Ninja Infiltrates Skepchick

Hi Everyone! My name is Jamie and way back in 2008 I sent an email to Rebecca saying I was interested in becoming a Skepchick. Needless to say, I didn’t make the cut. I spent the next five years slowly infiltrating my way into the inner circle of Skepchicks, by way of alcohol and unrestrained enthusiasm, until they finally gave in and asked me to join the team. I cannot tell you how excited I am to start blogging here on Skepchick.

First of all, a little bit about me. My background is in statistics and economics, especially as it relates to public policy. You might have heard my name mentioned here before. That’s because I’m the Vice President and Skeptical Ninja (yes, official job title!) of Women Thinking, inc. My official job duties are helping to run the Hug Me! I’m Vaccinated campaign, saving the world, and fetching drinks for our president, Elyse.

But, enough about me, lets talk about social science! This is my area of expertise, which I like to think makes me half-scientist. Often social science is called “soft science” as compared with the “hard science” of the natural science fields. But, these terms can be quite misleading. In fact, the social sciences, which deal with human behavior and societies, are often attempting to answer far more complex problems than the natural sciences. It also means that the answers researchers come up with to big social science questions often have far less evidence to back them up and are more controversial than is typical in natural science fields. Conclusions tend to be much fuzzier and not based on experimental data. This is why, as a skeptic, you should be extra careful when interpreting the results of a new and exciting social science study.

As an example, let’s take the question “What caused violent crime rates in the US to rise through the 20th century, peak in the early nineties, then steadily decline?”

This question is far too complex to be able to design an experiment that answers it. Instead, social scientists must use a variety of methods to build evidence for various theories. Unlike most science, however, they will never come up with a solid answer. Ever. What they will do is come up with a variety of likely answers weighted based on their likelihood of contributing to the answer.

Kevin Drum, a blogger at MotherJones, recently attempted to answer this question by claiming that the rise and later decrease in crime was caused, at least in part, by changes in lead in the atmosphere caused by gasoline. He provides a whole slew of convincing evidence:

We now have studies at the international level, the national level, the state level, the city level, and even the individual level. Groups of children have been followed from the womb to adulthood, and higher childhood blood lead levels are consistently associated with higher adult arrest rates for violent crimes. All of these studies tell the same story: Gasoline lead is responsible for a good share of the rise and fall of violent crime over the past half century.

This is, perhaps, a best-case scenario for a social science theory. There is overwhelming evidence of the connection between lead and crime. So, mystery solved, right? That’s at least what the media seems to be reporting. Even the anti-vaccine bloggers at AutismOne have picked up on this as proof of the evils of toxins in the atmosphere.

Not so. The truth is that, like all social science problems, we will never really know the answer. We will never get to a Theory of Evolution-esque amount of surety. Even in a best-case scenario like this one, the most we can say is that lead almost certainly had something to do with the increase and subsequent decrease in the crime in the last 50 years, though how much is unclear.

Does this mean that social science is useless? I certainly don’t think so. Just because we can never get to the amount of evidence necessary to prove a theory doesn’t mean we can’t learn enough to act on it. Policies to lower lead in the atmosphere are much cheaper to implement with potentially greater gains than other crime-fighting measures. We might never be able to completely quantify it the way we might be able to quantify the mass of an electron or calculate the orbital period of an exoplanet, but using social science we can make fuzzy estimates good enough for use in making decisions that make our world a better place to live.

Ninja image by the talented Jill Powell.

Jamie Bernstein

Jamie Bernstein is a data, stats, policy and economics nerd who sometimes pretends she is a photographer. She is @uajamie on Twitter and Instagram. If you like my work here at Skepchick & Mad Art Lab, consider sending me a little sumthin' in my TipJar: @uajamie

Related Articles


  1. Welcome, and happy blogging ; )

    I know it drives Statisticians crazy when non-Statisticians attempt complex statistics because they did a six month undergraduate course at some point in the past. Does it annoy you that everyone engages in social science all the time? for example, after this latest shooting there theories everywhere on the causes and the best way to prevent it from happening again.

    1. Yah, it really does! One of the problems with social science is the complexity means that often you need an advanced math and statistics background to understand how a study was done and how they came up with the results. Unfortunately, this can make interpreting a study near impossible for a layperson. Tack onto that the fact that practitioners will take advantage of that knowledge gap to push terribly done studies and the end result is utter misunderstanding from the media and general public.

      Instead, it seems most people just gage whether the idea makes intuitive sense. If it fits with already held beliefs, people tend to believe it. If it goes against your beliefs, a person might reject it outright. Usually none of this has anything to do with the actual facts of the study.

      Social science is just super fuzzy so it’s easier to mold it to what you believe. It’s just…really complicated.

  2. Oh no! Skepchicks is being attacked by a ninja! :)

    Welcome, UAJamie. I’m looking forward to reading more stuff by you here.

  3. Congratulations, Jamie!

    As to the question of lead and violent crime, do we even know the suppositions are correct? Lead in the atmosphere, in children and adults is measurable, and it seems reasonable that levels increased since the industrial revolution, and drastically when most gasoline contained lead, but has been declining since the 1970s when unleaded gasoline was first introduced and then mandated and lead house paint was banned and due to other environmental regulations, but how accurate are the violent crime statistics? I was under the vague impression (mostly due to skimming reviews of Steven Pinker’s latest book), that violence (except for state-sponsored wars?) has been declining for centuries.

    OTOH, getting rid of lead, like limiting CO2 emissions, has many other benefits, so even if this hypothesis is a “definite maybe”, it is one more thing on that side of the ledger.

    Maybe I’ll follow up later if I have time to look into this further.

    1. Drum’s article is more of a listing of all the studies connecting lead to crime. Likely each study uses different ways of meassuring “crime” but my guess would be that most are meassuring it by arrests. Obviously, this is rather problematic in a lot of ways.

      Any one of these studies by itself would be interesting but almost nothing more. These types of studies are just way to fuzzy to be able to make clear conclusions. But, lots and lots of studies all looking at different types of correlations between crime and lead have been done and every single one of them shows or suggests a positive correlations. Together, it is quite convincing. It’s rare to ever collect this much evidence in these types of endeavers.

      As you mention, this is just one more good reason to try to further decrease lead exposure. In terms of policies to decrease crime rates though, this is one of the cheapest and easiest to implement even if only for that reason.

      It could be, though, that we’ve already reached the lower threshold of lead exposure and that further decreases will not have much of an effect on crime rates. Even so, it’s relatively inexpensive to implement and would have many other positive benefits for human health and the environment.

      1. So, perhaps lead just makes criminals careless and more likely to get themselves arrested? Or it might make cops more efficient!


  4. I’m curious as to what you think about these explanations for growth and decline of violent crime rates:
    1) source: not sure, something I heard on Free Speech TV–
    Violent crime rates started steadily declining approximately 17-20 years after Roe V Wade.
    2)source: The Late George Gerbner, a long time media analyst–Violent crime is the result of “mean world syndrome,” a phenomenon in which TV news induces a state of mistrust and fear in the general public by over-exaggerating the risks of victimhood, a view also espoused by Michael Moore in Bowling for Columbine.
    3)source: Harvard professor Martin Novak–This director of the Program for Evolutionary Dynamics uses math and game theory to look for evolutionary advantages to cooperation rather than competition. It’s not directly linked to violent crime, but certainly related.
    4)source:The Thom Hartmann Program–violent crime is a result of socioeconomic inequity.
    5)source: author Barbara Kingsolver–In her essay “Life is Precious or It’s Not,” written in response to the Columbine massacre, Kingsolver says the question is not “How could something like this happen?” but “Why doesn’t it happen more often?” She bases her stance on the amount of gratuitous violence in our culture, mainly for entertainment purposes, as well as the horrible message our kids get from global leaders who can’t refrain from bombing one anther’s countries.

    They all seem like plausible explanations, although difficult to prove, as you said.

    1. Hi! Sorry I didn’t reply earlier. I went on a vacation around the time you posted this and somehow missed it. Anyways, I’m glad I saw it now because I secretly really wanted someone to ask me lots of details about things I know a lot about so I can have an excuse to show off!

      1) Yes, abortion being linked with lower crime rates is a well-known theory from Steven Levitt (and popularized in his book Freakonomics). He was able to show that a correlation between drops in crime rates and states that legalized abortion 20 or so odd years earlier. However, as far as I’m aware this is an interesting but far from proven theory that isn’t particularly well considered within the crime literature. It doesn’t mean it is wrong. It just means there is not much evidence for it and some of the evidence is questionable. It is also possible that abortion does have an effect on future crime, but that this effect is quite small and difficult to detect.

      2) I hadn’t heard of this theory specifically so I can’t comment on the evidence for it. However, I do want to make clear the fact that the connection Drum makes in his article between lead and crime isn’t that lead causes all or even most crime. There is a baseline amount of crime that would exist even if we got every last piece of lead out of our homes and air. Drum’s claim is that lead caused extra crime on top of the baseline crime, that lead to the super high crime rates of the early 90s. There were likely lots of other factors involved as well that all came together to form a “perfect storm” of crime. Drum’s claim is merely that lead was likely a large player in causing the crime increase.

      3) Oh, you said my favorite words! Game Theory!! There is quite a lot of theoretical research into cooperation and altruism, the flip side to crime. I took an entire semester just on Cooperation in International Relations. If you’re interested in the subject, I highly suggest reading up about Robert Axelrod’s tournaments.

      4) Within the crime literature, socioeconomic inequity is widely considered to be a cause of violent crime. The weird thing in the last 15 or so years is that crime continues to drop even though socioeconomic inequity is getting worse. This doesn’t mean that poverty and economic inequity does not cause crime, it just means that other forces that are decreasing crime are doing so at a rate that is faster than the added crime from increased inequity.

      5) Total departure, but I totally ran into Barbara Kingsolver once at a symphonic concert back when I lived in Tucson, AZ. As to her theory, maybe the idea that violence in entertainment causes violence out in the world is flawed.

      The main takeaway is that crime is super complicated. There are lots and lots of things that cause crime. In the case of lead, all we can really say is that it played a part (probably along with other mechanisms, perhaps the abortion theory you mentioned or increases in the police force) in the rise in crime rates throughout the second half of the 20th century. But, it certainly isn’t responsible for all or most crime, just some of the change in crime.

  5. Your post looking at the scientific and statistical modeling of gender/sex bias was the first article I saw that drew me to this site. I enjoy your writing (we’ll leave aside my envy for your job title), so thanks, I look forward to reading more!

    I found this article in particular very interesting because I’ve been looking a lot lately (I’m rather interested in the intersection of public health/security and personal liberties) at violent crime trends, and also found the recent Mother Jones article about lead exposure to be particularly fascinating (and having looked at some of the original studies cited, I’m rather convinced by their argument linking that as a factor to be considered). If only solutions to other contributing factors to our high murder rate (given that our overall violent crime rate is rather low) were as relatively simple and straight-forward as lead remediation . . .

    Anyway, congrats on your acceptance, keep up the good work!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Back to top button