[From Rick Marken (950123.1010)]
Me to Bruce Abbott:
WHY did you ask the question that set this all off, which was:
Do the PCT-analysis programs you have written estimate the reference value
from the data?
Bruce Abbott (950123.0915 EST) replies:
I simply wanted to know if this is typically done in your analyses and, if
so, how.
Let me show your complete statement:
Do the PCT-analysis programs you have written estimate the reference
value from the data? When the data set contains only cursor and handle
positions, (i.e., lacks the disturbance tables), the disturbance values
can be calculated exactly only if the reference values are known
If you just wanted to know whether reference (or target) value estimation is
typically done in our analyses, then why did you go on to say that
disturbance values "can be calculated exactly only if the reference [target]
values are known". Given the close proximity of your question about estimates
of reference [target] values to your statement about needing these values to
reconstruct d, it seemed to me (and to Tom too, apparently) that you thought
you had to estimate the reference [target] value (rather than just look it up
in the program code) in order to reconstruct d. That is, it seemed like you
didn't understand that c = h+c+t.
In fact, it's not true that you have to know the "target value" in order to
reconstruct d. What you have to know is whatever constant happened to be
added to c before it was displayed. In the THREECV1 program it just happens
that the contant added to c (what we are calling t) before display was
equal to the horizontal position of the target lines. But the target lines
could have been placed somewhere else on the screen (say at maxx/2-50); in
this case the target value is not equal to the value of t added to c before
display, which is maxx/2. So it is not true, in general, that you have to
know the "target value" in order to reconstuct the distrubance; you have to
know what was added to the disturbance to get c in order to reconstruct the
disturbance. This is just basic math and it has nothing to do with PCT or
with "estimating" reference or target values.
Me:
By the way, do I ever get to hear what you think these analyses of the
THREECV1 program data...imply about the experimental and statistical
methods used in psychology? I've asked three times now. You're not
avoiding answering, are you?
Bruce:
No. Yes.
Thanks for sharing:-)
Actually, I'm waiting to see what happens when something more powerful than
multiple regression is applied to the data, specifically, path analysis.
GREAT!!! I think it would be a good idea to try the whole nuclear arsenal of
social statistics on this little tracking experiment. Even if none of these
analyses can salvage the cause-effect model of behavior, it would show that a
little tracking experiment can cause BIG problems for psychological
"science".
Best
Rick