On a popular US poll twitter account, I found an interesting phenomenon. (Similar thing also has been happening on other US poll twitter accounts.)

Most of the polls it has (re)tweeted are favoring Biden by 10% nationally or so (except those conducted by https://twitter.com/Rasmussen_Poll). But all the polls which it itself conducted on twitter in past months always show 30%+ lead by Trump, and the most recent one is https://twitter.com/PpollingNumbers/status/1310007036294430725:

enter image description here

At least half of comments under the tweet are anti-Trump or pro-Biden, and so are half or most tweets on Twitter in general. My impression has been that Twitter is a social media with majority accounts having (neo)liberal ideology. I was wondering how to understand the contradiction which happens consistently and is echoed by another user? Is it because of the silent majority and shy Trump supporters on Twitter? Or is it because of interference by Twitter accounts outside US?



This is just a random twitter poll where anyone can vote regardless of being able to vote regardless of their actual voting status for the election. There is nothing that stops anyone from voting especially bots and the voters are all self selected which means the votes are not reliable.

The other poll you mentioned has procedures in place to make sure they only poll registered voters and work to make sure they get a representative sample in order to get accurate results.

  • You don't say it as strong as the other answer, but you still mention bots so I will also ask a similar question here. While bots very well may be able to vote, do you have evidence that this actually affected the result of a poll? – user29681 Sep 27 '20 at 20:05
  • @Chipster There is no evidence either way because outside of twitter itself there is no way to determine who voted on a poll. That being said there has been plenty of evidence in the last couple of years that bots are being used to support Trump so it would not be a bad guess to suggest the possibility here. – Joe W Sep 27 '20 at 21:27

In addition to what Joe W said, it's possible that it's not just poor sample selection. Online polling, especially on Twitter, is extremely easy to sabotage. Many votes are likely cast by bots or people with multiple accounts. In addition, neither bots nor people with multiple accounts are likely to post comments. It's not just selection bias, but willful disruption. Like the time Justin Beiber held an online poll to determine his next concert location, and pranksters voted for North Korea and won handily.

  • Many votes are likely cast by bots, eh while possibly true, do you have a source to back up that claim? – user29681 Sep 27 '20 at 20:02
  • @Chipster The use of automated bots to manipulate everything from online shopping recommendations to political trends is well known and documented. IIRC Twitter and Facebook have both banned million of accounts they identified as bot-controlled in recent years. I think the IT industry generally regards the banned accounts as the tip of an absolutely enormous iceberg. – StephenG Sep 27 '20 at 21:17
  • 1
    @Chipster lol there are literally websites for it now, it's ridiculous buytwitterpollvotes.com – Thymine Sep 29 '20 at 9:00

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .