Condemnation is fun, but numbers are boring. What are we supposed to do, then, when condemnation requires numbers? Suppose you want to condemn a parent for putting his child in grave danger, but your knowledge of actual risks is hazy at best. The simplest approach is to do what Kahneman calls “answering an easier question.” Leap to morally judge the parent, then pretend your moral judgment implies a risk assessment.
Does anyone really use this silly shortcut? Yes! Thomas et al.’s “No Child Left Alone: Moral Judgments About Parents Affect Estimate of Risk to Children” (Collabra, 2016) runs a series of experiments to test for the presence of what I call “moral causation.” Background:
Research in other domains has shown that moral judgments do affect people’s estimates of harm. For example, intentional actions that result in harm are seen as more harmful than unintentional actions with the same outcomes [17, 18]. Moral intuitions also affect judgments about cause: A driver who gets into an accident while speeding home to hide his cocaine is said to have ’caused’ the accident more than a driver who was speeding home to hide his parents’ anniversary gift [19]. People have also been shown to seek what is called ‘moral coherence’: they modify their factual beliefs to match their moral intuitions. For example, after reading an argument that capital punishment is morally wrong no matter the consequences, people are less likely to believe that capital punishment deters crime [20].
We hypothesize that a similar process may be at work when people imagine the harm likely to befall unsupervised children. That is, people may overestimate the danger to unsupervised children in order to justify their moral condemnation of the parents who allow the children to be alone.
The authors give subjects a series of vignettes, then experimental vary quantitatively irrelevant but morally pertinent details:
The vignettes differed only in the reason for the parent’s absence. In the ‘Unintentional’ version of each vignette (‘Unintentional’ condition), the parent was involuntarily separated from the child by an accident. In the other four versions, the parent intentionally left the child in order to work (‘Work’ condition), volunteer for charity (‘Volunteer’ condition), relax (‘Relax’ condition), or meet an illicit lover (‘Affair’ condition). After reading each vignette, participants were asked to estimate (on a scale of 1 to 10) how much danger the child was in during the parent’s absence.
Respondents inclined to blanket paranoia:
Estimates of risk were high overall. The mean estimate of risk across all situations (on a scale of 1-10) was 6.99 (SD = 2.63), and the modal estimate was 10.
Marginal effects, however, worked in the predicted directions:
This makes no sense at all.
In reality of course, children who are left alone in circumstances approved by their parents are likely to be safer than children who find themselves alone by accident, because parents can take steps to ensure their child’s well-being in their absence (e.g., making sure the baby is securely buckled into a car seat; that the car is parked in a shady spot; that an older child has a cell phone, knows when to expect the parent back, etc.) The fact that participants considered children left alone by accident safer than those left alone on purpose strongly suggests that participants’ moral condemnation of parents skewed their risk estimates.
Thomas et al. consider an array of competing hypotheses, and find them all wanting. Most notably:
In Experiment 4, we asked participants to make explicit moral judgments about the behavior of the mothers in the vignettes. This served as a manipulation check, confirming that subjects did consider leaving a child alone on purpose to be less morally acceptable than leaving a child alone by accident. Leaving to meet one’s lover was also considered less acceptable than leaving to work or relax. A second reason for including the moral question was to allow participants to make separate evaluations of the risk to the child and the morality of the mother’s actions. We thought that by giving participants a way to express their moral disapproval separately from their estimates of risk, they might produce less biased estimates of risk. In fact, the opposite turned out to be true: Risk estimates in Experiment 4 were more affected by moral judgments than in Experiments 1-3. It seems that the explicit moral question simply primed respondents to pay more attention to morality, producing even more exaggerated estimates of risk.
People don’t only think that leaving children alone is dangerous and therefore immoral. They also think it is immoral and therefore dangerous. That is, people overestimate the actual danger to children who are left alone by their parents, in order to better support or justify their moral condemnation of parents who do so.
If the experts stopped making excuses for laymen’s innumeracy, would laymen change their minds? I don’t know, but I’d like to find out. Expert apologists, repent!
READER COMMENTS
ybel
Sep 6 2016 at 1:26am
It would seem to me that there is a basic problem here. The explanation they provide is interesting, but it is hard to be sure and the reason is that there are too many confounding factors. The core of the problem is that the morally deviant behavior is also a signal for irresponsible behavior more generally.
So take the cocaine driver for example. Isn’t it reasonable to assume that–regardless of your views on drugs–a person like that is more likely to have a higher level of future discount than the ‘good son’ and so that his reckless driving may be more reckless than that of the good son?
konshtok
Sep 6 2016 at 4:26am
when a study caters to your prejudice about how stupid the public is you should be suspicious
so
get 3 more replications of this study done on different populations and call us in the morning
john hare
Sep 6 2016 at 4:37am
It would also be useful to know the age group of the children involved as well as the particular society they live in. A 12 year old in an upscale neighborhood is safer than a 4 year old in the bad part of town.
Michael Savage
Sep 6 2016 at 4:55am
For further evidence of this thesis, look at any discussion of bicycle helmets. It gets boring quickly, but the rhetoric on both sides is interesting – absolutely about condemnation irrespective of actual evidence.
phil
Sep 6 2016 at 8:24am
@ybel
yeah, exactly
I don’t think these really demonstrate layman innumeracy
in fact, I actually think they show relatively impressive Bayesian reasoning
—————
“A driver who gets into an accident while speeding home to hide his cocaine is said to have ’caused’ the accident more than a driver who was speeding home to hide his parents’ anniversary gift ”
sort of demonstrates that nicely, that one person was engaged in an (risky) activity far outside the bounds of social norms and the other wasn’t, seems like a Bayesian relevant detail, which, all else being equal, should actually change your assessment of each drivers risk management tolerance/skills
————————-
————————-
while on a normative level, I agree that the hysteria over unattended children is overblown
I’m not sure those survey result demonstrate layman innumeracy all that well
The work v volunteer v relax v affair distinction seem fairly Bayesian (though, I have to admit, that’s not how I would have intuitively ordered it)
the intended v unintended is trickier
I think what might be going on is viewing the risk as systemic v nonsystematic risk
I think people can accept that some nonsystematic risks are baked into the process of having kids (“a parent is going to show up looking for this kid any second now…”)
but react differently to what seems like systemic risk (“this is the 3rd time this week that this kid has been left alone while the mom has her affair, one of these times…”)
though I’ll agree, that as to each particular instance, the risk calculation seems off
[Do I get my gold star as a expert layman apologists?]
——————-
——————-
——————-
As to the larger question of “And most experts are too eager to reinterpret laymen’s ubiquitous irrationality as “rationality in disguise” to accept the straightforward explanation.”
I think we’re evolved to have pretty good Bayesian reasoning skills, but yeah, its not hard to come up with laboratory examples (tricks?) that test and break most peoples reasoning skills, especially since the stakes tend to be abstract and low in laboratory settings.
I think the more interesting question is how often those occur outside the laboratory
obviously they occur a lot, but do people figure them out when real world stakes are involved, when they can taste the carrots and feel the sticks?
I’m pretty agnostic toward the that question, but I think its interesting
How many times do people have to be burned before they can spot the scam?
[A small formatting suggestion from the Econlib Editor: Using long series of horizontal hyphens or underbars to demarcate and subdivide your comment can cause your comment to be captured and delayed by our spam filter. Might you actually instead mean to use more familiar formatting, such as bulleted or numbered lists? Or, different levels of indention–so, say, indenting some materials to show they are subsidiary? We offer a lot of options for formatting your comments. You can use our comment form buttons, or you can enter basic html list formatting directly.–Econlib Ed., or perhaps one Bayesian to another]
Maxim
Sep 6 2016 at 1:10pm
@ybel @phil: I was also thinking that, but what about the “volunteering” question? Surely that should imply more safety.
I guess one could argue that the person would have to be really abnormal to leave kids alone to go volunteer, and that the kids are in more danger because their parent is really abnormal. Seems a bit of a stretch… if that were the case it would perhaps show a strong “anti-abnormal” bias…
Dylan
Sep 6 2016 at 1:20pm
I’d echo Konshtok and add that you need to double that skepticism when the authors of the study also clearly seem to be biased towards the reported result. I read an interview with the authors on NPR a couple of weeks back where it was clear they think the current level of societal risk assessment is off the charts when it comes to children. I happen to agree with them, but given the replicability crisis in the social sciences, and the role that we know the unconscious views of the author can have on biasing results, I’m going to stay skeptical until I see the results repeated.
jc
Sep 6 2016 at 2:46pm
I doubt it.
We’re wired the way we’re wired, and that includes each of our brains being spin-doctor machines.
Interestingly, this may involve – though I’m not saying it’s the case here – bigger picture, system-wide optimization as opposed to seeking answers that seem rational and logical when analysis occurs using a simpler, isolated, linear lens.
In other words, what seems to be irrational at the local level may actually be the most logical choice when everything else in the world is not held constant…and everything in the world rarely is held constant, i.e., it’s one big complex adaptive system.
Consider our herd mentality and pretend (I say pretend, because I don’t know if it’s true, though it should suffice for illustrative purposes) that an artifact of this mentality is the creation and enforcement of irrational and immediately counterproductive religious/political/tribal beliefs. Say a sacred norm says we must all take a work break every 30 minutes to admire and appreciate the sacred color pink…
A linear thinker might say this reduces work efficiency, for a silly reason to boot, as pink is nothing special…and even if a reason later arises to show that it was efficiency-neutral or even enhancing – say breaks every 30 minutes help our brains perform at a higher level – this was just an accidental gain, i.e., we got lucky by doing something irrational as opposed to understanding and then rationally weighing the benefit of rest against the time taken off work.
A systems thinker might concede that this is irrational at the local level. And it’s not consciously rational at any level. But evolution is smarter than you, and it optimizes its evolutionary logic at the systems level. Here, the gains of tribal cohesiveness outweigh any loss of efficiency.
For the record, I’m more of a free-range-children kind of person. And I do think folks are being irrational, and behaving badly, by feeling moral outrage first and then finding their reasons and/or skewed evidence in favor of these reasons. I just also happen to think that this is what we, as humans, are hardwired to do.
Now, its expression may change, i.e., 100 years from now we may feel outrage when children’s lives are micromanaged as we hinder needed development. But the reflexive drive and ability to feel first and then seek and spin evidence in order to justify? Well…that’s just who we are. And if “experts” provide statistics that prove our outrage is misplaced? Well, we’re more apt to burn them at the stake, or simply ignore them, then we are to change our minds.
Game of Thrones: “It is known”. That’s all the justification we need.
Michael
Sep 6 2016 at 7:22pm
This doesn’t seem to make a lot of sense. Bayesian relevant to what? It seems you’re just agreeing that the cocaine guy is a bad person, so deserves more condemnation, which is exactly what’s at dispute.
If you’re claiming one reason or the other is a better predictor for future accidents, yeah, I’d have to see some great studies to believe that one. There has been plenty of “deviant behavior” in the past that’s now accepted and does not appear to, when attached to other bad outcomes, better predict the unrelated bad outcomes.
phil
Sep 8 2016 at 8:12am
@Michael
1.
“It seems you’re just agreeing that the cocaine guy is a bad person, so deserves more condemnation, which is exactly what’s at dispute.”
tbf, I didn’t double reference the original study in question, so its possible that I’m taking it out of context
but,
in the scenario where I have two fact patterns, and my job is to figure out in which of the two fact patterns the person involved is more likely to be at fault
fact pattern A is exactly the same as fact pattern B (they’re going the same speed, weather conditions were the same, the car had been inspected exactly as long ago … etc out to infinity),
except for the 1 difference that in fact pattern A, the person was acting well inside social norms, and in fact pattern B they were acting well outside social norms
that change to the fact pattern gives me information about who was more likely to be a fault
it might not be especially high value information (a change in the fact pattern about 1 driver going faster than the other driver is probably more high value information), it might be information I’m overly likely to take out of context, but it is information
if I’m able to contextualize it correctly, it brings me close to the answer to my question, than before I was aware of it
That’s what I mean by Bayesian relevant
2.
my value judgment as to the social norm, and whether that makes them a bad person, isn’t really relevant
Comments are closed.