Talk:Rasmussen Reports

Conservative

Previous editors had concerns about describing the pollster as objective. This was revisited several times, including during a serious of edits describing the pollster as right-wing. Newsweek has now described the pollster conservative. This on-top of the other analyses reported by sources describing the pollster as having a significant and persistent bias. Any edits to the alternative will need to provide several independent sources.

As per WP:RSP "Unlike articles before 2013, post-2013 Newsweek articles are not generally reliable." Wikipedia uses reliable sources. An unreliable source can't be introduced by a single editor to unilaterally overturn a closed RfC. Chetsford (talk) 20:59, 22 September 2020 (UTC)

Point of discussion

Does this analysis [1] of the content of a tweet violate WP:OR? Chetsford (talk) 05:25, 17 December 2020 (UTC)

You people are unbelievably ignorant.

Do you describe the hundreds of other polls as "left wing"? Because they are

 You just don't recognize that because you have been manipulated to believe that left wing is just "the norm."

This article is the embarrassing. Do you really believe that things such as this make people trust Wikipedia more? Wikipedia is becoming a joke, and it's becaus of articles like this. 12-year-old children are running this circus. 199.66.65.246 (talk) 08:50, 23 October 2022 (UTC)

I removed 'conservative' from the first sentence because it's unwarranted and consensus from a couple years ago was to omit the similar term 'right-wing'. But in general you will get better results from a reasonable argument rather than just coming here to insult everybody. –CWenger (^ • @) 19:31, 23 October 2022 (UTC)

2024 and they still won't budge on the "bias" mention even after Rasmussen polls predicting the 2024 election with great accuracy and no detectable bias.

The best I've gotten from them is "Washington post says so" 173.238.207.7 (talk) 00:33, 8 November 2024 (UTC)

Yes, as I have explained to you several times, "Washington Post says so" is the only justification needed. That's how Wikipedia works.

Also, I would suggest that if you're trying to get consensus for your proposed additions, immediately accusing everyone around you of bias may be more harmful than helpful to your own goals. AntiDionysius (talk) 00:36, 8 November 2024 (UTC)

It’s ok to be white poll

The Poll That Did in Dilbert Creator Scott Adams Is Even Dumber Than You Can Imagine Doug Weller talk 19:21, 28 February 2023 (UTC)

Here's another one. I think the WP:RS coverage of this incident moves the balance on Rasmussen from "objective, impartial pollster" towards "political activist who provides polling that supports conservative positions". This is a classroom example of a leading question:

https://www.washingtonpost.com/media/2023/02/28/rasmussen-poll-scott-adams-dilbert/

A poll asked if it’s ‘OK to be white.’ Here’s why the phrase is loaded.

Scholars raise questions about the Rasmussen poll that Scott Adams cited in the racist rant that led to ‘Dilbert’s’ cancellation

By Marisa Iati and Scott Clement

Washington Post

February 28, 2023 at 11:36 a.m. EST

The survey question was asked by the conservative-leaning Rasmussen Reports, whose head pollster described it as a “simple” and “uncontroversial” query. But, in fact, the phrase in question has a freighted history that implies more than its face-value meaning....

Survey takers familiar with that background may have wanted to avoid expressing approval of wording co-opted in that way, experts said....

In recent years, Rasmussen has shifted from serving primarily as a right-leaning polling firm to more actively amplifying conservative causes, with a website featuring commentary from conservative and libertarian pundits. [emphasis added] In the video about the recent survey question, Mitchell also hyped polling results that he said showed “nearly half the country is concerned that vaccines are causing a significant number of unexplained deaths.” (The U.S. Centers for Disease Control and Prevention has said there is no evidence that coronavirus vaccines are causing deaths.) On Twitter, the firm also elevated misinformation about alleged fraud in the 2020 presidential election and highlighted conspiracy theories suggesting that the Jan. 6 insurrection was a “set-up.”...

“You can see that phrase and easily recognize that someone’s trying to get a rise out of you by using it,” Pitcavage said. “Disapproving of that statement and disapproving of whiteness or White people are two very different things.”

--Nbauman (talk) 17:42, 6 March 2023 (UTC)

More crazy stuff

Doug Weller, Nbauman: you may be interested to learn that they recently stated that the COVID-19 "vaccines killed more people worldwide than Jews killed in the Holocaust", after polling people on whether they agree with the statement "China lied. Fauci lied. People died". Wtf lol. Endwise (talk) 07:26, 13 June 2024 (UTC)

@Endwise Crazy. But the world is turning crazy, so no surprise I guess. Doug Weller talk 07:31, 13 June 2024 (UTC)

Major Conservative Poll Cited by Media Secretly Worked With Trump Team

[2] Doug Weller talk 08:27, 28 September 2024 (UTC)

Blatant misinformation cited in Rasmussen bio

False allegations of republican bias cited without proof, or any mention of the fact that the accusations are unproven.

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

The Final RCP map shows us that Rasmussen was among the most accurate, even underestimating Donald J Trump's win by 0.2 percent.

I think it is imperitive that any allegation of bias from a proven accurate pollster is UNPROVEN. Biased moderators refuse to add this important distinction for no logical reason whatsoever. 173.238.207.7 (talk) 00:30, 8 November 2024 (UTC)

The closeness of a pollster's numbers to the final election result is entirely unrelated to questions of whether or not they're biased.

But frankly that doesn't matter anyway, because for us to look at Rasmussen's numbers and decide they prove a lack of bias and thus remove the accusation of bias even though it comes from a WP:RS would be a stark violation of the original research policy. AntiDionysius (talk) 00:38, 8 November 2024 (UTC)

Formatting and Citation Problems

I wanted to get community input, or at least document my rationale, on a few of these edits before I make them because, as most politically-related articles, it is extremely contentious especially in the weeks following an American presidential election.

The formatting of this article seems to be ignored in several places. For example, in the "Polling Topics," the following statement is included:

After Trump lost the election, Rasmussen suggested that Vice President Mike Pence should overturn the election results.

It's unclear to me how this statement, regardless of its truth, relates to "Polling Topics." I think it makes more sense in the "political commentary" section.

Also in the "Polling topics is the following excerpt:

After the election, James Rainey of the Los Angeles Times wrote that "Some conservative media outlets used the Rasmussen polling to prop up a narrative in the final days of the campaign that Romney had momentum and a good chance of winning the White House.

I again do not see how this information is relevant to the "Polling Topics" of Rasmussen, I think it either makes more sense in the "Business Model and Methodology" section or the "political commentary section.

The Rasmussen polls are often viewed as outliers due to their favorable Donald Trump approval ratings.

This pretty clearly belongs in the "Evaluation of accuracy and performance" section, not the "polling topics" subheading.

I think the "Political Commentary" itself is incredibly pointless. The section, as it stands currently, essentially just says "Yep! They have it!" It's two sentences long. If we choose not to merge some of the commentary Rasmussen does to this category, I think the subheading should be removed altogether.

in 2010 Silver concluded Rasmussen was the least accurate of the major pollsters

The following excerpt is included in the "Favorable" criticism of Rasmussen. Why, is beyond me, it seemingly is included to detract from his previous statement that he was favorable to Rasmussen, which would defeat the point.

And now the more substantive issues.

The article claims:

Rasmussen Reports has frequently claimed that COVID-19 vaccines are lethal

And cites this HuffPost article as proof. First of all, the "Frequently" modifier I believe to be misapplied; the source lists at best 2 examples, one of which was republishing a Covid-19 vaccine-skeptic comic. The other was a poll they conducted asking people if they had known someone who was injured after taking the Covid-19 vaccine, and showing the difference between the poll respondents and the official reports. I don't think either of these actions could be considered "Claiming the COVID-19 vaccines are lethal." Thus I think this claim should be removed or supported with a better source.

Also the following claim:

The Rasmussen polls are often viewed as outliers due to their favorable Donald Trump approval ratings.

Has five citations. However all five citations point to a different particular poll which happened to be an outlier in its given data set, none of them purport that "Rasmussen polls are viewed as outliers" or the subtle implication that Rasmussen tends to produce more outlier polls than any other polling company. Anybody who understands how statistics works will see this is quite absurd: all polling companies who conduct thousands of surveys will, through no fault of their own, have several polls which are not just outliers, they are objectively wrong. This is how polling works, it's how p-curves work, it's how we calculate statistics. Even if a company had a 99% confidence interval, which would be unbelievable and frankly unattainable, still for every 1000 polls conducted 10 will be flatly wrong. Can we cite those 10 polls and present them as evidence that clearly they're "often viewed as outliers?" Yet seemingly some author has concluded, of their own accord and not by relying on any source, that if 5 polls exist which were outliers in their respective measurement, then that means that "Rasmussen polls are often viewed as outliers." If we use this logic that "any time some individual polls are outliers in a larger data set, then the polling organization itself is viewed as producing outliers," then every polling institution will fall to this. The author who wrote this is inserting their own conclusions based on an incomplete understanding of statistics, not reliable sources. As such this snippet should be removed.

These are the changes I'll make in a few days unless there's a community census as to why they shouldn't be made. Just documenting my reasons. BabbleOnto (talk) 19:56, 20 November 2024 (UTC)