Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: graphing ordinal and continuous variables


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: graphing ordinal and continuous variables
Date   Mon, 23 Nov 2009 17:17:57 -0000

Thanks to Maarten for the plug. I'll add a -sysuse auto- line to the next public version of the help for -stripplot-. 

On the original question I note that even if one variable is ordinal, applying regression treats it as measured! My mental picture of an otherwise unspecified ordinal variable defaults to a graded variable with about 5 classes. What does a quadratic with such a grade on one side or the other tell you? 

Nick 
n.j.cox@durham.ac.uk 

Maarten buis

--- On Mon, 23/11/09, Robin L Jacobowitz wrote:
> I am trying to create a scatterplot showing the
> relationship between two variables, one ordinal and one
> continuous.   A regression indicates a quadratic
> relationship, but when I use the twoway scatter syntax,
> I get data points lined up along the categories of the
> ordinal variable. 

I like -stripplot- by Nick Cox for these kinds of graphs.
you can download it by typing in stata -ssc install stripplot-.
There are many examples in the help file to play with. 
Note that the help file assumes that you recognize the 
variable names as comming from the auto.dta dataset that 
comes with Stata, and which you can load by typing in 
Stata -sysuse auto-.


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index