Election Fact Checking Resources and Tips

Mick West

Administrator
Staff member
Election Metabunk Header.jpg
It was inevitable that this was going to be a race characterized by rumors, bunk, lies, and disinformation. The closeness of the race makes fact-checking all the more important. Here are some resources that can be used to figure out what's real and what's not.

Tip: many of these sources will be viewed as being partisan, or being run by a "deep state." Where possible dig down and find the most neutral source (such as the state election commission), or a source that's more likely to be accepted by your friend.


GOOGLE - Google should be your first step. Because of the huge amount of attention on this subject, any disinformation is being very quickly exposed as such, often in great detail by multiple sources. Just type in a few key words regarding the subject, include the word "election" and you will often get multiple high-quality results immediately. For example, try "sharpie election"
Metabunk 2020-11-05 07-31-03.jpg


FACTCHECK.ORG - Their main page has general election fact checks, and there's a page of Viral Voting Misinformation
[Update Nov 6, 2020] - New page: Trump's Wild, Baseless Claims of Illegal Voting

FIRST DRAFT NEWS - Up to the minute tracking of disinformation. Focussed heavily on the election now.

CISA.GOV - The Cyberstructure and Infrastructure Security Agency is a US Government agency, which may or may not make it more palatable. They have a detailed section on Election Security, which can supply very useful context. There's a sub-section on "rumors" that focuses on disinformation.

SNOPES.COM - While you might a "LOL Snopes" response, and they do lean left, they have a great fact-checking team who give understandable explanations with references. Here it's generally best to lead with the facts in the Snopes article and save the link for later. Quick lists on their Election Day Live Blog, and Election Aftermath Blog. Use their search, or browse the "Hot 50" list, which is pretty much all election right now.

POLITIFACT.COM - Again you'll get pushback because of the source. Focus on the verifiable facts. Cut-and-paste and use screenshots rather than just dropping links.

BBC.COM/NEWS - Foreign news sites might be seen as a little more neutral. The BBC has a few pages of fact-checks. (Nov 20)

Site-Specific Searches - Many outlets are hard to browse for specific lists. You can use Google to search a site, and curate your own list. This is useful if you want to use a particular source because you know it will be listened to. To do this add "site:" and the URL of the site, like factcheck site:https://www.usatoday.com

JOURNALIST'S TOOLBOX - A large selection of additional resources.

TIME LIMITED IMAGE SEARCHES - A useful way of finding current information is to do a Google image search limiting the result to the last 24 hours. During developing events this gives you an easily scannable set of relevant results. Here's how:
Metabunk 2020-11-05 08-04-30.jpg

Before limiting time, "sharpiegate" gives mostly hurricane-related results. Clicking on "24-hours" gives a full page of useful links regarding the election rumor.
 
Last edited:
Snopes also has liveblogs dedicated to election claims:

Article:
Live Updates: Snopes Coverage of 2020 U.S. Election Day
If the campaign season is any indication, misinformation is likely to fill social media feeds Tuesday. Get the facts here.

Trump Falsely Claimed Victory | Biden's 'Green Mile' Remarks | California Holiday Gathering Guidelines | Trump Tweeted "Racist Baby"? | Social Media Platform Election Efforts | McConnell and Cocaine? | Why Voters Wore Pearls | Amish Buggies for Trump | Obama Admin. Built Cages | Penn. Poll Worker Controversy | Biden on Work Visas | Trump in Dumpster? | Giuliani Misattributed Quote | Kullberg Network Reemerges | Trump's "Losers," "Suckers" Remarks | Pennsylvania Absentee Rumor | Georgia Election "Cheating" | Memphis BLM Shirt Controversy | 'Antifa's Plan?' | Sam Elliott Endorses Biden? | 'Illegal Campaigning' in Philadelphia | Voting Misinfo Robocalls Target Michigan Voters | Biden Campaign Logo is 'Three Red Banners?' | Philadelphia Polling Sign 'Violation' | FBI Investigating Biden Bus Incident? | A Guide to Voter Intimidation | 'Super-Predators?' | Doctors Get Paid for COVID-19 Diagnoses? | 'America is Great Because She is Good?' | Biden Bans Fracking? | Michigan's 'Man of the Year' | Background and Previous Claims

That blog is now closed, reporting continues here:
Article:
Live Updates: Snopes Coverage of the Aftermath of the 2020 U.S. Election, 11/04/2020
As states counted ballots on Nov. 4, misinformation spread rapidly online. Here are facts.

Claims from Nov. 5 | Moving Truck at White House? | Detroit Voter Fraud? | National Stress Awareness Day | Trump Warned Against Recounts in 2012 | AP Reports Biden Wins Michigan | 40k Georgia Ballots Rejected? | Trump's 'Major Fraud' Speech, Fact-Checked | Trump Wins Pennsylvania? | Biden Beats Obama's Vote Record? | What Would a Presidential Transition Look Like? | No, Biden Did Not 'Admit' Voter Fraud | Biden Wins Wisconsin | Ballots Marked With Sharpies Disqualified? | More Votes Than Wisconsin Registered Voters? | Did Biden Receive 100% of New Votes in Michigan? | Stalin: 'It's Not the Vote of People Who Count…' | Aria DiMezzo Loses | Trump Falsely Claims Victory With Millions of Votes Uncounted | Background and Previous Claims |
 
Last edited by a moderator:
POLITIFACT.COM - Again you'll get pushback because of the source.
Conservative View point and tips:


Don't post pictures that look like you immediately clicked on Snopes but bypassed Fox News. Believe it or not Fox News articles (vs. Tucker pundit show or Hannity pundit show) quite often do the debunking themselves. This is the first source you should check as it will be accepted the most. Then Brietbart or Blaze etc. Don't assume you can't find a debunk or explanation in a conservative source. I've always only used conservative sources on FB etc and i always find at least one that backs me up.

BBC is a joke and not even remotely neutral.


Politifact actually isn't 100% biased. Some articles do twist a convoluted explanation and then they mark it "false", but this only happens maybe 20-30% of the time. ex: they used experts to explain that California likely has only 2-4 electoral votes because of illegal immigrant census counts, vs the 6 electoral votes that was claimed. You can point such things out to your friends and they will trust the source much more.

But whatever you use, you cant just grab screen shots or cut and paste without providing a source link. We don't fall for those tactics.

edit add: and if it is family or friends you are arguing with, it would also help if you periodically debunk the Left bunk that's out there too. So you aren't just dismissed out of hand as being a rabid liberal shill.
 
TIP: Use numbers and unique phrases to find explanations of images and memes.

Example:

2020-11-05_10-21-34.jpg


Specific 4-digit numbers can be searched for, and add a little text. So search for: 4,788 duplicate
https://www.google.com/search?q=4,788+duplicate

This gives us both a Snopes article and links to the original news stories.

2020-11-05_10-24-32.jpg


So here we can see the lawsuit was in 2019, and it was dropped in January 2020 after the problems were fixed.
Article:
The Public Interest Legal Foundation, which filed its original suit in Detroit federal court in December, dropped it Tuesday after discovering Detroit Clerk Janice Winfrey and Elections Director George Azzouz had taken "remedial action."

"Defendants have taken action on the list of likely deceased registrants provided by the plaintiff," according to a Monday motion to dismiss the suit. "Further, almost all of the duplicate registrations that Plaintiff brought to defendants' attention have been corrected."
 
Good post.
It's all too easy to google your way into loquacious meaningless posts/ideas.
 
oops.. i was gonna recommend the Wayback Machine to prove information was not recently faked. But when looking at the maricopa County FAQ on ballots (which now says Sharpies are fine)
i got this
1604603539904.png

https://web.archive.org/web/20201001004918/https://recorder.maricopa.gov/site/faq.aspx

the new wording first recorded by wayback Nov 2nd does say they updated their machines. I think they just never updated their FAQ, as the same one as above shows in 2018. Which, of course, i can suggest using Wayback as a source.
 
oops.. i was gonna recommend the Wayback Machine to prove information was not recently faked. But when looking at the maricopa County FAQ on ballots (which now says Sharpies are fine)
i got this
View attachment 42038
https://web.archive.org/web/20201001004918/https://recorder.maricopa.gov/site/faq.aspx

the new wording first recorded by wayback Nov 2nd does say they updated their machines. I think they just never updated their FAQ, as the same one as above shows in 2018. Which, of course, i can suggest using Wayback as a source.
That seems to be a case of "Our old machines and the old ballot layout had issues with Sharpies, which we've fixed and sharpies are now fine.
According to: https://www.usatoday.com/story/news...e-controversy-arizona-false-claim/6164820002/
Maricopa County also explained in a Twitter video that the new tabulation machine only reads the ovals, so bleed-throughs are not a problem and that the new ballot style has off-centered columns, so bleed-throughs won't impact counting.

The county said it preferred Sharpies at voting centers because they dry much quicker, while ballpoint ink can smudge onto optical readers or other ballots which can create issues.
 
Not a pointer to a particular useful resource, but a pointer on how to recognise a useful resource. Alas, a bit of a two-edged sword:
Prefer high-information-content graphics, ones that don't just show you the overview that you're interested in but also convey the detailed data that is behind the aggregate data that makes the headlines.

Here's a very good example of an extremely high information content graph that I'd like to see more of:
7c629e71-a444-4eda-8239-c3c0e577923c_1137x825.png

I'll start with 3 very minor reasons why this particular example is sub-optimal before the why it's good parts:
- It's comparing a state-level election of local posts with a national election for president. However, when comparing changes over time as we are, picking the most relevant time is more important than picking the most relevant prior election, and it's at least clearly marked.
- It should be square, it seems the graphics guys are just addicted to graphs in landscape mode even when it makes no sense. A movement by N pixels in the X direction should be exactly the same in terms of what's being represented as a movement by N pixels in the Y direction, as they are the same measurement.
- Perhaps county-level dissection of the data might be too fine granularity, so it's a little noisy. Something slightly chunkier might be better.

Why it's good:
- it's fairly easy to get a feel for the magnitude of the swings in the various states from the bold lines; the -25 dem Georgians went to +/-0 dem, and the +/-0 dem Georgians went to +25 dem. So in middling constituencies there was about a 12% swing away from rep to dem.
- looking past the lines to the circles, you can see the fine granularity data. The smaller circles should be larger in number and broader in spread than the large circles, which is what we do see.
- you can even distinguish the likely-urban vs. likely-rural differences. The larger counties are urban and semi-urban, and indeed you see a larger proportion of those to the right (pro-dem) of the graph.

There are other features that you would expect to see, and which are present in this graph, but it's quite hard to describe (a) what they are and (b) why you should see them, but their presence again reinforces a presumption that the elections were conducted and counted fairly.
For example, the pinching of the curves towards the x=y line at the extremes - making this particular example a plano-convex lens, or split lentil - is exactly what you would expect. There's effectively a scaling factor of (m+100)*(m-100) (in margins) that prevents the data-points from deviating too far from x=y as m approaches -100 or +100. To see this, notice that a party receiving 10% of the vote one year (so -80 margin) would need to *double* its popularity in order to receive 20% (-60 margin), and a party receiving 90% of the vote (+80 margin) would need to achieve a clean sweep (+100 margin); neither is that likely.

I mention that particular final property because I seem to remember that in 2020/2021 that lenticular/parabolic shape was a core component of one of the claims *for* election fraud, despite it being evidence *against* it (weak, admittedly, but "it's just as you'd expect from a fair election" *is* a valid counter to "that distribution shows it's not fair"). That's one of the drawbacks of graphs that have high information content - if you don't know what you're looking at, a bad actor can tell you there's something there that isn't. Hence the "two-edged" caveat.
 
Back
Top