelevant. 

We’ll still ID politicians by party, because it is a piece of information that readers or listeners can take in and decide how much weight to give. We can also add context in stories about whether the topic reflects the position of the political party, in order to differentiate between newsworthy actions of a single legislator and that lawmaker acting as a representative of a larger political platform. But, we’ll avoid it when possible in a headline. We’ll assess where in the story to put party affiliation, based on how relevant to the story it is. We’ll be alert to use of party ID when readers or listeners could perceive a story to be “negative” or controversial.  We’ll work to make sure we’re consistent across political parties. Whether the person is a Democrat, Republican or other party will make no difference in any decisions we make.

These are a staple of political campaigns, allowing candidates to meet people on their terms and, in the case of stump speeches, to control their message. That’s especially true of presidential campaigns – and Pennsylvania is going to see a lot of President Biden, Donald Trump, and their political surrogates in 2024. In 2020, because of then-President Trump’s demonstrated history of falsehoods and inaccurate statements, we adopted a strategy of not simply reporting what a candidate said, and finding other ways to cover an appearance. That will be reflected in our strategy for 2024:

• Limited staff means we cannot commit to covering every candidate appearance, so we will make choices about which appearances to cover. Those choices will be made based on factors including whether the candidate is addressing an issue of significant local interest or is proposing a new policy that affects central Pennsylvania. Party politics will play no role in those choices.
• When we do cover appearances, our guidance includes:
  • We’re not a national media organization, and we will look for the value we can bring to our audience.
  • We’ll resist amplifying candidates’ statements, instead focusing on voters and their issues at these events. If a candidate is addressing an issue of local interest or one that affects central Pennsylvania, our focus will be to report on what voters say, not on repeating what the candidate says from behind the microphone.
  • We’ll consider fact-checks as the mode of coverage.
  • We’ll fact-check within a story when necessary/appropriate.

Generally, we are comfortable reporting when a candidate declares they are running. If we miss such an announcement, we can report on them after the filing deadline. In some races, however, prospective candidates like to talk (or have others talk for them) about the possibility that they might run. It’s a way of getting extra news coverage and exposure. We’ll refrain from reporting or running those kinds of speculative stories. And when candidates do declare they’re in a race, we’ll work to frame a story around issues, not around the person.

A lot of candidates like to get interest groups or other candidates to endorse them, and use that to send a message to voters. Often, though, the endorsements are predictable – like, the teachers’ union endorses a Democrat, while an anti-abortion group endorses a Republican. They can also be too numerous to keep up with. But we do hear from people through community engagement that endorsements are pieces of information that can help them understand where candidates are coming from. Here’s how we’ll handle endorsements:

• We’ll use news judgment, reporting when an endorsement could tell us something significant about a candidate.
  • In those cases, we’ll use language in stories like, “This endorsement is significant because…” and explain the context/facts that can help you assess what the endorsement means to you.
• As a practical matter, we won’t report individual stories about most endorsements. But we will make an effort to gather endorsements and list them in stories about particular races, so you can have that information as you’re deciding who to vote for. 

If you hear something like that, please let us know (see contact information above). But the most likely reason is going to be that what you heard was an NPR report on Morning Edition or All Things Considered. We do not control, or have influence over, what NPR airs on their time slots during that programming. We do control our local reporting and the local newscasts. 

For example, NPR might air a story about the latest presidential poll results. Because we air their programming as part of ME or ATC, you will hear that story. But we pledge that you will not hear a story like that produced by us on our local newscast.

This is a living document that will be added to, updated and/or revised. We invite you to raise issues we can address here, ask questions about what you’ve read, or make other suggestions. We hold ourselves to high standards, and we count on you to help hold us accountable.

You can reach me at scott_blanchard@witf.org or at 717-910-2805.

These guidelines were created by WITF editors Scott Blanchard, Tim Lambert and Randy Parker, and WITF Vice President of Journalism Tom Murse, with input from WITF staff; Joy Mayer, executive director of Trusting News; and Sue Robinson, Helen Firstbrook Franklin Professor of Journalism at the University of Wisconsin.

Media researcher Claire Wardle of First Draft News, an independent nonprofit that works to fortify journalistic institutions and the public against misinformation, defines misinformation as false information spread by someone who believes it is true.

Disinformation, she writes, is “intentionally false and designed to cause harm.” People who create and spread disinformation may do so for money, political power, or to cause chaos.

Once disinformation is unleashed on social media, however, users who believe and share it may unknowingly spread false information. Friends and family members who trust those users are more likely to trust the information.

Wardle, who co-founded First Draft, describes a third category, “mal-information” — something that is true but disseminated to cause harm, rather than to serve the public interest.  

First Draft’s materials package all these terms into a spectrum of “information disorder,” which includes lies, conspiracies, and rumors as well as “hyperpartisan content” and doctored images, photos, or audio.

Wardle divides misinformation into seven types that range from satire, whose creators aren’t trying to fool anyone but sometimes do, to content that is completely invented and intended to cause harm.

Much of the misinformation out there includes some things that are true and some things that are false, which makes it more difficult to recognize. Here are some of the ways misinformation has spread this year and in past election cycles, and some tips for discerning it:

Posts that misrepresent voting information: Among the ads, posts and fake accounts Russia used to influence the 2016 election, according to evidence tech companies submitted to Congress, was a post that told people to “avoid the line” and tweet their vote. Other graphics mimicked Hillary Clinton’s campaign and encouraged people to vote by text. Another report on the Russian Internet Research Agency’s disinformation campaign discovered a major focus was encouraging Black voters to boycott the election. Similar tactics targeted Mexican American and LGBT voters to a lesser degree. And the agency targeted conservatives with messages intended to stir outrage, division, and support for President Trump, according to the report. 

Another hoax (this one not attributed to Russian interference) showed a fake photo of Immigration and Customs Enforcement arresting people at the polls, but the nonprofit news organization Propublica found the ICE agent and the person being arrested were photoshopped into an image from a previous election. The account that posted the doctored photo was also fake, and had been featured in previous news stories. Reporters found the original images by a simple Google search, but the idea was pervasive enough that the same rumor (still completely false) surfaced again in 2018.  

Tip: If you have any questions about how to vote, call your county election office or visit its website. You can also go to the Pennsylvania Department of State’s voting website, www.votespa.com or, if it’s election day, the state hotline, 1-877-868-3772. A coalition of civil rights and lawyers groups also takes calls from voters who encounter questions or problems via several hotlines for English, Spanish, Arabic, and Asian languages. You can also check WITF’s election page for voting information.

Posts that promote false claims of election cheating: This year, the most prominent examples of voting misinformation surround mail-in ballots, and Pennsylvania is a hotspot. 

Propublica and First Draft analyzed Facebook posts earlier this year, finding that the platform “is rife with false or misleading claims about voting, particularly regarding voting by mail.” These posts saw some of the highest engagement among voting-related content and included “conspiracy theories about stolen elections or outright misrepresentations about voting by mail by Trump and prominent conservative outlets.” The report described claims from the political right and left that opponents would steal the election.  

Cybersecurity researchers and tech companies worry, above all, that disinformation surrounding mail-in voting and voter fraud could jeopardize the election, especially if it takes longer than usual for votes to be counted. NPR spoke to experts who believe “bad actors” within the U.S. may be a greater threat than Russia, which has been discovered interfering this election cycle. A group of Harvard researchers says Trump’s use of mass media, primarily television networks, is a more influential source of disinformation than what happens on social media.

Tip: Remember that it will likely take longer than usual to count the votes, and it will be more important than ever to fact-check any claims of fraud or corruption. PolitiFact and Snopes are well-known fact-checkers; below you’ll find a longer list of fact-checking websites and tools for evaluating online sources.

Mislabeled and doctored media: Deepfakes, or false video or audio clips that use artificial intelligence, haven’t been as much of a problem this year as some experts feared.

As NPR reported, misinformation has been spreading via less advanced, less expensive means. That includes everything from photos, videos, quotes, and articles taken out of context or misrepresented on social media to edited content (like the ICE agent dropped into a photo of a polling place) or made-up captions.  

Tip: Remember that articles, memes, photos, and videos shared on social media might be severed from their original context. Go back to the source and then see if you can verify the information somewhere else. You can find out where a photo came from by doing a reverse-image search; fact-checkers are also useful in verifying viral posts. As for deepfakes, you can learn about efforts to detect them here.

Fake accounts: Disinformation campaigns rely on bots, or fake accounts, to exacerbate divisions in society. 

A recent example: a group working to combat disinformation in Latino communities discovered that bots were amplifying the controversy around Goya Foods this summer, after the CEO praised Trump during a White House visit and critics of the president promoted a boycott. Twitter recently suspended a group of fake accounts purporting to be Black Trump supporters; the accounts picked up tens of thousands of followers in less than a week and collectively more than 265,000 retweets “or other amplifying ‘mentions.’” 

Tip: Keep in mind the source and motivation behind information, especially when it confirms your beliefs or generates an emotional response.Multiple studies have described how misinformation plays on our preconceptions and our psychology. We tend to believe claims we see (or hear) repeatedly, for example, and frequently assume things are true if they haven’t been corrected. We are more likely to believe something that aligns with our beliefs and more likely to share information that elicits joy, rage, awe, or disgust. 

Agenda-driven “news” sites: Groups with political or economic agendas sometimes set up sites where they publish misleading information to support their interests, using the guise of a local news outlet and mixing in reporting from other sources to gain readers’ trust. These sites have proliferated in 2020. Here is a list of digital outlets to treat with skepticism.

Tip: Other helpful resources include a list by Melissa Zimdars, an assistant professor of communication at Merrimack College, of suspicious “news” sites and ways to recognize them. Journalists behind a tool called News Guard evaluate sites based on standards of credibility and transparency. And a company called Ad Fontes Media created a “media bias chart” which can give you an idea of what that company believes is the political bent and reliability of a variety of sources.

Conspiracy theories: Between the pandemic and the proliferation of QAnon, conspiracy theories abound this year and may have an impact on some voters’ decisions on Nov. 3^rd. 

Trump’s coronavirus diagnosis sparked conspiracy theories and rumors, many of them coming from the political left. 

According to the Pew Research Center, about half the adults in the U.S. have heard of QAnon, which falsely claims that Trump is fighting a sex-trafficking network made up of members of the Democratic elite and an expanding lineup of famous people. The conspiracy has fueled threats and violence, and the Federal Bureau of Investigation has labeled it a domestic terrorism threat. In recent months, QAnon supporters have spread misinformation in online groups dedicated to fighting human trafficking and potentially influenced voters. Facebook just announced it would take down Instagram accounts, pages, and groups promoting the conspiracy theory.  Twitter removed 7,000 accounts this past summer, but researchers say 93,000 still exist on the platform.Tip: Many of the previous tips could apply here. Conspiracy theories can provide a sense of belonging, according to Cambridge University researcher Thomas Roulet, and tap into values like independence and even truth-seeking, but they can also be dangerous, eroding public discourse, distracting from real, pressing issues, and making it harder for people to get information they need. Having conversations with friends and family and leaning on fact-based information is a start.

New York University researchers predicted late last year that Instagram would be the “vehicle of choice” for using memes to spread disinformation and that WhatsApp would play a significant role. 

WhatsApp has become a major source of news for people around the world, a Reuters report found this year, and young people especially are relying on Instagram and Snapchat for information. A study of Russian disinformation in 2016 found that it was more successful on Instagram than Facebook. Conspiracy theories have also flourished on YouTube and TikTok, which is confronting a problem of widespread hate speech.

WhatsApp presents special challenges because information can spread via private chats, where misinformation is harder to identify and track. The impact of misinformation on the service and others like it has caused alarm around the world. As Politico noted in a story about misinformation circulating in Latino communities in Florida, WhatsApp is popular among immigrant communities because it allows people to communicate from around the world. Organizations like Univision and the Turkish fact-checker Teyit have rolled out stickers that users can send to family members to politely alert them to misinformation. 

Tech companies have become more aggressive in their approaches to misinformation. Facebook, which owns Instagram and WhatsApp, has partnerships with third-party fact checkers and donated to an international fact-checking organization to combat misinformation across its two subsidiaries. It also cut the number of times a user can forward a message on WhatsApp from 20 to five. YouTube and parent company Google are also rolling out anti-misinformation policies amid the election, as are Twitter and Facebook. This article from Vox includes tips for how to use the two platforms’ settings to get more reliable information on your feeds.

The good news is you don’t have to be a professional fact-checker to help combat it. Researchers at the Massachusetts Institute of Technology are finding that small groups of “regular people” can be as effective as fact-checkers at deciphering false content online. Crowdsourcing, they say, could be a useful way to help share the burden.

On an individual level, you can help by pausing before you share content that incites an emotional reaction, “anger, disgust, or fear” as well as hope or excitement, especially if the information supports your views. It’s also important to let friends and family members know, if you can, when they share something false or misleading.

Here are some resources to help:

Fact-checking websites & pages

• FactCheck.org
• PolitiFact
• Snopes
• The Washington Post’s Fact Checker
• The New York Times’ Daily Distortions
• First Draft News’ Briefing

Tools and training for recognizing false or misleading sources:

• WITF & Trusting News: Defining Credibility & Trustworthiness in Journalism
• The News Literacy Project
• The Center for News Literacy
• NewsGuard Ratings
• Ad Fontes Media’s Media Bias ChartList of dubious local news sites
• List of false and misleading sources and ways to recognize them

Sources for voting information:

• VotesPA
• County election offices
• WITF’s Election 2020 page