How do you think you compare to others? That’s the question posed by the ThanAverage website in what it calls a “small unscientific investigation into how we value and compare ourselves to each other.” It prompts:
Imagine you are in a room with 100 strangers. Imagine they’re similar to your peers and neighbours. Based solely on your own instinct, perceptions, and self-reflection, answer the following questions.
For each question, you click a button to indicate how you think you compare to the average: do you use social media less, are you more punctual, do you have better dental hygiene, are you more humble, and so on. The site presents 95 questions in random order (and welcomes suggestions for more). Response rates are robust, ranging from approximately 16,000 for newer questions to over 40,000 for those that have been in place longer.
Most amusing was the classic, “Are you a better driver than average?” When I answered it (above average, of course), the ThanAverage audience revealed itself to be far more self-aware than average, with precisely 50% of respondents claiming to be above average. Perhaps they are aware of other studies showing that up to 80% of people claim to be above-average drivers, which triggers Dave Barry’s quote: “The one thing that unites all human beings, regardless of age, gender, religion or ethnic background, is that we all believe we are above-average drivers.”
Our Do You Use It? polls have long revealed how self-selection influences survey results. I’ve always believed that, like the children of Garrison Keillor’s Lake Woebegon, TidBITS readers are all above average, and now you can quantify that for yourself at ThanAverage. Or at least get an opportunity to reflect on polling weaknesses and psychological biases. Hat tip to Recommendo for the link.
To some (perhaps large) extent, I believe this is because each driver has his or her own belief system about what makes a good driver. The person who swerved in front of me on the interstate with four lanes going my direction might think that a good driver can judge clearance well. (I haven’t believed this for many years, but since getting a vehicle with adaptive cruise control that I cannot turn off, I really don’t believe it.) Someone else might think that a good driver uses or doesn’t need to use turn signals. And many people seem to have different opinions about how much faster than the posted speed limit the traffic should go. “Everyone who drives faster than I do is a maniac; everyone who drives slower than I is a moron.” An acquaintance claimed to be a really good driver and the two accident he/she had been involved in were the other drivers’ faults. I failed to make a good impression when I observed that a really good driver might have avoided the other drivers. (At least I didn’t say “the other really good drivers.”)
@Will_M has a really good point which dovetails with my view. What the heck is an “average driver”?
This is not a problem of statistics (where more than 50% of a population can be above average*) but one of perception. The same applies to the majority of questions I encountered on that site.
The word “average” means different things to different people.
Your example is the arithmetic mean, and you’re 100% correct in that context. Outliers can shift the mean such that a clear majority of samples are above or below it.
But I suspect that in the context of these kinds of questions, most people are treating it as the median, which is, by definition, the point at which half of the samples fall on either side.
Nevertheless, you’re right that this is a matter of perception, not reality. We don’t actually know what the whole population is really like, and our impressions of others are always relative to ourselves.
The problem is the premise that the 100 strangers are “similar to your peers and neighbors”. First of all, peers and neighbors are different people – well, my peers and neighbors are totally different people. Second, I suspect that people who look at this kind of site, including their peers and neighbors, are probably more educated or otherwise different from the average (global?) citizen, so you get a skew right there.
The first question I got confirmed that: “Are you more or less religious?” – 83% of participants say that they are less religious, which is exactly the opposite of the “average American” (78% of Americans are religious, according to Gallup 2024).
It would be indeed an amazing site if they asked some background questions to benchmark answers to some reference :)
I gave up after about 8 questions - just seemed a bit nonsensical. “Do you dream more than average?” How can you possibly answer something based on how other people dream?
Most of the questions are just subjective opinions (which is of course, imho ).
I gave up after three questions to which I wanted to answer ‘average’. It’s all a bit binary for me, lumping together people who think they are ever so slightly better or worse than average with those who are convinced they are substantially so. I therefore find the results meaningless
It’s a recent Kia, and I expect that the feature exists across the Kia lineup (and I wouldn’t be surprised if many manufacturers do not allow disabling this “safety” feature).
Kia actually warns the customer that the adaptive cruise control can get false returns from a median barrier, and I have experienced that in mountain driving. I’m driving at the speed limit, the road curves somewhat sharply due to the terrain, the sensor notices the median barrier, and the brakes come on.
This is one of the reasons I noted that the site was “an opportunity to reflect on polling weaknesses and psychological biases.” :-)
There are all sorts of others, like the apropos Better-Than-Average Effect, the converse Worse-Than-Average Effect, Selection Bias, and so on. In the end, it’s mostly fun.
It sounds like you’re talking about collision avoidance. Adaptive Cruise Control is the feature where you set a maximum desired speed and the car will automatically slow down as necessary based on the car in front of you.
Very much my, and others’, point. What does your average person believe “average” means? Both definitions are very precise and I think, perhaps, most people wouldn’t give either definition if pressed. Outside of a group of numbers it’s an amorphous concept that just acts as a peg to hang ideas on. Not data.
).