This is the first time I've ever heard of ads for a "free IQ test" actually leading to the real thing!
The paper uses the word "significant" quite a few times, but never mentions the significance level, or confidence intervals, or any other indication of the statistical quality (or lack thereof) of this data. The large number of subjects (100K) suggests that the results may be valid, but it would be nice to see this quantified.
Also, the author says "the test worked properly in all of the major browsers", but doesn't explain how he established that, which is important to rule out the possibility that a problem with the test in some browsers skewed the results.
Still, the graph at the end (showing distribution of browser usage across IQ percentiles) is entertaining.
Furthermore, what would the usage of adblock be by IQ percentile? Opera's high showing may be due to it's adblock extension still not being as mature as those of Chrome and FF, and so the average Opera user might be less likely to use it. There's no doubt that Opera users would tend to be geeks and nerds anyway, but you could say the same thing of the top FF users. I'm not convinced that the methodology supports the apparent conclusion that the smartest people use Opera, just that of those who took the test, Opera users scored best.
So I get putting a lower bound on it, as it would take exposure to technical communities (even if it's just 4chan) that usually comes from inquisition and information-seeking, but what do you propose as the means by which there would exist an upper bound?
I imagine that people of high intelligence would be more likely to understand the greater implications of using ad blockers (driving the creations of paywalls, etc.) and so would be less likely to block ads, even when they know how to do so. It's hard to draw conclusions, though--who knows if this chain of speculation actually leads back to a significant confounding factor?
They would also be aware of game theory, and may choose to use adblock but not tell anyone, or even actively campaign against it while using it themselves.
I'm not suggesting smart people are unethical, but rather that ethicality and intelligence are fairly orthogonal, and that your proposed means seems to have more to do with the former than the latter.
As to the speculation, you are of course right. I'm just putting off studying. Speaking of which...
It's also potentially the case that intelligent people recognize that it's silly to be completely upset over ads and that few sides with really obnoxious ads have good content in the first place. Perhaps, to them, it's just not worth the trouble.
But neither is trying to convince anyone on the internet that you're smart in the first place.
I find that people with very high IQs tend to care less about practical computer usage skills. Anyone noticed that Math (or even CS theory) professors are terrible with computers?
There's a major validity problem with the sampling method, and exterm points out a pretty glaring bias, so high numbers don't necessarily improve validity, only reliability. Random sampling would be much better than opt-in sampling, but it's obviously pretty difficult to do that in practice.
A couple of points. First: In the paper, Aptiquant claims that over 100,000 took a Weschler (IV) test in 4 weeks. According to wikipedia, a Weschler test takes about 15 minutes. That's an awful lot of people looking for free IQ tests, and spending 15 minutes each on the site. Second: Check out Aptiquant's site. Maybe I'm stupid, but I was unable to find a single active 'Free IQ Test' link.
And another - Aptiquant's results (if they are real) are from a self-selected group. Does the need to take a free online IQ test relate to your IQ?
This study almost seems believable until you read the last paragraph. Since when are Web Standards "common knowledge"? How does understanding how IQ relates to browser choice help IT companies save the extra money they spend making websites work in IE? And does this conclusion mean that only IT companies have websites, or only IT companies invest the extra money to make them work in IE?
I didn't click on the IQ test... but I did click on the study. Dang.
Amusing, but it's kind of a truism. Changing your browser requires cognitive capacity. Those who change it the fastest have the highest cognitive capacity. Even without other explanations, that explains variations of version within a single browser, and to some extent Chrome and Firefox.
The really surprising thing for me was Opera. You can't explain that with mere willingness to update or try something new.
Here's the trend I see: all the browsers after IE pretty much go in order of market share. Safari isn't that special because it's the default on Macs, so even though its market share is only slightly above Opera's, the score there is dampened because there's still greater diversity in the people using it. But way fewer people are using Opera rather than Chrome or Firefox...I bet if you went back to the times Chrome/Firefox had a similarly sized market share, you'd see the same spike.
So it's not sampling bias, per se, but...I guess it's something along those lines?
Ah, now I see. It makes perfect sense. Think of it this way.
1. People who pioneer the use of alternative browsers always have high intelligence.
2. However, some browsers lack the ability to appeal to anyone beyond that initial market.
So know the graph is just saying things we already knew. Opera has a tiny fanclub, but the browser doesn't appeal to many people, or Opera fans are just bad at evangelizing to their friends. Meanwhile Chrome and Firefox have a much greater mainstream presence, and hence an attenuated average IQ.
So Opera's concentration among intelligent users isn't an indication of quality, but rather, some problem in their marketing approach.
TBH the I love the new Opera, its my browser of choice on Windows now. I still do most of my dev work in Firefox as its a swiss army knife, but opera otherwise.
On OSX, despite its memory leaks and crashing, the OS integrations of Safari can't be beat, the multi touch zooming, and the new lion features... I can put up with a crash or two a day.
And besides I like the feel of a high horse being at the to of that chart :P
not necessarily. Let's say my browser is targeted toward people with low IQ, and I get 10% of the market, biased heavily toward low IQ. Suddenly, people with high IQ realize that it's a great browser for them as well, and they all start using it - I increase market share AND raise correlated IQ levels.
It is more accurate to say that increasing market share will tend to shift average IQ score toward the population mean.
Your last statement is indeed the phenomenon I was intending to describe.
However access to the internet (and tools) has gone to the wealthy and educated first, who tend to have higher IQ (as measurable by most tests) than the poor and uneducated. Hence adoption of new browser technology is often among the cognitive elite (for access reasons) and penetration toward the masses (average IQ) depends on the overall appeal and not just a small fad among those with early access.
I've been using Opera for about 11 years now. When I started, there was nothing that was better. Even with that huge advertisement widget. It still is extremely competitive, if not leading in most things. For casual browsing it is unmatched.
It's not about one feature; it's the completeness of the whole browser. Things that you'd expect in Firefox/Chrome in extensions are built in here. It's still as fast as Chrome to the user, never mind silly benchmarks. Give it a spin some time.
I agree with balakk on all of his points, since that's when i started using Opera too, around version 2. It still is the most complete browser experience without requiring any extensions.
Another thing that makes me stick to it is the integrated mail client. It's extremely minimal, out of the way if you don't need it. My browser is always open, so having a built in client, that indexes my email, and is searchable even if i'm offline is definitely awesome. I don't need to use Thunderbird, etc, since the minimal features are all i need. Give it a spin!
I gave it shot for a while because I needed the Turbo feature. I liked using Opera, it was fast and the new chrome-like GUI is nice. However... as soon as my internet was back at full capacity I changed back to Chrome. You say you like it because of the completeness, but I found that it was missing quite a few small things which I cannot live without. Off the top off my head I can only remember two things -
Tabs cannot be dragged in and out of windows
Text cannot be selected and dragged into the URL bar
Tabs CAN be dragged - at least on Windows and Linux - not sure about Mac.
I also fell in love with tab stacking. It's just easier to group tabs into stacks in a single browser window than having separate windows. In later case I spend waaay too much time clicking wrong window.
It has some features that are fairly unique and I appreciate their experimentation. For example, Unite makes it very easy for your average person to share music, files, a streaming webcam, or whatever from within their browser. It's a great way to do something like let a friend listen to a cool album you just discovered without either of you ever leaving the browser. I still prefer Chrome on the desktop but I respect the developers and what they're trying to do.
On the other hand, I do use Opera on my phone. I find tab switching to be much more natural than it is with the default Android browser and I also appreciate the option of proxying traffic through their servers where it is compressed and stripped down to a mobile format. It cuts down on bandwidth and cpu usage which I like, particularly if the data connection is being spotty.
I've been using Opera as my main browser for years, but also uses Chrome and Firefox.
Chrome; very fast, but crashes too often, and uses a LOT of memory (and messes up my Task Manager)
Firefox; lots of interesting add-ons, but slow and memory-hungry (and too often updated, with too many unstable updates)
Opera; very fast, very stable, best with regards to memory use. However, has problems with a few web sites (not compatible, maybe because they haven't tested "leading edge" technologies on Opera ?)
> it is a clear indication that individuals on the lower side of the IQ scale tend to resist a change/upgrade of their browsers.
AptiQuant needs to review their correlation/causation definitions. Resisting change is one plausible explanation, but the only thing the data clearly show is that the people who scored low on this IQ test tended to be using out of date browsers at the time they took the test. "Resistance" implies agency, which correlation studies do not establish.
I choose to go with the reverse explanation: older browsers actively reduce user intelligence.
> I choose to go with the reverse explanation: older browsers actively reduce user intelligence.
Do you have any reason for that? Because frankly it just sounds ridiculous.
It would make sense that someone with a higher IQ would be able to consider the possibility that their browser was replaceable and take the necessary steps to replace it. But I don't see how which browser you use could have an appreciable effect on your intelligence.
Excellent. Now I can convert all of the people I know who still use IE to opera. "SMART PEOPLE USE OPERA, YOU WANT TO BE CONSIDERED SMART, DON'T YOU?!"
Perhaps some browsers are particularly hard to use and you need to use those extra IQ points figuring out how to get them installed or running. In other words, this might be a dubious accomplishment.
It is common knowledge, that Internet Explorer Versions to 6.0 to 8.0 are highly incompatible with modern web
standards. In order to make websites work properly on these browsers, web developers have to spend a lot of
unnecessary effort. This results in an extra financial strain on web projects, and has over the last decade cost
millions of man-hours to IT companies
Or, dumb people reduce functionality for the rest of us because work has to go into handling their mistakes rather than coding the better product the devs really wanted to make.
The paper uses the word "significant" quite a few times, but never mentions the significance level, or confidence intervals, or any other indication of the statistical quality (or lack thereof) of this data. The large number of subjects (100K) suggests that the results may be valid, but it would be nice to see this quantified.
Also, the author says "the test worked properly in all of the major browsers", but doesn't explain how he established that, which is important to rule out the possibility that a problem with the test in some browsers skewed the results.
Still, the graph at the end (showing distribution of browser usage across IQ percentiles) is entertaining.