#subconsciousmind #biasindata #studies #baitingheadlines
And so are people with a Mercedes Benz… It’s true, recently I was learning about how easy it is for us to fall prey to catchy headlines. Headlines that tell us what we want to hear, or that strengthen our points of view… hence the cheeky post if you are a dog-owner 🙂
You have all read the headlines to which I’m referring. “Drinking xyz is good for your heart”, “Chocolate helps you lose abc”, “People with 123 are happier”, etc, etc. At a conscious-level, we know they are all b.s., except, we mostly function at the sub-conscious. We seldom take into account the influence our own biases have in everything we do, from the subjects we pay attention to, to the way we interpret the subject itself, to the very data we choose to include. The latter is a perfect example of the long lineage of headlines exalting dog-ownership.
According to Professor of psychology Hal Herzog, there is very little evidence to prove dogs, or pets, are good or bad for us. In fact, the RAND corporation followed up on this very question and concluded the same.
Pet owners and non-pet owners differ across many socio-demographic variables, such as gender, age, race, living arrangements, income, and employment status. These differences are also associated with health, so when trying to draw causal inference about pet ownership using a general population sample, selection bias should be accounted for (or at least acknowledged), as it could lead to an over- or under-estimation of pet ownership’s true effects. In our analyses, it appears that it may inflate them, as pet owner characteristics are associated with better mental and physical health outcomes. This is not a new problem, as selection issues have plagued observational research, with many methodologists and statisticians advancing new methods to deal with this problem that used to confound any meaningful analysis. We recommend propensity score matching utilizing boosted regression since the exact relationship between socio-demographic characteristics and pet ownership is unknown.
2017 Saunders et al under CC BY
Getting unbiased data for what appears to be a simple question is more difficult than you would think. And that is the point I’m trying to make. The point is that often we look at a headline and we gravitate to them using our sub-conscious and manipulate it to agree with us. Not only do we use our oldest part of the brain to make these decisions, but worse, these biases can happen even by the very people we trust to provide us with information.
Next time you look at a headline, specially those catchy ones, think twice. Think once to get your conscious brain into gear, and another time to remind that conscious brain of yours the foundation of the article may be compromised.
