I don't know what Phase 1 and Phase 2 mean here. I
assume just two measurements.
I don't think that the graph I think you have in
mind is necessarily going to be the most effective.
In particular, exchanging axes may be effective.
The main logic to that is that cholesterol as a response
is better displayed vertically; in particular differences
are then explicit as vertical distances as well. (If
ratios are of interest, use a log scale.)
See
McNeil, D.R. 1992. On graphing paired data.
American Statistician 46, 307-11.
which in your case should be accessible through
www.jstor.ac.uk
For Stata implementations of graphs for comparison,
see (e.g.) -pairplot-, -parplot- and -linkplot- on SSC.
Some relevant material also at
http://www.stata.com/support/meeting/9uk/compare_gph.pdf
Nick
[email protected]
Zoue Lloyd-Wright
>
>
> I would be most grateful if anyone can help me, apologies for the
> stupidity.
>
> I have 42 subjects, I would like to graph their cholesterol levels
> Phase 1 and Phase 2.
>
> BUT!!
>
> I would like to place the 42 subjects individually on the yaxis and
> their corresponding Phase 1 and Phase 2. data on the xaxis.
>
> It does not matter if it is a dot or line representation
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
