Skip to content
 

One-way street fallacy again! in reporting of research on brothers and sisters

There’s something satisfying about seeing the same error being made by commentators on the left and the right. In this case, we’re talking about the one-way street fallacy, which is the implicit assumption of unidirectionality in a setting that actually has underlying symmetry.

1. A month or so ago we reported on an op-ed by conservative New York Times columnist Ross Douthat, who was discussing recent research exemplified by the headline, “Study: Having daughters makes parents more likely to be Republican.” Douthat wrote all about different effects of having girls, without realizing that the study was comparing parents of girls to parents of boys. He just as well could have talked about the effects of having sons, and how that is associated with voting for Democrats (according to the study). But he did not do so; he was implicitly considering boy children to be the default.

2. A couple days ago, liberal NYT columnist Charles Blow (link from commenter Steve Sailer) reports on a similar study, this one by Andrew Healy and Neil Malhotra, finding that “that men with sisters are more likely to be Republican. . . . A report from Stanford about the study concluded, ‘Watching their sisters do the chores “teaches” boys that housework is simply women’s work, and that leads to a traditional view of gender roles — a position linked to a predilection for Republican politics.’”

The “report from Stanford” is actually a press release from last year, and it again takes a unidirectional perspective: lots of discussion of what happens to boys with sisters to make them vote Republican, nothing about what happens to boys with brothers to make them more supportive of Democrats. I don’t see any logical reason for brothers to be the baseline, I just think this is how people tend to frame the comparison, without thinking about it. For example, one of the sections of the Healy and Malhotra paper is entitled, “The Effect of Sisters on Political Attitudes,” but there is no corresponding section on “The Effect of Brothers.”

And this framing has an effect, especially in the popular reception of the work, for example in the above-linked op-ed by Charles Blow, who talks all about the effect of sisters while just treating brothers as some sort of gray default, of no interest in itself.

Also (this comes up in the paper and the discussion in a few places), the difference between significant and not significant is not statistically significant.

P.S. When I was a kid, one of my sisters rebelled at one point because she said that the girls got all the gross chores like scrubbing the bathroom. Personally, I hated—hated—mowing the lawn and I swore this would be something I’d never do when I grew up. And indeed I haven’t.

17 Comments

  1. Rahul says:

    You just need a bigger lawn. Those bigger mowers you can actually drive around are kinda fun. :)

  2. Slugger says:

    We all do this. We have our closely held beliefs and winnow through data looking for confirmation. We might encounter a dozen studies that do not support or even contradict our position, and we ignore this to pounce upon a supportive study when we find it. It is a common failing.
    It is certainly common to see medical studies where the total effect of the experimental arm is no different than the control arm touting the “fact” that the treatment is better for some subset. They never emphazise that the treatment is worse than the control for other subsets.

  3. Lord says:

    I was wondering what the reverse case would be, boys without sisters as single boys, boys with only brothers, which would be smaller sets, and would girls with brothers be more likely to be Republican than whom, or would they be more likely to be Democrats and what of girls with sisters. What would girls without brothers be? Is it any sisters, or is there a weighting of brothers to sisters, or is it a larger family effect? Or is the reference all people? It does leave an excess of Democrats to appear somewhere and must be symmetric (with that reference) but where?

  4. So according to the nomenclature established last time (http://andrewgelman.com/2013/12/20/discovered-new-fallacy-now-need-name/) this would be an instance of both the one-way street fallacy and the Smurfette fallacy. :)

  5. jonathan says:

    It isn’t the mowing: it’s the prep and clean-up which amount to a mental investment I just don’t want to make.

    In a related vein, I ran across a hilarious Noam Chomsky quote from Manufacturing Consent – it was on Raw Story so I’ll quote it as a whole from their site:

    “This is an oversimplification,” Chomsky argues, “but for the eighty percent or whatever they are, the main thing is to divert them. To get them to watch National Football League. And to worry about ‘Mother With Child With Six Heads,’ or whatever you pick up on the supermarket stands and so on. Or look at astrology. Or get involved in fundamentalist stuff or something or other.”

    The point is to “get them away from things that matter. And for that it’s important to reduce their capacity to think.”

    “Take, say, sports — that’s another crucial example of the indoctrination system, in my view. For one thing because it — you know, it offers people something to pay attention to that’s of no importance. That keeps them from worrying about — keeps them from worrying about things that matter to their lives that they might have some idea of doing something about.”

    “I remember in high school,” Chomsky continues, “and already I was pretty old. I suddenly asked myself at one point, why do I care if my high school team wins the football game? I mean, I don’t know anybody on the team, you know? I mean, they have nothing to do with me, I mean, why I am cheering for my team?”

    “It doesn’t make sense. But the point is, it does make sense: it’s a way of building up irrational attitudes of submission to authority, and group cohesion behind leadership elements — in fact, it’s training in irrational jingoism. That’s also a feature of competitive sports. I think if you look closely at these things, I think, typically, they do have functions, and that’s why energy is devoted to supporting them and creating a basis for them and advertisers are willing to pay for them and so on.”

    So you can take x and frame it to fit in your ideology and that makes it true? This fascinates me for several reasons:

    1. The way we tell ourselves stories
    2. How these stories can go off the rails into error and even lunacy.
    3. The inability to recognize something that’s been discussed for over 100 years: the concept of frame of reference and thus the relative truth of your particular frame. It’s like no one has heard of Special Relativity, Lorentz transforms, etc., let along the entire post-WWII critical analysis methodology … at least that doesn’t exist when you are the one thinking about truth.
    4. The way this reflects underlying reality. What is truth? This gets into related areas like what can we perceive, implanted memories, memories and perceptions in contexts, etc. We tell ourselves stories all day long. We have no choice.

    So x or y see the same thing as running in opposite directions. And if you replace the word Chomsky and a few other words in that excerpt from Raw Story, you have a guy talking about the hidden cabal running the world, like Jewish domination as described in the Protocols of the Elders of Zion. And the guy can’t see he went off the rails. That’s fascinating.

    • Zach says:

      “that’s why energy is devoted to supporting them and creating a basis for them and advertisers are willing to pay for them and so on.”

      So Chomsky thinks that when the executives at Coca Cola decide whether to sponsor a sporting event they reason: ‘We spend $X and the publicity will result in sales of N bottles and the sporting event will reduce the probability of a socialist uprising by P%.’

    • Corey says:

      Noam Chomsky commits the one-way street fallacy of thinking that irrational jingoism isn’t the default. It’s my view that most people don’t need to be trained to develop irrational jingoism, but rather the opposite — most people will never develop the skill of avoiding that pattern of thought on their own.

      • Entsophy says:

        Yes Corey, but in the case of America, that default is perfectly rational since America rules! USA! USA! USA! USA! USA! USA! USA! USA! USA! USA! USA! USA! USA! USA!

        No need to thank me for raising the tone of the discourse. The thanks of a grateful nation is enough.

      • Steve Sailer says:

        One interesting phenomena is how old urges (e.g, jingoism) and newer fads (e.g., Communism, feminism, gay lib, transgender rights, etc.) can feed off each other.

        For example, the Soviets and East Germans invested heavily in their women’s Olympic teams, doping many female athletes with artificial male hormones, to boost the prestige of their countries and ideologies. So, Americans didn’t get all that invested in international women’s sports because we kept losing to the East Germans.

        After the fall of the Berlin Wall, however, the American public would become temporarily obsessed over America’s women’s sports teams, seeing US victories in basketball, soccer, ice hockey, softball, and so forth as proof of the superiority of American feminism over the oppression of women Paris, Milan, and London, where they weren’t allowed to trade in their Jimmy Choos for soccer spikes.

        Of course, these bubbles evaporated and various pro women’s leagues founded in the enthusiastic aftermath have had a hard road to travel since Americans aren’t really all that interested in female team sports unless either their daughter is playing or the national team is crushing foreign countries.

        The American press and public has been gearing up for the upcoming winter olympics in Russia as World War G, in which our gay male figure skaters stick it in the eye of Putin just like Jesse Owens humiliated Hitler in 1936 (actually, that didn’t really happen, but it makes a good story).

        After World War G … My guess is that the inevitable next step is World War T which will be fought out over the right of men to declare themselves women and maul women in sports.

        • Rahul says:

          I’m not so sure that the American public is indeed so worked about the winter Olympics as you make it sound. I wouldn’t be surprised if a survey showed that the word “Sochi” means nothing at all to 9 out of 10 Americans.

          To further color it as a Gays versus Putin event is ridiculously fantastic. Most Americans probably just don’t care.

          • Steve Sailer says:

            The New York Times very much cares about the denouncing the Russian government’s law restricting the distribution propaganda among minors judging by the obsessive coverage of gay rights in Russia, Slate is writing Sochi-related headlines about “Gay Holocaust,” and Glenn Beck announced “I will stand with GLAAD … against hetero-fascism” in Russia.

            Just because you haven’t noticed the media’s latest obsession doesn’t mean you won’t eventually start being influenced by it. In fact, media influence works better when you don’t notice the patterns, because you might otherwise have a chuckle at their promotion of World War G.

  6. Slugger says:

    A bit more (muddled) thinking from me. My personal experience suggests that affiliation with the right or the left side of the spectrum occurs by the time one is a young adult. I don’t think that there are many changes of heart after age 22. This means that leftists produce more Y chromosone bearing gametes while rightists are shooting a preponderance of X’s if these studies are more than just chaff. Unless you are a confirmed Lamarckian, you should be quite dubious about these results.

  7. Jonathan says:

    Somewhat related? Maybe you are rubbing off on journalists who are discussing statistical drawbacks on their own:

    http://www.bloomberg.com/news/2014-02-03/obamacare-wonks-flunk-data-analysis.html

  8. tko says:

    I was really hoping you’d take the opportunity to cement “Smurfette fallacy” into the nomenclature. The name alone is likely enough to pique the interest of at least a few journalists.

    http://andrewgelman.com/2013/12/20/discovered-new-fallacy-now-need-name/#comment-152545

Leave a Reply