Pew: Partisan Divides Over Political Values Widen, Part I

Pew has a survey out which measures, sort of, our increasing political and cultural apartness (a.k.a. apartheid).

The gap between the political values of Democrats and Republicans is now larger than at any point in Pew Research Center surveys dating back to 1994, a continuation of a steep increase in the ideological divisions between the two parties over more than a decade.

There are people who self-identify as D or R, it’s true, but what Ds and Rs represent has not been constant.

At the top levels, in Congress, the there is thismuch difference between the two groupd. Yes, they have small battles to divide the spoils, and have different emphases, but that there are spoils to divide there is certain agreement.

At the bottom, where you and I dwell, dear reader, there is more agreement than you might think, too. And a lot of it in cultural and social matters.

It’s only in the middle, in places like universities, bureaucracies, professions, Twitter and the like where the gap is widest.

All this makes interpreting their results difficult. We must keep in our mind there is no permanent metaphysical definition of a D or R. If there was, then the results would not change in time, given a decent sample. How could they? If beliefs are constant, the answers about them should be, too.

So a lot of what we’re measuring is the changing definition of what it means to be a D or R. Also, it is really only useful to speak about the differences. On the set of subjects and propositions there is perfect or near perfect agreement, it is useless to create the political categories in the first place.

Scan the picture above (the gray is the average of all respondents). All are valid questions to ask about D and R differences, because there are no agreements.

The thing to note about the questions, which you might not see at first, is there is no unanimity inside Ds and Rs.

Take “Homosexuality should be discouraged by society”. Now, just over a third of Rs agree, but still do about 1 in 10 Ds!

Incidentally, only 1 in 4 now agree with that statement. The number has oscillated historically.

Knock Knock!

“Who’s there?”

“P.U. surveyors.”

“What do you want?”

“We wanted to ask whether you think homosexuality should be discouraged by society?”

“I sure do.”

“Thanks. Huh. You’re the only one so far.”

“Being alone’s my lot in life. When are the results of this survey going to be available?”

“Should be posted at the South Gate of Gomorrah tomorrow morning — that’s if these ominous storm clouds hold back.”

If the trends in this question hold, then in about eight years it will be an absolute requirement to be a D to think homosexuality should be encouraged, tacitly or actively. Rs would join sixteen years after that. And then there will be no use asking this question of Ds and Rs, because there would be complete agreement.

That’s assuming the trends hold. But don’t forget: motus in fine velocior. Things accelerate toward the end. Twenty-four years is thus likely an over-estimate.

This is emphasized because if there are left at that time a significant proportion of people who think the question worthy of disagreement, then D and R may merge and another party — call it A for another — might be created.

To keep Ds and Rs intact would require, we might guess, agreement not just on one question, but several. Now we have to assume Pew (or P.U.) asked the most representative of the important questions.

There are still three questions where the gap between D and R is widening: on poor people having it easy because they can live on government taking people’s money and distributing it how it wishes, on blacks being able to kneel at professional sporting events, and how environmentalism hurts business.

These areas seem sufficient to keep distinctive differences between Ds and Rs. This next picture of Pew’s agrees with that assessment.

This is crude, but illustrative. Pew asked 10 questions, gave numerical answers to them, then showed the distribution of total score for Ds and Rs. Again, Ds and Rs are not concrete; they are fluid. But it appears Ds are more fluid than Rs.

The distribution of R scores is about the same since 1994, with some more spread. But the Ds have moved fairly far leftward. (I like, too, that Pew uses the median and not mean as a “gap” summary.)

But we already knew we all loathe one another, so this graph doesn’t add much.

Islamic violence and much more in Part II!


  1. Gary

    P.U. surveys

    Truth in advertising

  2. Ken

    PEW Topic: ‘The govt today can’t afford to do more to help the needy’
    R – 69% agree (govt’s funds are tapped out)
    D – 76% disagree (govt can/should spend lots more)

    With just the interest being paid on the national debt approaching $50-100B in a single month (some months), and the national debt resoundingly exceeded annual GNP (with the debt growing still), that the govt cannot afford to expand discretionary expenditures ought to a “no-brainer” (i.e., that one ought need little to no brain power to realize the Fed’ bank is tapped out and then some, and, the amount of available other people’s money to spend is dwindling).

    That some three-fourths of a group believe the govt can [and should] spend lots more for new, never before implemented, discretionary do-gooder programs indicates a sizable proportion of some demographics lack understanding of economic and fiscal realities (education cures ignorance), or, may very well lack a brain.

  3. Levantine

    Off topic, I want to bring to your attention these data:
    “More than 75 percent decline over 27 years in total flying insect biomass in protected areas”: (Germany)
    via the comments on that page,
    Similar record of decline in a national park in Finland:
    Some simple notes about insects’ declines in Britain and the USA:

  4. Plantagenet

    Totally off topic;
    A Chemist, a Biologist, and a Statastician go hunting.
    The Chemist misses the deer five feet to the left.
    The Biologist misses five feet to the right.
    The Statistician shouts “We got it”

  5. Plantagenet

    Sorry Briggs nothing personal I just heard that this morning…probably an old one.

  6. Ye Olde Statistician

    It is not at all unusual to see changes to the opinions of Ds and Rs that are actually constant opinions changing their declared affiliations. Being a D or R is not like being an M or F or like being a B or W; that is, not an inherent characteristic. No one has to provide a Party Registration card to call himself one or the other; esp. to say he is leaning one way or the other. If he can figure out which way that is. Parties have been known to lean multiple ways.

    Not addressed is the Heisenberg Uncertainty Principle of Questionnaires. What do some of the these questions mean? Do they mean the same thing to different respondents? Take two of them:
    a) Government regulation of business usually does more harm than good.
    Does “usually” mean more than 50% of the time? 50% of the regulations? Or some other frequency? What does “more harm” mean? How much more? Harm to whom? How does one answer “it depends”?

    c) Poor people have it easy because they can get government benefits without doing anything in return
    What if they can get government benefits without doing anything in return, but nevertheless do not have it easy by any stretch? Or what if they have it easy, but not because of government benefits? (What if you know someone who works in a government benefits office and you know what some of the stories actually are, as opposed to liberal-conservative fables?)

Leave a Reply

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