Has anybody done a properly controlled study on typist speed between QWERTY and Dvorak keyboard layouts? I am curious about whether people actually achieve significant speed improvements, but I've yet to read anything that is non-anecdotal. A typical conversation:
Dvorak: "I switched over two years ago and never looked back! My colleagues hate me! I think I might type faster now."
QWERTY: "My colleague switch over two years ago. I sometimes have to use his computer, and I get very annoyed when I have to switch it back to QWERTY. I still type faster than him."
Dvorak: "QWERTY was designed to slow down typists to prevent jams in obsolete typewriters! We don't use typewriters, so why is our keyboard layout catered to them?"
QWERTY: "The keyboard layout is a well-established convention, much akin to the use of English units of measure in the US. Sure, a transition is possible, but is it possible to recoup the retraining costs in a realistic time frame?"
... and from there, a pointless dialectic ensues in which QWERTY and Dvorak neglect to perform any valid assessments of such minor empirical details as who can achieve better typing speed or has a reduced risk of RSI.
What I'd like to see are the results of words-per-minute tests - for instance, those that force the typist to retype random bits of literature - split among users that rate themselves as light, moderate, or heavy typists, with information on when the QWERTY-Dvorak transition was performed to try to understand the magnitude of QWERTY's historical advantage. Ideally, the study could include pre-transition QWERTY speeds for comparison with Dvorak speeds, and perhaps measurements of Dvorak speeds for various periods of time post-transition.
I should not need to say it, but just in case: It should be obvious that a questionnaire on a topic with as much fervent zealotry behind it as the QWERTY vs Dvorak discussion cannot control for selection bias. The study must perform actual, controlled tests, and is ideally longitudinal.
Anyone know of any such studies?