R help - Aug 2009 - R graphics

Dear sir,

I am trying to create a barplot for two propotions, with proportions successful
on the y and age on the x.  So my data is in the formate of one vector
containing 0s and 1s (i.e.successful or not) with a corresponding two level
vector of age.

success        age
1                     old
0                    young
0                     young
1                     old

Howeber i cannot find a way to stop r stacking the data using the code: 
plot(successfulYN~maleage) as it ends up showing both proportions as i only want
the proportions of success to show.
also entering it as a table results in x axis being removed maybe though this is
the only way
could you please offer any help?

Grant McDoanld

On Thu, 13 Aug 2009, Mcdonald, Grant wrote:
> Dear sir,
>
> I am trying to create a barplot for two propotions, with proportions
successful on the y and age on the x.  So my data is in the formate of one
vector containing 0s and 1s (i.e.successful or not) with a corresponding two
level vector of age.
>
> success        age
> 1                     old
> 0                    young
> 0                     young
> 1                     old
>
> Howeber i cannot find a way to stop r stacking the data using the code: 
plot(successfulYN~maleage) as it ends up showing both proportions as i only want
the proportions of success to show.
> also entering it as a table results in x axis being removed maybe though
this is the only way
> could you please offer any help?
A reproducible example would have been helpful (as the posting guide 
explains).

Using the data from ?spineplot:

   treatment <- factor(rep(c(1, 2), c(43, 41)), levels = c(1, 2),
     labels = c("placebo", "treated"))
   improved <- factor(rep(c(1, 2, 3, 1, 2, 3), c(29, 7, 7, 13, 7, 21)),
     levels = c(1, 2, 3), labels = c("none", "some",
"marked"))

I think you have a plot which is similar to

   plot(treatment ~ improved)

showing P(treatment | improved) on the y-axis and P(improved) on the 
x-axis. And you want the same y-axis but an x-axis with equal widths on 
the x-axis, right? Then you can do

   ## store the underlying contingency table
   tab <- plot(treatment ~ improved)
   tab
   prop.table(tab, 1)

   ## plotting with P(improved) x-axis
   spineplot(tab)

   ## plotting with equi-sized bars
   spineplot(prop.table(tab, 1))

(Note that the order of the variables would typically be exchanged but 
this ordering makes the difference between the different plots more 
obvious.)

hth,
Z
> Grant McDoanld
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
>

________________________________________
From: Mcdonald, Grant
Sent: 13 August 2009 13:08
To: r-help at R-project.org
Subject: R graphics

Dear sir,

I am trying to create a barplot showing propotions, 

 So my data is in the format of a vector containing 0s and 1s (i.e.successful or
not) with a corresponding vector with two  catogorical levels.

success        age
1                     old
0                    young
0                     young
1                     old

I want to show a bar plot with a ylim=c(0,1) and an x-axis showing the two ages
and the proportion of success only as filled bars.

after making success a factor and plotting against age using: plot(success~age)

the graphic shows age on the x and proportion on the yaxis as i want.  However
for each age it shows the total proportion split into both success and failures.
I would however just like to show the prop successes without the failures stajed
on top.

data frame is below:

successful	maleage
1	O
1	Y
1	Y
1	Y
1	O
1	O
1	Y
1	O
1	Y
1	O
1	O
1	Y
1	O
1	Y
1	Y
1	Y
1	O
1	O
1	Y
1	Y
1	O
1	Y
1	O
1	Y
1	Y
1	Y
1	O
1	Y
1	Y
1	O
1	Y
1	O
1	O
1	Y
1	O
1	Y
1	O
1	Y
1	Y
1	O
1	Y
1	Y
1	O
1	Y
1	Y
1	Y
1	O
1	O
1	Y
1	O
1	Y
1	Y
1	Y
1	Y
1	Y
1	Y
1	Y
1	Y
1	Y
1	O
1	Y
1	O
1	Y
1	Y
1	O
1	O
1	Y
1	Y
1	O
1	Y
1	O
1	Y
1	Y
1	Y
1	O
1	Y
1	O
1	Y
1	Y
1	Y
1	O
1	Y
1	O
1	Y
1	Y
1	O
1	Y
1	Y
1	O
1	O
1	Y
1	Y
1	Y
1	O
0	Y
NA	O
0	O
NA	Y
1	Y
1	O
NA	O
NA	Y
0	O
NA	Y
0	O
NA	Y
1	Y
1	O
0	O
0	Y
NA	O
0	Y
NA	O
0	Y
0	Y
NA	O
NA	O
NA	O
NA	Y
NA	Y
0	O
NA	O
NA	Y
NA	O
1	O
0	O
0	Y
NA	O
NA	Y
1	O
0	O
0	O
0	O
NA	Y
0	O
0	Y
NA	Y
0	O
NA	O
0	Y
NA	O
1	O
0	Y
NA	O
NA	O
NA	O
0	O
0	O
0	O
0	Y
0	O
0	O
0	O
0	Y
NA	O
0	Y
0	O
NA	Y
NA	O
0	O
NA	Y
NA	O
0	O
NA	O
NA	O
0	O
NA	Y
NA	O
NA	Y
0	O
NA	O
1	Y
NA	Y
NA	O
1	O
1	Y
0	O
0	Y
0	O
0	Y
0	Y
1	Y

On Thu, 13 Aug 2009, Mcdonald, Grant wrote:
>
> ________________________________________
> From: Mcdonald, Grant
> Sent: 13 August 2009 13:08
> To: r-help at R-project.org
> Subject: R graphics
>
> Dear sir,
>
> I am trying to create a barplot showing propotions,
>
> So my data is in the format of a vector containing 0s and 1s
(i.e.successful or not) with a corresponding vector with two  catogorical
levels.
>
> success        age
> 1                     old
> 0                    young
> 0                     young
> 1                     old
>
> I want to show a bar plot with a ylim=c(0,1) and an x-axis showing the two
ages and the proportion of success only as filled bars.
>
> after making success a factor and plotting against age using:
plot(success~age)
>
> the graphic shows age on the x and proportion on the yaxis as i want. 
However for each age it shows the total proportion split into both success and
failures.  I would however just like to show the prop successes without the
failures stajed on top.
You can play around with the graphical parameters of ?spineplot. For 
example, I copied your data into a "dat.txt" file and read the data
via

   ## read data
   dat <- read.table("dat.txt", header = TRUE)

   ## transform labels
   dat$successful <- factor(dat$successful, levels = 1:0, labels     
c("success", "failure"))
   dat$maleage <- factor(dat$maleage, labels = c("old",
"young"))

And then you can produce various types of plots changing color, labeling 
etc.

   ## simple spine plot
   plot(successful ~ maleage, data = dat)

   ## hide failures and suppress y-axis labels
   plot(successful ~ maleage, data = dat, col = gray(c(0.3, 1)),
     border = "transparent", yaxlabels = "")

Or you can also use barplot() and create this by hand:

   ## create table
   tab <- xtabs(~ maleage + successful, data = dat)

   ## simple barplot of success proportions
   barplot(prop.table(tab, 1)[,1])

   ## barplot with proportional x-axis and labeling etc.
   barplot(prop.table(tab, 1)[,1], width = margin.table(tab, 1),
     col = gray(0.3), space = 0.1, ylim = c(0, 1),
     ylab = "proportion of succsess")

hth,
Z
> data frame is below:
>
> successful	maleage
> 1	O
> 1	Y
> 1	Y
> 1	Y
> 1	O
> 1	O
> 1	Y
> 1	O
> 1	Y
> 1	O
> 1	O
> 1	Y
> 1	O
> 1	Y
> 1	Y
> 1	Y
> 1	O
> 1	O
> 1	Y
> 1	Y
> 1	O
> 1	Y
> 1	O
> 1	Y
> 1	Y
> 1	Y
> 1	O
> 1	Y
> 1	Y
> 1	O
> 1	Y
> 1	O
> 1	O
> 1	Y
> 1	O
> 1	Y
> 1	O
> 1	Y
> 1	Y
> 1	O
> 1	Y
> 1	Y
> 1	O
> 1	Y
> 1	Y
> 1	Y
> 1	O
> 1	O
> 1	Y
> 1	O
> 1	Y
> 1	Y
> 1	Y
> 1	Y
> 1	Y
> 1	Y
> 1	Y
> 1	Y
> 1	Y
> 1	O
> 1	Y
> 1	O
> 1	Y
> 1	Y
> 1	O
> 1	O
> 1	Y
> 1	Y
> 1	O
> 1	Y
> 1	O
> 1	Y
> 1	Y
> 1	Y
> 1	O
> 1	Y
> 1	O
> 1	Y
> 1	Y
> 1	Y
> 1	O
> 1	Y
> 1	O
> 1	Y
> 1	Y
> 1	O
> 1	Y
> 1	Y
> 1	O
> 1	O
> 1	Y
> 1	Y
> 1	Y
> 1	O
> 0	Y
> NA	O
> 0	O
> NA	Y
> 1	Y
> 1	O
> NA	O
> NA	Y
> 0	O
> NA	Y
> 0	O
> NA	Y
> 1	Y
> 1	O
> 0	O
> 0	Y
> NA	O
> 0	Y
> NA	O
> 0	Y
> 0	Y
> NA	O
> NA	O
> NA	O
> NA	Y
> NA	Y
> 0	O
> NA	O
> NA	Y
> NA	O
> 1	O
> 0	O
> 0	Y
> NA	O
> NA	Y
> 1	O
> 0	O
> 0	O
> 0	O
> NA	Y
> 0	O
> 0	Y
> NA	Y
> 0	O
> NA	O
> 0	Y
> NA	O
> 1	O
> 0	Y
> NA	O
> NA	O
> NA	O
> 0	O
> 0	O
> 0	O
> 0	Y
> 0	O
> 0	O
> 0	O
> 0	Y
> NA	O
> 0	Y
> 0	O
> NA	Y
> NA	O
> 0	O
> NA	Y
> NA	O
> 0	O
> NA	O
> NA	O
> 0	O
> NA	Y
> NA	O
> NA	Y
> 0	O
> NA	O
> 1	Y
> NA	Y
> NA	O
> 1	O
> 1	Y
> 0	O
> 0	Y
> 0	O
> 0	Y
> 0	Y
> 1	Y
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
>