Social penumbras predict political attitudes

More crap from PPNAS:

The political influence of a group is typically explained in terms of its size, geographic concentration, or the wealth and power of the group’s members. This article introduces another dimension, the penumbra, defined as the set of individuals in the population who are personally familiar with someone in that group. Distinct from the concept of an individual’s social network, penumbra refers to the circle of close contacts and acquaintances of a given social group. Using original panel data, the article provides a systematic study of various groups’ penumbras, focusing on politically relevant characteristics of the penumbras (e.g., size, geographic concentration, sociodemographics). Furthermore, we show the connection between changes in penumbra membership and public attitudes on policies related to the group.

Jargon alert! Sounds like gobbledygook to me.

And look at these results:

Pretty noisy, huh? And don’t get me started on the problems with their identification strategy. B-b-b-but . . . they have some robustness checks! It is to laugh.

On the plus side, the data are openly available so you can wade through the ugly code yourself and marvel at the unrecognized potential for nonresponse bias. Enjoy.

P.S. To be serious for a moment . . . I actually really like this paper. Yotam and I worked hard on this project because we think that the concept of penumbra that we introduce here is a helpful way to understand a lot of things in politics.

16 thoughts on “Social penumbras predict political attitudes

  1. Edit: I see that an earlier version of this paper was called “The Political Significance of Social Penumbras”, which means that at some point the decision was made to go with the more confident title “Social penumbras predict political attitudes”. But then you must actually feel quite confident about the results, despite the issues raised in this blog. Or maybe the PNAS editor forced you into it somehow? Interested in the story there.

    • Michael:

      What happened was that PNAS told us they wanted a declarative title. In our revised title, we used “predict,” which is indeed confident but avoids causal implications. I’m confident about the predictive part!

      • The abuse of the word “predict” has been bothering me for a while.

        “Intention to engage in climate activism in the United State predicts familiarity with Greta Thunberg.”
        “No shit, Sherlock.”

        “Familiarity with Greta Thunberg predicts intentions to engage in climate activism in the United States”
        “Data proving Greta Thunberg right. You are never too small to make a difference.”

        At least “political attitudes predict social penumbras” is less trivial.

      • I wonder if this could partially* explain the rural-urban divide, as many rural communities tend to be more “homogeneous” whereas people in say suburbs and diverse cities are likely to know people from all sorts of groups?

        *Although I think there is a “practical” component too – large cities need more government services in day-to-day life to function, vs. many rural communities where people are on private well water and use septic tanks & may not have municipal garbage collection, etc.

  2. Where did the chart format come from?

    Reading the caption I got interested in the numbers but it’s really hard to get them from the plot. Is there some sort of radius going on here? Where I’m supposed to be reading distance from origin not area?

    This seems like a good place for a parallel dot plot: https://statmodeling.stat.columbia.edu/2020/08/30/an-example-of-a-parallel-dot-plot-a-great-way-to-display-many-properties-of-a-set-of-items/

    I also thought of https://statmodeling.stat.columbia.edu/2021/01/19/subtleties-of-discretized-density-plots/, but maybe the discretization is too coarse?

  3. I glanced at the figure of estimates and CIs and thought “huh, at least they didn’t present a table with stat sig stars…odd he didn’t mention that.”

    Nicely done sir.

Leave a Reply to Andrew Cancel reply

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