Plots a week's worth of browsing data by age group

gistfile1.r

pdf(file="testing.pdf", width=11, height=4.5, family="Helvetica")

# read in table and remove rows with NAs
agedata <- na.omit(read.table("sessions_by_age_10min.txt", sep="|", header=T))

color <- c("#0000FF", "#FF1D25", "#009103", "#F7941E", "#00A8E2", "#FFA4D6")

# (probably an easier way to do this)
agedata$color[agedata$q6==0] <- color[1]
agedata$color[agedata$q6==1] <- color[2]
agedata$color[agedata$q6==2] <- color[3]
agedata$color[agedata$q6==3] <- color[4]
agedata$color[agedata$q6==4] <- color[5]
agedata$color[agedata$q6==5] <- color[6]

ages <- c("<18", "18-25", "26-35", "36-45", "46-55", "55+")

survey = read.table("survey.csv", sep=",", header=T)

plot(agedata$stimestamp, agedata$total, type="n", ylim=c(0, .4), ylab="", xaxt="n") # for scales

# add a series for each age group
for(i in unique(agedata$q6)) {
  t = agedata[agedata$q6==i,]
  x = length(which(survey$q6==i))
  print(x)
  t$freq <- t$total/x
  # t$freq <- t$total
  # points(t$stimestamp, t$freq, col=t$color, type="p", pch=19, cex=.12)
  lines(lowess(t$stimestamp, t$freq, f=.04), lwd=2, col=t$color)
}
axis(1, at=c(1288756800, 1289361600)) # x axis
legend("topleft", legend=ages, lwd=1.5, col=color, cex=.7)

dev.off()