Dogs and men: A toxic combination?

Posted on January 10, 2012

0



Image: Danielle deLeon (via Wikipedia)

Dateline: 3 November 2011 // Posted by: Cillian Ryan

Source: Discovery News, Daily Mail, AllPetNews.com, The Science Bit

Story: “Dogs Walked By Men Are More Aggressive”

Summary: A new study published in the academic journal Applied Animal Behaviour Science claims that dogs walked by men are four times more likely to threaten and bite other dogs and that dogs walked on a leash are more likely to act aggressively than dogs off the leash. The study conducted by Petr Rezac and his colleagues at Mendel University in the Czech Republic studied close to 2000 dog interactions on owner-led walks.

They observed across 30 different locations in the city where owners frequently walk their dogs during mornings and afternoons. Whenever two dogs walked past each other, they recorded a set of specific observations on a checklist. They also restricted themselves to single-dog/ single-owner cases, excluding owners with multiple dogs and single dogs with multiple owners.

Petr Rezac and his colleagues wrote: “We propose that the occurrence of threat and biting in dogs on a walk may have some connection with aggressive tendencies and/or impulsivity in people,” adding that “dogs are able to perceive subtle messages of threat emitted by another dog. Simultaneously, dogs are unusually skilled at reading human social and communicative behaviour.”

Why is this PseudoNews? There are certainly some aspects of this article that lead us to think that this is pseudoscientific news. In particular, some of the research findings: the researchers found that male dogs “sniffed” female dogs more than female dogs “sniffed” males, puppies played with other puppies more than they played with adult dogs, dogs behaved more boisterously when on their leashes than when off their leashes and that male dogs appeared more threatened by other male dogs, and female dogs by other females. These findings certainly don’t seem like anything that we didn’t already know about dogs in particular, and also about sexual behaviour in other animals for that matter too. The “sniffing” they speak about is of course an attempt to start mating, and it’s known that males would usually be more likely to initiate this. As for the dogs appearing threatened by other dogs – dogs, males in particular, are always going to feel more competition from their same sex. Dogs are also going to feel a lot more confident and cocky when on a leash being led by their owner, compared to off the leash where they may feel more timid.

However it was the main finding of this research that leads them down the pseudoscientific road, which was, that dogs are four times more likely to bite other dogs when both owners were male. This suggests that because of males being somewhat more boisterous and aggressive than women, somehow pollute the mind and the morals of their dogs and force them to become more violent towards other dogs.

What features of pseudoscience are on show? The first feature of pseudoscience that can be seen on show in this article is, in terms of the statistical significance of hypothesis tests, a risk of false positives. This can be seen when the investigators tested such a large number of separate research questions, they failed to use any statistical method to control for multiple testing. When so much statistical analysis is conducted probability means that eventually a result will emerge. This result may look like a finding, however, it is not really a “true” finding – it is a false positive.

Confirmation bias can also be seen in this study. The research project looked at over 1870 dogs and the analyses related to only one dog from each meeting (i.e. to half of 1870, being 935 dogs). Of these 935 dogs only 3% were seen biting. This means that the researchers were working with a sample of just 28 dogs when looking at the impact of the owner’s gender on the dog’s aggression. This finding is in line with prevailing social prejudices about men and masculinity, therefore we can see that confirmation bias is evident here.

The moral of the story: Though this study may have had some minor promise at the beginning, the bottom quickly fell out. There will not be too many people hanging up their leashes and retiring from dog walking in fear that they may run into a dog whose mind has been polluted with violence courtesy of his male owner.

Advertisements