I admit, I have sometimes wondered what leads people into the career euphemistically we might euphemistically call “pest control” or more realistically declare “extermination”. I have been greatly relieved that people have chosen such career paths and are so available to handle wasp nests in my attic, and unable to fathom why anyone would opt to spend their time getting up close and personal with stinging insects, snakes, and who knows what else, even if only briefly. Despite the best efforts of one overdramatized article from Psychological Science, though, I am not willing to declare all exterminators sadists.
The topic of the article is certainly compelling: the prevalence of horror films, violent video games, and super-slow-motion instant replay of every injury in major sporting events suggest what the researchers call “everyday sadism”, a degree of pleasure in cruelty that goes beyond mere schadenfreude but not quite to the extreme of being a psychological disorder. The authors are probably right that the tendency for some “normal” people to have more sadistic tendencies has not been studied enough; unfortunately, they seem to become more focused on showing the sex appeal of their research than on sticking to careful conclusion about what their studies actually show.
I’ll focus on Experiment 1, which is where the exterminators come in. Volunteers, believing themselves to be in a study investigating how personality relates to “tolerance for challenging jobs”, were asked to choose between four possible career paths: extermination, where they would be asked to kill bugs; extermination assistant, where they wouldn’t have to kill themselves but would have to help; sanitation, where they would have to clean toilets; and a proxy for any outdoor job in the winter, where they would stick their arm in a bath of ice water.
Volunteers who chose the extermination job were given a modified coffee grinder and three live pill bugs, each in a pretty paper cup labeled with whimsical names such as “Muffin”. Their task was to drop the pill bug in the grinder and push down to get it working; sound effects were provided to mimic the crunching of a bug, even though the grinder was rigged so the pill bugs were never in any danger. (I swear, someone did this study just so they could include “No bugs were harmed in this experiment” in their paper). Extermination assistants had to hand the pill bugs in cups to the experimenter, but didn’t have to actually work the coffee grinder, just watch and listen. Those who chose sanitation or cold were off the hook, and ultimately didn’t have to do anything (although a toilet plunger and cleaning supplies were actually in the room).
To evaluate why some people opted for extermination over cleaning toilets, several personality tests were given. One was the Sadistic Impulse Scale, which asks people to indicate how strongly statements like “Hurting people is exciting” reflect their attitudes. There was also a straightforward question of “Are you afraid of bugs” to exclude that as a motivating factor. From these scores and the volunteers’ choices of jobs, the researchers calculate that having above-average sadism makes people twice as likely to head for an extermination job, where the pleasure of killing can be had, over the merely disgusting job of cleaning toilets or the actually panful proxy of an ice water bath.
Sadly, this is also where the researchers start overstating their case. It begins with the sudden and unheralded switch from what “higher sadism scores” were linked with, to referring to a subset of the volunteers as “sadists”. On the one hand, I can sympathize with the desire to speak concisely and the need to stay within journal word limits; on the other hand, “sadists” is definitely a loaded term with implications far more severe than the “everyday sadism” the researchers declared themselves interested in. It is certainly going too far given the only cutoff mentioned for “high sadism” (1 standard deviation above average) meant they were simply looking at the top 16% of self-reported sadism scores. And when we turn to the Supplemental Materials to find out what the average and standard deviation actually were, we learn that a score of just 2.18 on a scale of 1 to 5 is enough to mark you as “sadist”; that’s the equivalent of saying “slightly” to most of the questions. I don’t know about the rest of you, but I am not willing to declare that 1 in 6 people (16%) are “sadists” on the basis of “slightly” agreeing to statements instead of just picking a 1 for every single answer.
That might seem like a lot of quibbling over a label. But the label is just the beginning. There is also a tendency to ignore the fact that over half of the volunteers opted for a job in extermination, with only a third going for sanitation (and only 12% going for the painful ice bath). That’s 53.6% of the volunteers agreeing to exterminate, when only 16% of the volunteers could be declared above-average on sadism. What’s going on in the minds of the third of the volunteers who chose to exterminate despite not having particularly sadistic tendencies? No suggestions are offered. However, those non-sadistic people choosing to kill bugs does explain why the sadism scores for people who chose to kill bugs themselves were not that much higher than those who chose any other profession (about a 1.8 for the primary bug killers, compared to averages between 1.4 and 1.6 for the other professions). Those numbers, by the way, are not supported by the relevant statistical analysis, and are presented in a chart with a scale of 1 to 2 instead of the proper 1 to 5, suggesting that the differences between groups are much more dramatic than they actually are.
Oh, and before you give up on humanity, volunteers on average killed only 1 of 3 bugs, with many refusing to continue killing after that first experience. But we don’t know whether the number of bugs killed has anything to do with their sadism scores; I’m inclined to think that it must not be related, because such a connection would have been too good to pass up on reporting.
Granted, the purpose of the study wasn’t exactly to find out why people might choose to kill bugs over cleaning toilets or experiencing pain yourself. But making a compelling argument that sadism contributes to a choice to exterminate means considering what alternative motivations might also be at play. I would guess that the single question about fear of bugs did not get at the full range of emotions people have; I don’t have to be afraid of something to loathe it, or to find the option of ridding the world of a few insects more hygienic than cleaning up someone else’s excrement. A slight thrill of the kill might be at play here, but there are so many other possibilities and so many hints of author bias that I just can’t commit to the idea that sadism is a major influence in the volunteer’s choices.
There is an Experiment 2, and it does seem to have a more intriguing design showing that while many factors might influence a person’s willingness to “punish” an opponent with a blast of white noise (including narcissism, low empathy, and poor perspective taking skills, suggesting more thoughtless harm than deliberate harm), only people who score high on sadism are likely to increase the pain level when the opponent passes on a chance to retaliate, and to go through a chore of counting letters in a paragraph to be able to punish. Unfortunately, the problems that had leapt at me from Experiment 1 were still on my mind and primed me to be hyper-critical of this experiment as well: Why was the measure of sadism changed, without explanation? Why weren’t the raw data shown for the critical correlation, of sadism scores and number of times people worked through the letter counting task just to hurt their opponent – and look, that result is only marginally significant, when the conclusions drawn from it are made quite strongly.
I am easily convinced that people do vary in their tendencies to sadism, even in the smaller “everyday” levels we might not notice or think about. I am ready to be convinced that those inclinations toward or away from cruelty predict something meaningful about our behavior. But I am not quite convinced that those inclinations tell us something meaningful about who steps up to volunteer to squish the spider or behead the snake, and who gets rough in a one-on-one competition. Sadly, sometimes the coolest designs can be shot in the foot by their own authors’ enthusiasm for their theories.
Buckels EE, Jones DN, & Paulhus DL (2013). Behavioral confirmation of everyday sadism. Psychological science, 24 (11), 2201-9 PMID: 24022650