## clear the workspace
rm(list=ls())

## --------------------------------------------------
## Analyze vector of data: flying snakes

## 1.
myvector <- c(0.9, 1.4, 1.2, 1.2, 1.3, 2.0, 1.4, 1.6)

## 2.
hist(myvector, col='red')
hist(myvector, col='red', right = FALSE)

## 3.
x <- myvector*2*pi

## 4.
sum(x)/length(x) ## this calculates the mean
## let's wrap this into a function
my.mean <- function(vec) {
  mm <- sum(vec)/length(vec)
  return(mm)
}
x.mean <- my.mean(x)
x.mean
## and I can also apply this function to the original data
my.mean(myvector)

## 5.
mean(x)

## 6.
##
## my own variance function
my.var <- function(vec) {
  vv <- sum((x-x.mean)^2)/(length(x)-1)
  return(vv)
}
x.var <- my.var(vec)
x.var
## my own sd function
my.sd <- function(vec) {
  vv <- my.var(vec)
  return(sqrt(vv))
}
x.sd <- my.sd(vec)
x.sd

## 7.
var(x)
sd(x)

## 8.
median(x)

## 9.
sd(x)/sqrt(length(x))


## --------------------------------------------------
## Anolis lizards in a data frame

## 4.
mydata <- read.csv('https://www.sfu.ca/~lmgonigl/materials-qm/data/anolis.csv')

## 5.
str(mydata)
class(mydata$Island)
class(mydata)

## 6.
head(mydata)
head(mydata, 20)
tail(mydata)

## 7.
mydata <- read.csv('https://www.sfu.ca/~lmgonigl/materials-qm/data/anolis.csv',
                   stringsAsFactors=TRUE)
class(mydata$Island)

## 8.
##
## Note that, as an alternative to the $ sign, you can access columns
## by entering their name, in quotes in the 'column' position of the
## data.frame, as I do below.  I prefer this format to the $-symbol,
## but both are ok, and I use them interchangeably.
levels(mydata[,'Ecomorph'])

## 9.
table(mydata[,'Ecomorph'])
which(mydata[,'Ecomorph']=='Trunk-Crown ')

## 10.
mydata[mydata[,'Ecomorph']=='Trunk-Crown ','Ecomorph'] <- 'Trunk-Crown'
table(mydata[,'Ecomorph'])

## 11.
mydata[,'Ecomorph'] <- droplevels(mydata[,'Ecomorph'])
levels(mydata[,'Ecomorph'])

## 12.
anolis <- read.csv('https://www.sfu.ca/~lmgonigl/materials-qm/data/anolis.csv',
                   stringsAsFactors=TRUE,
                   na.strings=c('', 'NA'),
                   strip.white=TRUE)

## 13.
table(anolis[,'Ecomorph'])

## 14.
table(anolis[,'Ecomorph'], useNA = 'ifany')

## 15.
table(anolis[,'Island'])

## 16.
length(grep('Cuba', anolis[,'Island']))

## 17.
big.4 <- c('Cuba', 'Hispaniola', 'Jamaica', 'Puerto Rico')
## write a function to get the ecomorphs from one island
get.for.one.island <- function(s) {
  table(anolis[grep(s, anolis[,'Island']), 'Ecomorph'])
}
## test it
get.for.one.island('Cuba')
get.for.one.island('Hispaniola')
## do all at once
big.4.tab <- sapply(big.4, get.for.one.island)
big.4.tab
## transpose the table (switch rows and columns) for a more natural
## presentation
t(big.4.tab)
## this works because no single species appears on multiple big islands

## 18.
## there is lots of syntax we haven't covered in here... we'll learn
## these as we go
tab.not.big.4 <- table(anolis[!(anolis[,'Island'] %in% big.4),'Ecomorph'])
names(tab.not.big.4)[which.max(tab.not.big.4)]