QWERTY vs DSK: How to Test Them?

susupply@aol.com (SUSUPPLY): (on the Dvorak vs QWERTY studies)

The pro-Dvorak studies did not have proper controls, that is why 
they are suspect.  The studies that found little or no advantage 
did attempt to control,  I can't speak as to whether the controls 
were adequate.


>Tony Minkoff:
> off the top of my head, I don't see
>how to do this.

I don't blame you, but I applaud you for recognizing the difficulties 
inherent in constructing a valid study.  That puts you one up on a 
scientist (who should know better) who regularly posts to this thread.


Hi,

You mean that there is ANOTHER scientist who is posting on this
thread?  ;-)

Yes, I agree that the Navy data is suspect since the director
(A. Dvorak) had a financial interest in the outcome. And that
a valid study would require some care (and cost).

     DOING IT MY WAY

If someone will provide the money, I would be willing to conduct
such a study. I would want a group of about 100 high school students
who do not type but who want to learn (and don't care which keyboard
or even don't know that there are different kinds). They would
be paid to take part in the study, both for incentive to stay in
the study,  and because half of them may learn a skill they cannot 
use.  But using "virgins" (so to speak) would avoid the complications 
of prior training.

After signing up, they would be randomly divided into 2 groups (of
50).  One half would be taught typing on computers using QWERTY 
keyboards and the others on identical keyboards set for DSK, and 
where the key labels had been blanked out on all of the keys for
all of the computers. (I recall that when typing was taught
in my high school, typewriters with blank keys were used: the
students were to KNOW where each letter is without looking).

This would be done during the summer so as not to interrupt other
classes, and so the typing could be done for several hours a day.
And they would be instructed NOT to practice outside of the class.
(Which would prove to be very confusing for the DSK group!)

The teachers (probably 2, maybe 4) would rotate daily, each 
teacher with the QWERTY students on one day and the DSK the next.
Testing would be for speed and accuracy of typing, copying the 
same documents for all 100 students).

The "learning curve" for each student would be monitored until
each had reached a plateau. The rate of increase in speed and
the final plateau would both be measured  for each student, and
the grading and results compiled by someone who did not know 
which keyboard was being used by which students.  A special 
incentive bonus would be given to each student when/if they 
could type 50 wpm for 5 minutes, and another bonus when they
reached 70 wpm.


MY GUESS AS TO THE WINNER: ERGONOMICS 101.

Based on looking at the two keyboard patterns, I would
expect Dvorak to be better. But what advantage in speed or
hand stress would that translate into?  Would need a study
like that described above to say.

My prediction of the outcome is based on this: look at the
English letter frequency data at:

http://www.trincoll.edu/depts/cpsc/cryptography/caesar.html

Notice that the most frequently used letters in English are
e,t,n,r,o,i, a, and s. (as those of you who have played with
cryptography know).


Then look at the Dvorak keyboard at:

http://www.mwbrooks.com/dvorak/


And at your QWERTY keyboard.

Notice that all of these letters except "r" and "I" are directly 
under the finger rest positions on the DSK layout (and r is an 
up reach for the right hand and I is a left shift for the left
index finger).

On the QWERTY layout only "a" and "s" do not require reaching, 
and "n" requires reaching down, which is harder than reaching
up.

From this, I conclude that the Dvorak DSK requires less finger 
movement to touch type a given document. It uses the 
alternating right and left hands for the common alternating 
vowel/consonant pattern, which should be an advantage when
typing fast.

I can't say from this that DSK will result in faster typing
in my study, but that is the way that I would bet.


GENERAL CONSIDERATIONS

What general principles about understanding the Universe do
I see in this QWERTY vs DSK example?

Well, I think one needs to look not only at the various published
studies and data available, but then also at the basic theoretical
and "common sense" considerations also.  I think data and "studies"
can be misleading. See my web page section CAN YOU TRUST NUMBERS?
for some examples.

Here, I think the ergonomics of the keyboard plays the role that
common sense plays in the "does the minimum wage reduce jobs?"
and that basic physics plays in the "does CO2 change the climate?"
and the "does Freon reduce the ozone layer?" debates.  In each of
these, while the data may prove "common sense" or basic theory to
be wrong, I need more convincing that they are wrong than that they
are right.

                     ,,,,,,,
_______________ooo___(_O O_)___ooo_______________
                       (_)
jim blair (jeblair@facstaff.wisc.edu) Madison Wisconsin
USA. This message was brought to you using biodegradable
     binary bits, and 100% recycled bandwidth.