How effective is the Dvorak keyboard layout (or other alternative layouts) at improving efficiency/productivity at the computer? Has any research been done on improved typing speed or productivity?
Putting a group of QWERTY typists against a group of Dvorak typists is not proper research.
Dvorak research is usually biased, as to get a perfect measurement, you would have to compare a person typing QWERTY for X years against the same person typing Dvorak for Y years. And even then, the setting he will be in will most likely differ.
You should base yourself on pure analysis rather than actual typist research, as these are way better at showing the differences between various keyboard lay-out.
Let me show you some statistics for your post...
QWERTY lets your fingers travel
The home row is used about three times more and the bottom row is reduced a lot, rather than going from top to bottom over and over again.
is turned into
Let me show you some statistics for Josep's post...
QWERTY lets his fingers travel
Again, we see that Dvorak let's the home row used a lot more than any other row.
is turned into
Just by this non-biased, you can tell that it's just 4%.
Two notes I'd like to add to what's been said already:
Depends on what you're doing. I find that it makes typing general papers (reports, memos)—"formal" documents much easier. It just feels better and there is less fatigue. Oddly enough, less formal documents (chat, for example), might feel a bit weirder; there may not be any improvement. For coding, it is sometimes slower, especially as many symbols are at the nether regions of the keyboard and accuracy suffers there. YMMV.
I learned typing in QWERTY in high school but never had much need to type a lot. I broke my arm and lost some movement in my right arm. After I went to college I could not stand typing papers for more than an hour at a time because my wrist would hurt to much. I then started typing DVORAK to see if it would help with comfort and it completely has. I will have a little soreness if I am at the computer all day, but generally I'm fine.
That being said, I agree with the other people who said DVORAK is more for comfort than efficiency. You actually may lose some productivity because of shortcut keys moving around etc. I still type with QWERTY and DVORAK because I support other people's computers for a living but I am about 60 wpm in DVORAK and 45 wpm in QWERTY. My recommendation is if you are planning on making a living by sitting at a computer, switch to DVORAK for comfort but know that you will take a hit in productivity for at least 4-6 months.
For considering effectiveness of Dvorak layout, I think you should also consider the learning curve to "master" it.
We are so used with QWERTY that it feels natural for me. While that might make me type 4%, 10% slower than on Dvorak, my productivity bottlenecks are others, such as external distractions, not focusing on the right things (my own distractions), non-effective meetings, etc.
So, say I'm willing to master it. It will take a long time and a lot of effort (while having lot less productivity typing slow) and the gains would be very low. So, is this efficient? Is it really worth it?
I guess I would only learn that just if I typed a lot, to avoid wrist pains. Even though I think that ergonomics would matter more than that, in this case.
Dvorak is a great security bonus too. When someone finds my computer unlocked, they can't update my Facebook to something nasty our send out a company-wide embarrassing email :) . Plus I use the Das Keyboard which has unmarked keys :)
I type on Dvorak and have for the last 6 years. I have to switch back when at someone else's keyboard. I'm slower on qwerty, but that's changing. Dvorak is way more comfortable and I never get tired from typing.
In windows, I map Ctrl-Shift-1 to set keyboard to Dvorak and Ctrl-Shift-2 to set it to Qwerty. I disable the toggle shortcut as that causes more inadvertent switches.
For my phone's keyboard I use Swype. Much faster than any other keyboard on a phone. As a side effect, it's teaching me Qwerty layout.
I personally type in both. I prefer Dvorak, although that may be for status and preference reasons rather than actual efficiency. I believe that having learned Dvorak two or three years ago I may have put more effort into it. I know I'm quicker and more accurate, but it may simply be because I learned later.
Switching back and forth isn't much of an issue for me, because I typed in qwerty for so long. I generally have an minute or two of confusion and then it clears.
Most Windows 7 machines can be set to use Dvorak by using the Language bar. The only portability issues I run into are in other people trying to use my left handed mouse and weird unmarked keyboard. Generally, though, I convert really well now that I'm comfortable in both, but I consider it more a preference than an efficiency thing.
Qwerty touch typer for 10 years, Dvorak touch typer for 8 years.
Comparing myself to my peak speed at Qwerty and peak at Dvorak I would say it's about the same. The difference is negligible.
Dvorak vs Qwerty
The benefits really stop there
I usually don't recommend changing. I changed because of tendentious / RSI, slowed me down for 2 weeks while i learned it. Gave me a chance to heal some... but overall I usually do not recommend it.
I switched as part of a formal research study when I worked at Tektronix Computer Research Lab on human factors engineering in the 80's.
I don't have a copy of my final report, but to summarize:
Nearly 30 years later, I still use Dvorak, and won't go back.
As a security plus, I use Dvorak proximal patterns for passwords that are simple to remember and quick to type, but which are completely meaningless to Shoals typists. For example, "4puk7ghm" is the first-finger column of right and left hands in Dvorak but are all over the place on Shoals. It also tends to meet security systems' password constraints.
To those worrying about "two on a tube" pair programming, I actually found it to be an advantage, as long as you can quickly change with a hot key or physical switch.
I worked with someone who was constantly grabbing the keyboard or mouse, and having to switch layouts both tempered that person's "grabbiness," and also allowed for a "context switch" period in which to gather thoughts.
This is similar to requesting others first ask, "May I interrupt?" rather than just launching into asking something. I also switched to left-mousing for the same reason. In pair-programming, good fences make good neighbours… :-)
One other thing worth noting that I haven't seen mentioned here: the Dvorak layout was designed for English prose, not computer programming, which generally contains a lot of non-alphabetical characters. Other languages or a specialized vocabulary also may temper some of Dvorak's advantages.
More anecdotal evidence pro Dvorak: I have a programmer friend who switched to Dvorak from Qwerty a summer ago when he started having shooting pains in his wrists. Now his speed is about 20 wpm slower (85+ instead of 115) but is he going to go back? (Ans: Dvorak has completely eliminated his wrist pains.)
I was also at the same point: completely comfortable in Dvorak, but my typing speed was about 60-70 WPM compared to a former 80-90(?) in Qwerty. Even if I have to toggle it before others can use my computer, Dvorak is supremely more comfortable. (Then I found Typeracer and Typing.io and broke a barrier I didn't think I could.)
I don't buy the whole "Dvorak is better because alternating fingers." Would you rather type "asdfjkl;" or "ajskdlf;" ? To me, much of Dvorak's comfort comes from more sequences being of the first runny kind than in Qwerty. (I also don't buy the "more-used letters on home row" because what's more-used for me is different (cough Vim) and it's more natural to splay my hand than curl 8 fingers on the middle row.)
For what it's worth, Dvorak is a pain for programming, but an addiction to Typing.io fixed that. I just count my blessings I don't have to program on a German keyboard anymore.
Another marginal benefit to knowing Dvorak: I went to type something on my friend's computer (I was in my back-to-Qwerty phase at the time):
Finally, it's not how fast you type; it's all about your cognitive load.