teacher.pay_scan("teacher.pay.txt")
hist(teacher.pay)
stem(teacher.pay)
plot(sort(teacher.pay))

teacher.pay
teacher.pay[50]			# outlier

mean(teacher.pay)		# mean
mean(teacher.pay[-50])		# mean without outlier

sort(teacher.pay)[26]		# median
median(teacher.pay)
median(teacher.pay[-50])

sort(teacher.pay)		# find mode?

range(teacher.pay)		# gives min and max
diff(range(teacher.pay))	# computer range
summary(teacher.pay)		# to calculate IQR -- also 5 number summary
var(teacher.pay)		# variance
sqrt(var(teacher.pay))		# standard deviation

hist(teacher.pay)		# symmetric or skewed?
hist(teacher.pay,breaks=12)

boxplot(teacher.pay)		# box and whisker diagram

sort(teacher.pay)		# explore empirical rule
mean(teacher.pay)-sqrt(var(teacher.pay))     # not symmetric or bell-shaped
mean(teacher.pay)+sqrt(var(teacher.pay))     # 80% within 1SD
mean(teacher.pay)-2*sqrt(var(teacher.pay))
mean(teacher.pay)+2*sqrt(var(teacher.pay))   # 96% within 2SD

attach(cereal)
summary(calories)
summary(sugars)

plot(sugars[-58],calories[-58])
cor(sugars[-58],calories[-58])

summary(cereal)
carbo
sugars
potass
cereal_cereal[-c(5,21,58),]
attach(cereal)
potass
cor(cereal[,4:16])