R help - Oct 2010 - plot CI and mortality rate

Dear R Users:

       I have the individual mortality rate and 95% CI of 100 hospitals,
how  to do the  plot with the individual hospital in the Yaxis, and the
mortality rate and 95% CI in the Xais and a overall mean as a reference
line?

Thanks and regards,

Xin

	[[alternative HTML version deleted]]

XINLI LI <lihawaii <at> gmail.com> writes:
> 
> Dear R Users:
> 
>        I have the individual mortality rate and 95% CI of 100 hospitals,
> how  to do the  plot with the individual hospital in the Yaxis, and the
> mortality rate and 95% CI in the Xais and a overall mean as a reference
> line?
  Something like

library(plotrix)

plotCI(mortratevec,1:n,li=mort.lo,ui=mort.hi,err="x",axes=FALSE)
axis(side=1)
axis(side=2,at=1:n,label=hospitalnames,padj=0,las=1)
box()
abline(v=meanmort)

 
ggplot2 is another option: more complicated to get working,
but very pretty.  You'll want to look at
geom_point
geom_linerange
coord_flip
geom_hline

Hi:

Following up on Ben's suggestion re ggplot2, here's a manufactured
example:

# Fake data:
mortrate <- round(runif(100), 3)
dd <- data.frame(rate = mortrate, moe = 1.96 * sqrt(mortrate * (1 -
mortrate))/10,
                  hosp = factor(paste('H', 1:100, sep = '')))
dim(dd)
[1] 100   3

# Set up the plot: ymin and ymax are input parameters to geom_pointrange
g <- ggplot(dd, aes(x = hosp, y = rate, ymin = rate - moe, ymax = rate +
moe))

# Version with hospitals in lexicographic order (rather useless):
g + geom_pointrange() +
    geom_hline(aes(yintercept = mean(rate))) +
    scale_x_discrete(expand = c(0.01, 0.01)) +
    coord_flip() + theme_bw() +
    opts(axis.text.y = theme_text(size = 5, hjust = 1))

# Version with hospitals sorted by rate (better):
g + geom_pointrange(aes(x = reorder(hosp, rate, mean))) +
    geom_hline(aes(yintercept = mean(rate))) +
    scale_x_discrete(expand = c(0.01, 0.01)) +
    coord_flip() + theme_bw() +
    opts(axis.text.y = theme_text(size = 5, hjust = 1))

Hope you like it...
Dennis

On Tue, Oct 19, 2010 at 4:59 PM, XINLI LI <lihawaii@gmail.com> wrote:
> Dear R Users:
>
>       I have the individual mortality rate and 95% CI of 100 hospitals,
> how  to do the  plot with the individual hospital in the Yaxis, and the
> mortality rate and 95% CI in the Xais and a overall mean as a reference
> line?
>
> Thanks and regards,
>
> Xin
>
>        [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@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.
>
	[[alternative HTML version deleted]]

You could also use lattice + Hmisc:

# Fake data (taken from Dennis Murphy's reply):
mortrate <- round(runif(100), 3)
moe = 1.96 * sqrt(mortrate * (1 - mortrate))/10
dd <- data.frame(rate = mortrate, CI.lower = mortrate - moe, CI.upper =
mortrate + moe, hosp = factor(paste('H', 1:100, sep = '')))
# using packages lattice and Hmisc:
require(lattice)
require(Hmisc)
dd$hosp.sort <- reorder(dd$hosp, dd$rate)
Dotplot(hosp.sort ~ Cbind(mortrate, CI.lower, CI.upper), data = dd, abline =
list(v = mean(mortrate), col = "red"))

HTH
Heinrich.
> -----Urspr?ngliche Nachricht-----
> Von: r-help-bounces at r-project.org
> [mailto:r-help-bounces at r-project.org] Im Auftrag von XINLI LI
> Gesendet: Mittwoch, 20. Oktober 2010 02:00
> An: R-help at r-project.org
> Betreff: [R] plot CI and mortality rate
>
>
> Dear R Users:
>
>        I have the individual mortality rate and 95% CI of 100
> hospitals,
> how  to do the  plot with the individual hospital in the
> Yaxis, and the
> mortality rate and 95% CI in the Xais and a overall mean as a
> reference
> line?
>
> Thanks and regards,
>
> Xin
>
>       [[alternative HTML version deleted]]
>
> ______________________________________________
> 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.
>