Remarks On Pew’s Polarized Partisan Poll

Remarks On Pew’s Polarized Partisan Poll

Pew has a large survey out with the title “In a Politically Polarized Era, Sharp Divides in Both Partisan Coalitions: Partisanship remains biggest factor in public’s political values.” I have two statistical points to make about it. The survey itself is of interest and can be read at the link.

Point one

Partisanship continues to be the dividing line in the American public’s political attitudes, far surpassing differences by age, race and ethnicity, gender, educational attainment, religious affiliation or other factors. Yet there are substantial divisions within both parties on fundamental political values, views of current issues and the severity of the problems facing the nation.

That we’re at each other’s throats, and that a good chunk of the country is abandoning Reality and seeking to mandate all follow them, is well known. Measures of this are bound to be faulty and over-certain.

But one thing that seems fixable, or at least adjustable, is the way of sorting responses. The split Pew uses is self-identification of Democrat and Republican. This has some utility, more so with voting intentions, but about matters like “Gender & Sexuality” it’s far less useful. The foreign policy decisions of the leaders of these parties differs scarcely at all in practice. Both are for spending more, more, more, and then even more. And so on.

Among individuals there’s more spread, but still not sharp. Older, union Ds, for example, aren’t on board with transgender madness, for instance, where younger Rs are happy to pretend two men can marry each other.

Pew knows all this of course.

Democrats also are divided by race, with black Democrats much more likely than white Democrats to associate belief in God with morality and less likely to say that same-sex marriage has been good for society. Racial differences in attitudes are far less consequential for Republicans, who are predominantly non-Hispanic white.

Blacks are far less woke then baizuo on many questions.

Still, Pew’s focus is on D vs. R.

Across all 30 political values, the differences between Republicans and Democrats dwarf all other differences by demographics or other factors. The 39-point average gap is more than twice the difference between white and nonwhite adults (17 percentage points); people who regularly attend religious services and those who do not (14 points); college graduates and those who have not completed college (10 points); younger and older adults (also 10 points); and men and women (6 points).

What I would like to see, or what I would do if I had access to the data, is to cluster it. That is, group people who answered all questions the same or nearly the same.

Cluster analysis is a far from perfect tool, with many subjective decisions and a lot of uncertainties. But I think it might, and only just might, do a better job at separating people than D and R. In any case, that could be examined. If in the clusters the clear separation between groups coincides with D and R, then we’d have something. Not likely, though, given what commonsense tells us.

The cluster analysis can with the sampling plan give as a guess of how many adults belong to each cluster. That would be a lot of fun to know. How many crazed ultra ring wing reactionaries like your host exist? Just me? A million? Ten?

Point two

This one is against scientism, or at least the dangers of assuming. The ever-present voting fallacy is in all these surveys, but I don’t here want to go into that. The voting fallacy says that when enough people agree to a thing that that thing turns magically form immoral to moral, from wrong to right, from evil to good, and vice versa on all these. It’s natural to find this fallacy in democracies (my upcoming book goes into this).

Here’s what I do mean:

Americans generally believe that women continue to face obstacles that make it more difficult for them to get ahead than men. While there are sizable gender differences in these opinions, the partisan divide is even more pronounced.

Overall, 57% of adults say that “significant obstacles still make it harder for women to get ahead than men.” About four-in-ten (42%) say that the obstacles that once made it harder for women to get ahead are largely gone.

The public’s views on whether it is more difficult for women to advance have changed modestly since 2016, when 53% of the public said women still face significant obstacles.

Men and women hold very different views on whether obstacles continue to stand in the way of women’s advancement: Nearly seven-in-ten women (69%) say there are still significant obstacles that make it harder for women to get ahead than men, compared with 43% of men.

Now my idea of women getting ahead and exercising the true and highest power they have is to have kids and raise their families in the way oriented to Truth. A woman sacrificing herself to sit in a cubicle and create PowerPoint “decks” that nobody cares about had thrown away her best chance of influencing the future.

Many of Pew’s questions of these sorts of matters are loaded in favor of progressive ideas. Yes, progressive. That’s the point. The argument between Ds and Rs is loud but to a large extent trivial. The real differences are not heard from.

Except in places like this, of course.

To support this site and its wholly independent host using credit card or PayPal (in any amount) click here


  1. Sheri

    Women do face significant difficulties in getting ahead—it’s called being whiney, stupid, helpless fools because being strong is too difficult. And being strong does NOT mean you have to watch other whiney, stupid women tell you you’re smart (you’re not) or worthy (you’re not), etc. Women refuse to do anything to help themselves except try to castrate all men, change reality and keep on whining. Too bad Pew did cover that in the poll.

    The cluster analysis sounds interesting. Too bad the data is not available. I bet there would be some interesting clusters of beliefs out there.

    There are less than a million and more than 10 right wing crazies, which today is good since crazy actually means sane when used by the crazies (you know who you are and so do we). Love the doublespeak! Use it while you can!

  2. Sheri

    Women don’t get “equal pay” because they are so stupid and greedy and lazy they want to count sitting on their back sides in an office the same as mining coal at 2 am. Women who work in the trades GET THE SAME PAY as men.* But whiney, weak, stupid women don’t want to get up at 2 am in -40F weather and work. So tough luck—get off your behinds AND WORK. Then, and ONLY then, do you get equal pay. That SJW degree should mean minimum wage for a lifetime, by the way. It’s the laziest, stupiest, most meaningless thing a woman (or man) can get and they should suffer poverty all their meaningless lives for getting it.

    *If you work for a Democrat, you WILL get less pay if you’re a woman. They all preach the idea and then short every stupid woman that works for them. They understand human stupidity and use it to their advantage.

  3. c matt

    A woman sacrificing herself to sit in a cubicle and create PowerPoint “decks” that nobody cares about had thrown away her best chance of influencing the future.

    Given some of the women I know who work those cubicle jobs, I am not so sure that losing their best chance to influence the future is such a bad thing.

  4. JTLiuzza

    “Now my idea of women getting ahead and exercising the true and highest power they have is to have kids and raise their families in the way oriented to Truth. A woman sacrificing herself to sit in a cubicle and create PowerPoint “decks” that nobody cares about had thrown away her best chance of influencing the future. ”

    Couldn’t agree more about the first sentence. As far as the second; an empire in astronomical debt with no end in sight, whose females murder their own children because they don’t want to be bothered, that pollutes the children who do actually get to draw breath with the most vile perversions, and who opens its doors to anyone with a pulse regardless of intent, background… clearly doesn’t give a hoot about the future.

    I applaud “In God we trust” even though as a nation we clearly don’t and really never did. While there’s still time perhaps we should change it and stamp on all our coinage and so on, “Eat, Drink, and be Merry; for Tomorrow We Die.”

  5. Uncle Mike

    Dear Dr. Briggs,

    There are some other problems with the survey.

    1. Self-selected respondents. The on-line survey is of those who chose to participate via the Web and not of the population in general. Pew explains how they switched from telephone polling to on-line, but they do not discuss the problems with telephone surveys — which are also made up of people who happen to be home and who also self-select when choosing to respond.

    2. Tiny sample. Almost 10,000 people responded out of at least 200,000,000 American adults who might hold opinions. This is 1 out of 20,000 or a 0.005% sample.

    3. Opinions are temporary. People change their opinions as often as they change their underwear. Opinions are ephemeral, fugacious, fleeting, and transitory.

    4. Pew is a liberal alarmist media animal that constructs fake news for “entertainment” purposes. That’s their gig. They are modern day P.T. Barnums. They rely on suckers born every minute for their market.

    These sources of ERROR are not discussed; indeed they are ignored. The data are crap and not something to hold onto. As a statistician (to the stars!) making “statistical points”, you ought to know and do better. You don’t want to be a #4, do you?

    PS — Cluster analysis my a$$.

  6. Milton Hathaway

    Sometimes I feel like I’m on another planet. (And the older I get, the more often I visit.) People have their principles and beliefs, and they prioritize them, at least mentally. They compare the top items on their list to the platforms and actions of the political parties, and declare themselves (temporarily, at least) Repoltroonican, Corrupticrat , or Independent.

    If x obviously causes y, then measuring a high correlation between y and x is just an exercise in redundancy.

    Or did I miss the point?

Leave a Reply

Your email address will not be published. Required fields are marked *