# 1
pp<-read.csv('presidencies.csv')
pp$name[pp$state=='Virginia']

# 2 
pp$name[pp$order==33]

# 3
pp$start[pp$state=='Ohio']

# 4
pp$name[duplicated(pp$name)]

# 4, another solution
tt<-table(pp$name)
names(tt)[tt>1]

# 5
pp$name[pp$party=='Democratic'&pp$state=='New York']

# 6
pp$name[pp$party=='Republican'&pp$state=='Ohio']

# 7 
pp$name[pp$state=='Ohio'|pp$state=='Texas']

# 8
ii<-read.csv('nhanes.csv')
mean(subset(ii$bmxbmi,ii$hs=='History of smoking'))
mean(subset(ii$bmxbmi,ii$hs=='No'))

# 8, another solution
mean(ii$bmxbmi[ii$hs=='History of smoking'&!is.na(ii$hs)])
mean(ii$bmxbmi[ii$hs=='No'&!is.na(ii$hs)])

# 8, another solution (please see the pdf posted on the cle for further
# discussion)
mean(ii$bmxbmi[ii$hs=='History of smoking'],na.rm=TRUE)
mean(ii$bmxbmi[ii$hs=='No'],na.rm=TRUE)

# 9 
t.test(bmxbmi~military,data=subset(ii,ridageyr<=30))

# 10
sum(is.na(ii$hs)|is.na(ii$military)|is.na(ii$asthma))