Fake Polls

We need to chat about polls. See here...

https://www.realclearpolling.com/polls/president/general/2024/pennsylvania/trump-vs-harris

The latest polls for Pennsylvania show a slight Harris lead. But look more closely.

Of the last eight polls up until the Bloomberg poll yesterday, five showed that Harris and Trump were tied, two had Trump ahead by 1% or 2%, and one poll had Harris ahead by 2%.

The Bloomberg poll has Harris ahead by 5%. Real Clear Politics averages the results of polls so a poll result like this skews the results in favour of Harris.

Harris does not have a 5% lead in Pennsylvania. As noted in previous blogs, the editorial team here at mewetree.blogspot.com thinks she will win the state by under 0.5%. But it will not be a landslide, as is suggested by this Bloomberg poll.

So how did this poll come about?

Pollsters will skew results by polling people who they know will favour their own favored candidate. This can be done by polling people in geographic areas known to support the pollster's favourite candidate, or by having less balance to the demographic spread than is optimal, and polling demographic groups that are known to favour their favourite.

In short, pollsters actually know who likely favours whom in terms of voting preferences and where those people live, so to skew a poll they just make sure to ask those who they know will give them the answers that they want.

To see how easy it is to fake a public opinion poll, see here - this is the classic explanation...

https://www.youtube.com/watch?v=ahgjEjJkZks

The key is this, "...so they don't mention the first five questions, and they publish the last one..."

It's not a matter of "Trust no none!", but it might be a matter of "When things look out of the ordinary and/or look illogical, they are usually just so."