Daily Presidential Tracking Poll

[bcking 3,157]

6 minutes ago, Bill & Katya said:

Aren’t all polls subjective?  Even voting exit polls are subjective and reliant on folks actually answering the pollsters truthfully.  Unfortunately, these pollsters are not the FBI, CBP or USCIS officers, so there is no penalty for answering dishonestly.  Regardless, news organizations, and political parties pay a lot of money for polling.

Of course all questions are subject to the person answering honestly.

 

However there is still a different between a question that has a factual answer ("Are you registered to vote" is an objective yes/no answer...it isn't an opinion, it's a question asking for factual information) and questions that have subjective "maybe" answers ("Are you likely to vote in the next election?" Or something similar).

 

Wording questions in surveys is incredible difficult and is it's own science. Questions should be validated in smaller populations before used in large scale surveys, though that is rarely done.

[Dashinka 24,792]

8 minutes ago, bcking said:

Of course all questions are subject to the person answering honestly.

 

However there is still a different between a question that has a factual answer ("Are you registered to vote" is an objective yes/no answer...it isn't an opinion, it's a question asking for factual information) and questions that have subjective "maybe" answers ("Are you likely to vote in the next election?" Or something similar).

 

Wording questions in surveys is incredible difficult and is it's own science. Questions should be validated in smaller populations before used in large scale surveys, though that is rarely done.

What is funny is that we get all this analysis of polling now.  When it was Obama in the Whitehouse and the MDL pundits brought these questions up they were essentially marginalized.  The simple way to do it is to automatically assume all polls are inaccurate.

[bcking 3,157]

13 hours ago, Bill & Katya said:

What is funny is that we get all this analysis of polling now.  When it was Obama in the Whitehouse and the MDL pundits brought these questions up they were essentially marginalized.  The simple way to do it is to automatically assume all polls are inaccurate.

From what I can tell fivethirtyeight has been doing their analyses of polls, and their ratings of polls, long before Trump was elected. I wouldn't say polls were considered "gospel" prior to Trump's election. Maybe by some people, but to be honest those people just aren't well educated in what polls are and what they can/can't do.

[jg121783 3,921]

So let me get this straight. Trump matched Obama's approval rating and he did it with 3 times more negative news stories about him than Obama did and some here are complaining about the poll being skewed 5% towards republicans (if that's even true)? Given the full picture I would say Trump is doing far better than Obama was at this stage of his presidency.

[bcking 3,157]

1 minute ago, jg121783 said:

So let me get this straight. Trump matched Obama's approval rating and he did it with 3 times more negative news stories about him than Obama did and some here are complaining about the poll being skewed 5% towards republicans (if that's even true)? Given the full picture I would say Trump is doing far better than Obama was at this stage of his presidency.

The skew that was mentioned is just based on comparing their results to the metaanalysis of all polls combined. At least based on my interpretation. FivethirtyEight tracks every poll Rasmussen puts out (along with all the other pollers) and over the hundreds of polls they tend to skew right a bit.

 

Now just looking back over the last few months the Rasmussen polls are responsible for the vast majority of"likely voter" polls (compared to "registered voter" polls). So the skew may more reflect the difference between those two types of polls. The other pollers that have used "likely voters" over the past few months have had similar numbers to Rasmussen, which would make me think it isn't an issue with Rasmussen, but just a bias in their shared methodology. (Polling likely voters tend to skew slight right, or if it makes you feel better polling registered voters may skew slightly left...but we have many many more polls of registered voters so we can be more comfortable with their reliability).

[Amica Nostra 4,803]

1 hour ago, jg121783 said:

So let me get this straight. Trump matched Obama's approval rating and he did it with 3 times more negative news stories about him than Obama did and some here are complaining about the poll being skewed 5% towards republicans (if that's even true)? Given the full picture I would say Trump is doing far better than Obama was at this stage of his presidency.

You are basing your opinion on one  poll that is a consistently skewed 5 pts toward Trump and 5 points against Obama making a 10 point bias in your analysis. You have doubled down on the error.

[bcking 3,157]

https://www.washingtonpost.com/news/politics/wp/2017/12/29/trump-celebrates-tying-obamas-approval-rating-in-trumps-friendliest-poll/?utm_term=.d4b714b70456

 

I only provide this article for the figures that they have. This highlights that Rasmussen's polling has always been lower than the "average" for Obama, and higher than the "average" for Trump.

 

I like the distinction fivethirtyeight makes between registered voter polls and likely voter polls, but I don't know if realclearpolitics makes the same distinction.

 

 

Similarly you can look at fivethirtyeights comparisons where they use ALL polls (you can filter for just likely/registered voter polls instead of all people):

 

https://projects.fivethirtyeight.com/trump-approval-ratings/voters/

 

Average over all polls Trump has an approval rating of about 39%. Using the same "average" over all polls Obama was at around 49% at this time in his first year (it's the little graph down the page under "How does Trump compare to past Presidents?").

 

As I've said before, i'd love to see them separate out "likely voter" polls with "registered voter" polls because I'm not entirely sure the "bias" Rasmussen has towards the right is due to them actually biasing that way based on how they weight polls. At least for Trump (They dno't have a list of all polls for Obama, at least not that I can find), the vast majority are "registered voter" polls and those that are "likely voter" polls all fall along the same lines as Rasmussen (though there are far far less of them...only 2 other pollsters).

[jg121783 3,921]

A couple things here. First of all the left has no problem holding up polls skewed towards democrats by 10 to 20 points as proof Hillary was going to win by a landslide (we all know how that worked out for you). But when any poll comes out favoring Trump the left screams "bias". Secondly you don't even address the fact that the media had 3 times as many negative stories (most of it made up garbage) about Trump as they did about Obama which no one can deny has a heavy impact on polls like this against Trump. As I stated above with all the negative propaganda Trump has had to deal with I would say he is doing well versus Obama as far as approval goes.

[bcking 3,157]

1 minute ago, jg121783 said:

A couple things here. First of all the left has no problem holding up polls skewed towards democrats by 10 to 20 points as proof Hillary was going to win by a landslide (we all know how that worked out for you). But when any poll comes out favoring Trump the left screams "bias". Secondly you don't even address the fact that the media had 3 times as many negative stories (most of it made up garbage) about Trump as they did about Obama which no one can deny has a heavy impact on polls like this against Trump. As I stated above with all the negative propaganda Trump has had to deal with I would say he is doing well versus Obama as far as approval goes.

Yes the media's portrayal of Trump has an impact, but we aren't really talking about WHY their approval ratings are different (or similar) here, but we are just talking about whether or not they are (similar or different). Throwing in "why" are they different is a very different conversation but I would agree that the media being negative l likely contributes. I'd also say if you rated every Trump speech, interview, tweet etc... that Trump himself has been far far more negative than Obama was in his first year (or any year). It isn't one sided negativity. 

 

As for the actual topic -

 

"Bias" isn't automatically negative. Bias doesn't mean they are INTENTIONALLY skewing themselves (they may be, I don't know). Saying Rasmussen polls are biased towards Trump, or towards the right, is just a statement of fact when you are comparing them to all other polls. The bias may be caused by their personal views influencing their work, but their bias could also be caused by their methodology and be unconscious. As I've said multiple times I think it is entirely likely that the bias isn't due specifically to Rasmussen, but perhaps due to their methodology since there are a couple of other pollsters that use similar methodology and end up with similar results. It is very likely that trying to select "likely voters" by whatever method they use tends to bias their polls to the right.

 

While many of these polls are large and therefore likely have good internal reliability, it is ideal to combine them all together to help weed out the methodological variability and how that can contribute to variability in responses. That is true whether we are talking about now, or whether we are talking about last year during the election. People were noting (though they weren't heard the most loudly) late October/Early November last year that Trump was polling close to within the margin of error compared to Hillary (when you look at all polls). If you look at the approval ratings right now by taking an average of all polls of likely or registered voters (or all Americans), There is still a ~10 percentage point difference between Trump and Obama's approval rating. 

[JimandChristy 1,689]

3 hours ago, jg121783 said:

A couple things here. First of all the left has no problem holding up polls skewed towards democrats by 10 to 20 points as proof Hillary was going to win by a landslide (we all know how that worked out for you). But when any poll comes out favoring Trump the left screams "bias". Secondly you don't even address the fact that the media had 3 times as many negative stories (most of it made up garbage) about Trump as they did about Obama which no one can deny has a heavy impact on polls like this against Trump. As I stated above with all the negative propaganda Trump has had to deal with I would say he is doing well versus Obama as far as approval goes.

Isn't it obvious why Trump gets three times as many negative stories as Obama did? It's all self inflicted on Trump's part. His behavior alone justifies it. He has completely demeaned the office of the Presidency with his childish antics and unprecedented falsehoods and lies, many times over than any other politician in recent memory. Obama, Bush, Clinton etc etc never stooped this low on a consistent basis. His impulsive demeanor and narcissism is going to end up being his downfall because he doesn't know when to shut up.  

[bcking 3,157]

1 minute ago, JimandChristy said:

Isn't it obvious why Trump gets three times as many negative stories as Obama did? It's all self inflicted on Trump's part. His behavior alone justifies it. He has completely demeaned the office of the Presidency with his childish antics and unprecedented falsehoods and lies, many times over than any other politician in recent memory. Obama, Bush, Clinton etc etc never stooped this low on a consistent basis. His impulsive demeanor and narcissism is going to end up being his downfall because he doesn't know when to shut up.  

He's the loudest child in class complaining about why other children tell the teacher he is being loud. 

 

He is responsible for his actions. We shouldn't just ignore all the negativity that he spews from his mouth (or puts forth from his twitter). 

 

I'd love to see an analysis of his comments (twitters, speeches, interviews) since starting presidency and the number that are negative vs. positive. Even in many of his positive messages, he tends to throw in an insult towards at least one person, group or organization. 

[JimandChristy 1,689]

4 hours ago, bcking said:

https://www.washingtonpost.com/news/politics/wp/2017/12/29/trump-celebrates-tying-obamas-approval-rating-in-trumps-friendliest-poll/?utm_term=.d4b714b70456

 

I only provide this article for the figures that they have. This highlights that Rasmussen's polling has always been lower than the "average" for Obama, and higher than the "average" for Trump.

 

I like the distinction fivethirtyeight makes between registered voter polls and likely voter polls, but I don't know if realclearpolitics makes the same distinction.

 

 

Similarly you can look at fivethirtyeights comparisons where they use ALL polls (you can filter for just likely/registered voter polls instead of all people):

 

https://projects.fivethirtyeight.com/trump-approval-ratings/voters/

 

Average over all polls Trump has an approval rating of about 39%. Using the same "average" over all polls Obama was at around 49% at this time in his first year (it's the little graph down the page under "How does Trump compare to past Presidents?").

 

As I've said before, i'd love to see them separate out "likely voter" polls with "registered voter" polls because I'm not entirely sure the "bias" Rasmussen has towards the right is due to them actually biasing that way based on how they weight polls. At least for Trump (They dno't have a list of all polls for Obama, at least not that I can find), the vast majority are "registered voter" polls and those that are "likely voter" polls all fall along the same lines as Rasmussen (though there are far far less of them...only 2 other pollsters).

 

Thanks for the link, interesting analysis, like the fact 538 take ALL polls into account and don't just cherry pick on one single poll that Trump is doing, stupidly. Rasmussen is most certainly biased towards Republicans and there will be pollsters that have a Democratic bias but that's why you take all polls and get an average. So it looks like right now we're looking at a 35-40% approval on Trump, the worst of any President in their first year of office.