# input data into R

readincamp = read.csv(“competitivereadingcamp.csv”)

# Tell R to assume readincamp is dataset from now until detach

attach(readincamp )

# Summarize data with summarySE command

# Source; the excellent http://www.cookbook-r.com/Graphs/Plotting_means_and_error_bars_%28ggplot2%29/

# To do this have to install.packages(“bear”)

# summarySE gives count, mean, standard deviation, standard error of the mean, and confidence interval (default 95%).

# rc2 will be a new data frame.

# measurevar: the name of a column that contains the variable to be summariezed

# groupvars: a vector containing names of columns that contain grouping variables

# na.rm: a boolean option that indicates whether to ignore NA’s – missing values

# conf.interval: the percent range of the confidence interval (default is 95%)

install.packages(“bear”)

library(bear)

rc2 <- summarySE(readincamp, measurevar=”score1″, groupvars=c(“treatment”, “female”))

# In the new data frame rc2, make treatment and female into factor rather than numeric variables

rc$treatment2 <- factor(rc$treatment)

rc2$female2 <- factor(rc$female)

# Now use ggplot to make the bar plot

# need install.packages(“ggplot2”) and then library(“ggplot2″)

# Error bars represent standard error (se) or confidence interval (ci) of the mean

ggplot(rc2, aes(x=treatment2, y=score1, fill=female2)) +

geom_bar(position=position_dodge(), stat=”identity”) + # Thinner lines

geom_errorbar(aes(ymin=score1-ci, ymax=score1+ci),

size=.3, # Thinner lines

width=.2,

position=position_dodge(.9)) +

xlab(“Treatment”) +

ylab(“Score 1″) +

scale_fill_hue(name=”Gender”, # Legend label

breaks=c(“0”, “1”),

labels=c(“Male”, “Female”)) +

ggtitle(“The Effect of Treatment on Test Score1 (with confidence intervals)”) +

scale_y_continuous(breaks=0:20*2) + #control ticks on y axis

theme_bw() #make background white

# make a boxplot of the means

# Here is some great help http://www.r-bloggers.com/box-plot-with-r-tutorial/

boxplot(score1 ~ female*competitive, main=”Scores on reading test”,

xlab=””, ylab=”Score 1″,

col=(c(“white”,”gray”)), las = 2,

at =c(1,2,4,5),par(mar = c(8, 5, 4, 2)+ 0.1),

names = c(“men, noncomp”,”men, comp”,”women, noncom”,”women, comp”))

### Like this:

Like Loading...

##
About mkevane

Economist at Santa Clara University and Director of Friends of African Village Libraries.