给定变量或数据列的R直方图图密度

Question

I have a data set with a column for age and a corresponding column with lung capacity. How can I create a histogram showing the distribution of lung capacity with respect to age?

Here is an example of what the data looks like. I actually want to compare the distributions for those who don't smoke with those who do:

Caes Age Gender Smoke Height FEV

0 16 1 0 64.8 2.65

0 12 0 0 60.5 2.27

1 19 1 0 71.7 4.29

0 15 0 0 64.8 2.52

Answer 1

Histograms are usually used when you have a single vector (like lung capacity) and you want to show the distribution of values:

library(ggplot2)
foo <- data.frame(age=runif(1000,min=10,max=50), capacity=rnorm(1000,mean=10))
ggplot(foo, aes(capacity))+geom_histogram(fill="blue")

If you want to plot the relationship between two variables, scatter plot might be a better choice:

ggplot(foo, aes(age, capacity))+geom_point(color="blue")

Answer 2

Thanks for the responses. I realized that I wanted a barplot rather than a histogram. Here is the solution that I came up with:

smoke=read.csv("SmokingEffect.csv",header=TRUE)
smokes=subset(smoke,select=c(Age,Smoke,FEV))
library(plyr)
smokesmeans <- ddply(smokes, c("Age","Smoke"), summarize, mean=mean(FEV),
sem=sd(FEV)/sqrt(length(FEV)))
smokesmeans <- transform(smokesmeans, lower=mean-sem, upper=mean+sem)
smokesmeans[,2] <- sapply(smokesmeans[,2], as.character)
library(ggplot2)
plotation <- qplot(x=Age, y=mean, fill=Smoke, data=smokesmeans, 
geom="bar",stat="identity",position="dodge",main="distribution of FEV",
ylab="mean FEV")
plotation <- plotation + geom_errorbar(aes(ymax=upper,
ymin=lower), position=position_dodge(0.9), data=smokesmeans)
png(myplot.png)
plotation
dev.off()

The output looks like this: