At 12:01 PM 5/17/00 -0400, mbattagl wrote:
>I have data that measures light intensity with a number of different
>techniques. One of the measurements (a direct measurement and "true"
>measurement of light intensity) involves lots of time, labor, and expense.
>The other techniques are more practical in the sense of time and labor, but
>are indirect measurements (based on canopy structure (density, location of
>holes in the canopy, etc). My goal is to determine if the indirect
>measurements are valid estimates of the direct measurements. However, I
>would
>also like to predict light intensity based on the indirect methods.
you have a proxy measure ... just like we use a paper and pencil test (otis
quick scoring test of mental ability for example ... and that's a
mouthful!) as a proxy measure of a more elaborate individualized measure
... which takes more time and energy (but is probably more accurate)
>If the variables are bivariate normally distributed then I use Pearson,
>but if
>they are not normally distributed I use Spearman. Is this correct?
no ... the issue here is ... what are you really interested in ... the
EXACT score value equivalences between the 'true' and proxy measures ....
OR, only the relative positioning of these values
pearson is a measure of linearity in the paired data (true and proxy) ...
but spearman has a lesser goal ... is the relationship monotonic ... ie, as
one goes up, so does the other BUT, perhaps not in proportionately equal
increments
>The regression analysis is also somewhat confusing. Regression analysis is
>based on the fact that the Y (dependent variable) is random and the X
>(independent variable) is fixed with no error. For my case, both X and Y are
>random and have some measurement error. Is it correct to use simple linear
>regression for this analysis or is there another type of analysis to obtain
>predictions?
the problem with regression analysis ... at using r too ... is that lack of
regression (that is ... perfect linear relationship between true and proxy)
could be because of problems with either X or Y or both ... a simple
regression analysis will have an error term ... but, what is it due to? you
don't know ... even in your case ... is the 'true' REALLY true? or .. just
the best you can do?
>I apologize for such a long post, but I have been struggling with this
>analysis for sometime and the more information I obtain from Statistics
>books,
>the more confused I get.
what you really want is what we call an 'equating' study in measurement ...
you would like to be able to equate some value on the proxy with some value
on the true ... in a sense, you want them to be parallel measures ...
but, even if the correlation between the two is perfect ... that does not
mean that a value on the proxy is equivalent to the same value on the true
... there could be a 5 lumen systematic error in all values on the proxy
(for example) ... this could be adjusted after of course ...
my advice is this: do a sampling of n number of measurements where you have
(let different people do the measuring so as not to bias the data by
knowing what it is) both true and proxy ... then make a simple scatterplot
... and see what the relationship looks like ...
if it looks linear AND, if the r is very high ... think about 'equating'
proxy for true ... using regression ... then at least you will have a good
idea about what the connection is ...
you don't need fancy analysis to show you if there is some connection
between the two
>Thanks in advance, Mike
>
>
>
>===========================================================================
>This list is open to everyone. Occasionally, less thoughtful
>people send inappropriate messages. Please DO NOT COMPLAIN TO
>THE POSTMASTER about these messages because the postmaster has no
>way of controlling them, and excessive complaints will result in
>termination of the list.
>
>For information about this list, including information about the
>problem of inappropriate messages and information about how to
>unsubscribe, please see the web page at
>http://jse.stat.ncsu.edu/
>===========================================================================
Dennis Roberts, EdPsy, Penn State University
208 Cedar Bldg., University Park PA 16802
Email: [EMAIL PROTECTED], AC 814-863-2401, FAX 814-863-1002
WWW: http://roberts.ed.psu.edu/users/droberts/drober~1.htm
FRAMES: http://roberts.ed.psu.edu/users/droberts/drframe.htm
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
