R help - Mar 2020 - Syntax for geom_point in ggplot2, distinguishing point types.

I have an application in which I want to plot points with different 
colours and symbols to distinguish (three) types of point.  I thought I 
had it figured out, mostly by a trial-and-error approach and by 
simple-mindedly following instructions from others, without (sad to say) 
really understanding what I was doing.

I have now encountered an example where my syntax screws up, and I can't 
see how to fix it in a coherent and generally applicable way.  The 
problem arises if the data column (factor) specifying the point types is 
missing one level.  When this happens, the types get out of synch.

I have attached a source-able file "demo.txt" and the two data sets
upon
which it depends as "data1.txt" and "data2.txt" (dput()
format).

The result from using the data from "data1.txt" (plotObj1)
demonstrates
the problem, i.e. the unwanted phenomenon.  The result from using the 
data from "data2.txt" (plotObj2) is "as it should be".

In the data from data1.txt the "extremes" column, which specifies the 
point type, is missing the level "0":
>        table(Dat1$extremes)
> 
>        0 intermed        1 
>        0       39       43 
As you will be able to see, the plot of these data has the "0" symbol
(a
triangle) where there should be an "intermed" symbol (a solid dot),
and
the "intermed" symbol where there should be a "1" symbol (an
upside-down
triangle).

The data from data2.txt have an extremes column with all three levels 
and the corresponding plot is as desired.

I'm pretty sure that the solution to my problem is quite simple, but I 
am too stupid to figure it out.  Can anyone guide me in the right direction?

Thanks.

cheers,

Rolf Turner

-- 
Honorary Research Fellow
Department of Statistics
University of Auckland
Phone: +64-9-373-7599 ext. 88276

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: demo.txt
URL:
<https://stat.ethz.ch/pipermail/r-help/attachments/20200329/c9433604/attachment.txt>

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: data1.txt
URL:
<https://stat.ethz.ch/pipermail/r-help/attachments/20200329/c9433604/attachment-0001.txt>

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: data2.txt
URL:
<https://stat.ethz.ch/pipermail/r-help/attachments/20200329/c9433604/attachment-0002.txt>

My general approach is to import data without letting the import convert to
factors. Then I explicitly create my factors with the levels I want them to
have, with explicit vectors specified for the levels. Fixing this after the
analysis is possible but a headache.

On March 28, 2020 5:46:25 PM PDT, Rolf Turner <r.turner at auckland.ac.nz>
wrote:
>
>I have an application in which I want to plot points with different 
>colours and symbols to distinguish (three) types of point.  I thought I
>
>had it figured out, mostly by a trial-and-error approach and by 
>simple-mindedly following instructions from others, without (sad to
>say) 
>really understanding what I was doing.
>
>I have now encountered an example where my syntax screws up, and I
>can't 
>see how to fix it in a coherent and generally applicable way.  The 
>problem arises if the data column (factor) specifying the point types
>is 
>missing one level.  When this happens, the types get out of synch.
>
>I have attached a source-able file "demo.txt" and the two data
sets
>upon 
>which it depends as "data1.txt" and "data2.txt" (dput()
format).
>
>The result from using the data from "data1.txt" (plotObj1)
demonstrates
>
>the problem, i.e. the unwanted phenomenon.  The result from using the 
>data from "data2.txt" (plotObj2) is "as it should be".
>
>In the data from data1.txt the "extremes" column, which specifies
the
>point type, is missing the level "0":
>
>>        table(Dat1$extremes)
>> 
>>        0 intermed        1 
>>        0       39       43 
>
>As you will be able to see, the plot of these data has the "0"
symbol
>(a 
>triangle) where there should be an "intermed" symbol (a solid
dot), and
>
>the "intermed" symbol where there should be a "1" symbol
(an
>upside-down 
>triangle).
>
>The data from data2.txt have an extremes column with all three levels 
>and the corresponding plot is as desired.
>
>I'm pretty sure that the solution to my problem is quite simple, but I 
>am too stupid to figure it out.  Can anyone guide me in the right
>direction?
>
>Thanks.
>
>cheers,
>
>Rolf Turner
-- 
Sent from my phone. Please excuse my brevity.