The previous tests were brand new Clearview signs beating old-and-faded classic signs -- which seems like an unbelievably embarrassing mistake.
Only when someone points out to me that the signs they are replacing are probably not new would it become "obvious". (In fact, while I read my first article on this, my internal thought was 'so why isn't this better if it tested so well?', then they mentioned the old signs, and my thought was 'oh, right, that makes sense'.
So embarrassing mistake? No doubt. Unbelievably embarrassing mistake? No way, not with the sadly believable mistakes I've made in the past.
A/B testing is notoriously hard to do well, especially if you're comparing something the user already knows well with something different. Controlling for familiarity is pretty much impossible.
Basically a "you failed science 101" level of embarrassment mistake in my estimation. Testing two brand new signs with the different fonts should be a fairly obvious first idea. The hard/interesting part is taking into account learning effects for the previous font. There's a pretty interesting research question in how you'd age signs in there as well (as you want to know if the gains are positive over the entire lifetime of the sign) and some interesting ethics questions about field testing.
Meh. The American way would be "Build a street for font, and see which one people are willing to pay more for"
“Helen Keller can tell you from the grave that Clearview looks better,”
and thinks it could be a matter of cost according to the article[5]. TerminalFonts posted a hilarious tweet about it as well[4]. They seemed to have done a decent job researching it, and it was not as cut and dry as simply an a/b test with old signs, but it was rigorously documented. Although, I find reading font testing methodology absurdly boring so I skimmed it. Not really sure what the cost to the government was, but they are certainly killing the font according to every source.
According to the citylab article:
The FHWA has not yet provided any research on Clearview that disproves the early claims about the font’s benefits.
So, I have to apologize, I briefly insinuated that maybe these guys forged the test to get the contract. Instead, they seem like pretty passionate design nerds who meticulously researched this topic and then developed a font around their findings.
Apparently, they did a poor job though, according to some unreleased report.
[1]http://clearviewhwy.com/ResearchAndDesign/_articles/TRB_Pape... [2]http://www.terminaldesign.com/web-fonts/ [3]https://en.wikipedia.org/wiki/Clearview_(typeface) [4]https://twitter.com/TerminalDesign/status/693173610538799105
[5]http://www.citylab.com/commute/2016/01/official-united-state...
PRICING from www.citylab.com
==============
Jurisdictions that adopt Clearview must purchase a standard license for type, a one-time charge of between $175 (for one font) and $795 (for the full 13-font typeface family) and up, depending on the number of workstations. (Meeker and Associates disclaimed exclusive rights to the “Clearview” copyright.)
I hope those weren't the only tests that were conducted. Two groups of 12, 65 years plus, daytime and night time. Word recognition test+legibility test (2 studies with the same setup/subjects etc.). Tested words: Purcel, Dorset, Conyer, Bergen, Ordway, Gurley. Parked car (1993 Ford) at varying distances. Repeat-measures design. ANOVA for material and ANOVA + t-tests for fonts. They used the same sign with bolt-on names for all different cases so the issue linked in the OP doesn't apply to this test.
So we have no weather conditions, a very selective population (also notable that gender isn't mentioned at all which is normally standard practice in the subjects section), a questionable list of words (no mention what word/letter distributions look like and why this specific group was picked...they do explain why they built the word group the way they did however). Only one vehicle type. No discussion of sample size, power etc. in the linked studies.
It's a decent set of constraints for a single test but we're talking about highway signs for the general population I'd expect more extensive tests than that.
Also, they ran 2 tests and they somehow used, or read research by someone who had run, a computer simulation of font degradation to anticipate how it would age.
So, I am not sure this is nobel prize worthy, but a bit disingenuous to claim they they are complete morons.
edit: while I don't laud the amazing efficiency of our peer reviewed science system, you might also find these included works helpful, these ones seem relavent. I must admit, not being even remotely interested in traffic fonts or the testing of them, I have not read them.
Peer reviewed research papers by Donald Meeker as author or co-author have been published by:
Transportation Research Record, National Academy of Sciences (2 author, 3 co-author)
Ergonomics in Design, the Journal of the Human Factors and Ergonomics Society (co-author)
Transportation Engineering, the Journal of American Society of Civil Engineers (author)
We have drastically different tastes in comedy...