library(plyr)
library(ggplot2)
library(RColorBrewer)
data2 <- data1[c(11, 88, 23)]
data3 <- count(data2)
# for correct ordering by median given limits on plot
data4 <- data3[data3$freq <= 110, ]
# table of number of dissertations per major, minimum of five
pop.maj1 <- sort(table(data4$Cnty_ID), decreasing = T)
pop.maj1 <- data.frame(num_crash = pop.maj1)
pop.maj1$Cnty_ID <- row.names(pop.maj1)
head(pop.maj1)
## num_crash Cnty_ID
## 101 424 101
## 220 416 220
## 71 415 71
## 178 414 178
## 227 406 227
## 161 401 161
# subset data by majors with at least five
# add number of dissertations to major column
to_plo1 <- data4
to_plo1 <- merge(to_plo1, pop.maj1, by = 'Cnty_ID', all.x = T)
to_plo1$Cnty_ID <- paste0(to_plo1$Cnty_ID, ' (', to_plo1$num_crash, ')')
head(to_plo1)
## Cnty_ID Prsn_Age Year freq num_crash
## 1 1 (56) 54 2011 1 56
## 2 1 (56) 45 2013 1 56
## 3 1 (56) 86 2014 1 56
## 4 1 (56) 51 2009 1 56
## 5 1 (56) 48 2013 1 56
## 6 1 (56) 57 2010 1 56
# median page lengthy by dissertation
med_pgs1 <- ddply(to_plo1,
.(Cnty_ID),
summarize,
median = median(Prsn_Age)
)
med_pgs1 <- med_pgs1[order(med_pgs1$median),]
# reorder major as factor by median page length
to_plo1$Cnty_ID <- factor(to_plo1$Cnty_ID, labels = med_pgs1$Cnty_ID,
levels = med_pgs1$Cnty_ID)
# merge with median pages for fill
to_plo1 <- merge(to_plo1, med_pgs1, by = 'Cnty_ID', all.x = T)
to_plo1 <- merge(to_plo1, pop.maj1, by = 'Cnty_ID', all.x = T)
head(to_plo1)
## Cnty_ID Prsn_Age Year freq num_crash.x median num_crash.y
## 1 108 (92) 34 2011 1 92 32 NA
## 2 108 (92) 21 2012 1 92 32 NA
## 3 108 (92) 16 2009 1 92 32 NA
## 4 108 (92) 44 2014 1 92 32 NA
## 5 108 (92) 25 2009 2 92 32 NA
## 6 108 (92) 57 2011 1 92 32 NA
######
# boxplots of pages by major, collored by median
diss_plo1 <- ggplot(to_plo1, aes(x = Cnty_ID, y = Prsn_Age, fill = median)) +
geom_boxplot(lwd = 0.3) +
ylim(0,105) +
coord_flip() +
theme_bw() +
scale_fill_gradientn(colours = brewer.pal(11, 'Spectral')) +
theme(legend.position = "none", axis.title.y = element_blank(),
axis.text.y = element_text(size = 16))
print(diss_plo1)
######
# barplots, number of dissertations by major
to_plo1 <- pop.maj1
to_plo1$Cnty_ID <- factor(pop.maj1$Cnty_ID, labels = pop.maj1$Cnty_ID,
levels = pop.maj1$Cnty_ID)
diss_plo <- ggplot(to_plo1, aes(x = Cnty_ID, y = num_crash, fill = num_crash)) +
geom_bar(lwd = 0.3, stat = 'identity') +
theme_bw() +
scale_fill_gradientn(colours = brewer.pal(11, 'Spectral')) +
ylab('Number of Crashes') +
theme(legend.position = "none", axis.title.x = element_blank(),
axis.text.x = element_text(size = 8, angle = 90, hjust = 1,
vjust = 0))
print(diss_plo)
###
require(devtools)
## Loading required package: devtools
source_gist('5281518')
## Sourcing https://gist.githubusercontent.com/fawda123/5281518/raw/29a1ab398df96658fdedd966870fb95351c55d20/plot_qual.r
## SHA-1 hash of file is 5003a8077c070fcd75fd92a56d04c1e241336f4b
par(mar=numeric(4),family='serif')
plot.qual(dat8,rs.ln=c(3,15))
## Loading required package: scales
par(mar=c(0,0,1,0),family='serif')
plot.qual(dat8,y.locs=c(0.05,1),ln.cl=c('lightblue','purple','green'),
main='County Wise Slopegraph for Texas')
