When stereotyping creates the bias being researched

I am reading The Bias That Divides Us by Keith E. Stanovich.  Good summary of research but hard slogging owing to constant reminders of the author's own biases.  Or, better put, blinders.  

One of the most interesting things about the 2019 Van Boven and colleagues experiment was that it tested exactly the same subjects who had responded to the travel ban issue on a second controversial issue of contemporary importance, an assault weapons ban, meant to reduce mass shootings. The subjects were given a definition of “assault weapon” (any of a number of semiautomatic rifles, and other semiautomatic weapons equipped with attachments—such as a scope, pistol grip, or grenade launcher—or high-capacity magazines), they were told that a comprehensive bill banning a number of assault weapons had been introduced in Congress, and they were given the following statistics (presented as frequencies in order to disguise the parallel with the experiment’s other issue), based on current and historical data (S = mass shooting; A = assault weapon):

p(S): In the last few years, 6 out of 100 million American “adults committed a mass shooting.

p(A): In last few years, 12 million out of 100 million American adults owned an assault weapon.

p(A|S): Out of 6 American adults who committed a mass shooting, 4 owned an assault weapon.

P(S|A): Out of 12 million American adults who owned an assault weapon, 4 committed a mass shooting.

They were then asked which one of these statistics was most important to them personally in deciding whether to support or oppose the assault weapons ban.

Most subjects chose one of the two conditional probabilities as the most important. Clearly, the hit rate, p(A|S), of 4 out of 6 (67 percent) seemed to support the assault weapons ban more than the inverse conditional probability, p(S|A) of 4 out of 12 million (0.000003 percent). Subjects who had supported the assault weapons ban overwhelmingly chose p(A|S) as the most important, but subjects who had opposed the assault weapons ban, in contrast, overwhelmingly chose p(S|A) as the most important. As on the previous issue, the huge myside bias observed was roughly equivalent on both sides of this issue, and it was not attenuated in “those subjects high in numeracy (who in fact displayed a greater myside bias).

You have no doubt already intuited what is amazing about the results of the 2019 Van Boven and colleagues experiment. The subjects who opposed the travel ban tended to support the assault weapons ban (for simplicity, let’s call them “the liberals”)—and the subjects who supported the travel ban tended to oppose the assault weapons ban (for simplicity, let’s call them “the conservatives”). That means that both liberals and conservatives were switching their preference for types of evidence depending upon the issue in question. The liberals did not like focusing on the hit rate when the issue was a travel ban, but they did when the issue was an assault weapons ban. Conversely, the conservatives liked focusing on the hit rate when the issue was a travel ban, but they did not when the issue was an assault weapons ban. The Van Boven and colleagues 2019 experiment provides a particularly good demonstration of how people pick and choose which statistic they view to be most important based on which is most consistent with their prior opinion on the issue at hand.

And where I am working hard to follow the tangle of thinking is the travel ban.  Van Boven and colleagues have demonstrated the reality of the My Side Bias and reasoning switch.  That's the key point.  Which is why it becomes puzzling why Stanovich is imposing putative labels such as liberal and conservative.  Since the researchers are not asking about either liberal or conservative self-identification or partisan registration, what does this imposition of conjured labels add?

The mainstream media constantly portrays conservatives and Republicans as xenophobes and anti-immigration when that is, at best, a very weak case as discussed below.  And Stanovich seems to be working off this media stereotype.  

He seems to be conflating Republicans with conservatives and conflating support of a travel ban with opposition to immigration.  

It seems like he is assuming into existence "let’s call them “the conservatives”" a stereotype.  Otherwise, he would have said something more straightforward like:

Those who self-identified as conservative both supported the travel ban and opposed gun control, demonstrating that they used varying logic depending on the issue.  

If I am correct in assuming Stanovich is labelling based on media stereotypes, then it seems critical to ask whether the media stereotype which Stanovich is using has support in history and data.  

In 1965 the Immigration and Nationality Act was passed.  This new act replaced the earlier immigration policies which strongly favored immigration from Northwestern Europe and strongly constrained immigration from Asia and Africa.  It was passed with bi-partisan support.

In the House, 202 Democrats voted yes, 60 voted no, and 12 abstained, 118 Republicans voted yes, 10 voted no, and 11 abstained.[30] In total, 74% of Democrats and 85% of Republicans voted for passage of this bill. Most of the no votes were from the American South, which was then still strongly Democratic.

While it was bipartisan in support, 85% of Republicans supported the reform but only 74% of Democrats.  

The largest amnesty extended to illegal aliens at that time was signed into law with the Immigration and Reform and Control Act of 1986.  Signed into law by Ronald Reagan and again supported on a bi-partisan basis in the House and Senate.  

Using Republicans as a rough proxy for conservative, then we see a consistent modern history in which conservatives are supportive of immigration.  In other words, support for travel bans is not inherent in the definition of conservative.  Which is what calls into question Stanovich's otherwise seeming arbitrary choice to label support for the travel ban as a "conservative" position.  Republican position?  Probably reasonable.  Conservative position?  Not necessarily.

I am unable to find any sort of large-scale longitudinal surveying which establishes conservative versus liberal support/opposition for immigration.  There is this Gallup data showing support for immigration has been high and rising among all Americans.  In the 2000s there no partisan difference to speak of in support of immigration but by 2018, there was a gap.  In 2018, 65% of Republicans supported immigration, 75% of all Americans supported and 85% of Democrats supported.  But supported by strong majorities across the party spectrum.

Politic had polling on the travel ban in particular.  

Polling on the travel restrictions has varied wildly since the Trump administration unveiled the first executive order on travel in late January. But after months of litigation and controversy, 6-in-10 voters back the ban — and the survey suggests the actual policy may be more popular when separated from the president.

Asked whether they support or oppose the State Department’s “new guidelines which say visa applicants from six predominately Muslim countries must prove a close family relationship with a U.S. resident in order to enter the country,” 60 percent of respondents said they support the guidelines, and only 28 percent oppose them.

[snip]

Republicans overwhelmingly back the restrictions, the poll shows. Eighty-four percent of GOP respondents support the ban, while 9 percent oppose it. But the policy is also popular among independent respondents: 56 percent support it, compared with 30 percent who oppose it. Democrats tilt slightly against the ban, with 41 percent supporting it, and 46 percent in opposition.

If we are accepting the conflation of Republican/conservative and Democrat/liberal, then we need to take into account that the travel ban was supported by the majority of the population and that Democrats (when taking into account margin of error) were basically in equal support and opposition.  

So Stanovich's choice to identify supporters of the travel ban as conservative seems increasingly unrigorous and arbitrary.  

Which makes sense when one thinks all those classes of philosophical orientations who generally fall under the rubric "conservative" and where those groups tend to fall on immigration.  Libertarians are generally very supportive, as are big corporations, main street conservatives, patriotic conservatives, religious conservatives, Classical Liberals (who are now categorized as conservative), etc.  Not all categories of conservatives are supportive.  Defense oriented conservatives tend to be somewhat lukewarm, as might be nationalists.  But those tend to be smaller segments of the big tent.  

So if we are indeed talking about conservatives, we would not be surprised to find many/most conservatives supporting immigration.  

The travel ban was a targeted piece of legislation that is hard to separate from Trump and from the Republican Party and from the circumstances of the time.  Those same conservative groups who support immigration might be more concerned about the travel ban but it is difficult seeing many of them shifting hard away from their philosophical positions, particularly for something that was advanced as a targeted policy around specific perceived threats.  Defense conservatives would be reasonably supportive but everyone else would likely be judging on the specific perceived merits of the stated policies towards achieving a stated improvement in security.

Even if we accept that, Politico finds that the legislation was broadly supported or at least accepted across party lines.  

Which is all a longwinded way of saying that Stanovich seems to be buying into mainstream media stereotypes in order to draw conclusions about whether "conservatives" were using inconsistent reasoning in order to arrive at predetermined decisions.

I am comfortable accepting that people are inconsistent in their reasoning but this seems to be the cornerstone example of My Side Bias and Stanovich seems to be relying on imputed stereotypes rather than actual evidence in order to arrive at the appearance of a My Side Bias.

Perhaps all this is merely awkward wording on the part of Stanovich but it certainly raises a big red flag.  Or my misunderstanding of what he is saying.