Bias2012/12/10 - Author: Marco Neves - No Comments
We watch the world through our senses – mostly our eyes if we can see, but not exclusively – but it is our brain that see it, after filtering it to better fit in our image of reality.
We have bias – things we believe to be true, behaviors we think are universal, expectations people – specific or in general.
Our mind does the best it can to make the world around us fit with our view of the world, by applying what we believe over everything we see, ignoring as much as possible that contradicts that view and getting our attention to anything that enhances that view.
“A lie repeated enough becames true.” doesn’t mean that the true will change just because someone decides to repeat a lie as if it was a mantra. It means that if we hear the same “fact” from a few sources we start to believe it and after that it is a lot harder for us to see anything that says that we are wrong than anything proving us right.
And that is what bias are. Things we believe to be true, and that have an impact on our perception of the world.
Bias are not restricted to some subjects or types of information. Bias are not restricted to some groups of people and do go away with education. Even our behaviors and our interpretation of others behaviors are altered by bias.
Even our abilities are influenced by bias. If you believe that a subject is really hard to understand, you’ll not try as hard to understand it as if you thought it was really easy.
If you think a guy is a playboy you’ll see every time he talks with a girl as a flirt or a pickup attempt, but if you think that a guy is gay you would see the same conversations as normal.
If you think a brand is expensive, you’ll notice more the price differences if a product of that brand is more pricey that a concurrent product, but you’ll notice more the superior characteristics of the concurrent if it is cheaper.
Bias influence our decisions on a very large degree. We are more likely to buy a car from a more expensive brand if we believe that brand to be more reliable than the alternative brand (even if in the end both cars are made in the same factory using the same pieces).
Most of us today tend to believe in statistics and think they are serious, we were made believe that averages tell a story. However, more than 80% of all published statistics lays or misrepresentations of the facts (this was just made up – it’s a common joke in the internet, but not completely inaccurate). Most statistics published in the mainstream internet and in the traditional media are not really representative of the reality – averages (the common published metrics), median and modes often tell different stories, and often we see averages being used as if they were the modes, when they are two completely different things.
Often, and this is probably the most common problem with statistics, the sample used is too small to be really indicative of any true reality – statistics based in interviews of a hundred persons is only representative of a small universe – and sometimes it is even hard to understand which universe. One person goes to a shopping and asks people how much they usually drink – and found out that in average people drink 4 beers each day. A second person tries to reproduce the study and ask the same question to the same amount of people – and finds out that in average each person drinks 1 glass of wine per day. What was never published is that on of them made the interviews before a game in the club in one of the sides of the shopping, while the other made the interviews after service in the church in the other side.
Even in small universes is easy to get biased statistics. Imagine that you are in college and want to make some statistics about the students of your college – let’s say you want to find out how much do your colleagues read. Most likely you’ll try to interview the people you know better, and if you spend a lot of time in the library you’ll try to interview people next to it, but if you go often to the bar next to the college, you’ll try to interview people there. Can you imagine how that will make a difference in the results?
But, more important, can you see what’s the most important part on the previous examples? More than anything else, they show my own bias – that people who go to bars read less and people who read more go less to bars – and that football fans drink more beer while religious people drink more wine. In the end, if you are aware of your own bias, if you manage to notice them, they tell a lot more about you than they say about anyone else.
But, the important thing about statistics is that we get them in a daily bases and those statistics create bias in our mind. And we even apply those bias to statistics we see. We end believing easily in statistics that confirm our bias than in statistics that contradict them.
And that’s why bias are most of the time self-reinforcing entities, entities that we are not fully aware of, and that make a huge difference in our lives.
Categories: Definitions - Tag: bias, life, statistics