ibutton.analysis.R

## this file demonstrates how to use a collection of functions to load and
## analyze ibutton temperature data
## written by 
## Andrew MacDonald, Feb 2012


# load in required packages -----------------------------------------------

library(reshape)
library(lattice)

# load my functions -------------------------------------------------------

source('ibutton.functions.R')

# analysis ----------------------------------------------------------------

ibutton.data <- read.ibutton.folder("example.data/")

## which ones failed? you could just check the number of rows in each, and pick 
## the ones which are suspiciously short, indicating that the ibutton stopped 
## recording temperature.  This function automates this process to make it a bit
## more objective: it points out ones which recorded less than the median number
## of datapoints for the experiment.  It assumes that all the ibuttons were
## supposed to run for equal amounts of time
id.broken(ibutton.data)

## get the lengths for your own use:
sapply(ibutton.data,nrow)

## sometimes, fieldworkers record the data incorrectly -- for example, one 
## common mistake is to save data from the same ibutton twice with different 
## filenames.  However, each ibutton has a unique registration number.  check 
## for any number >1 in this table to identify this error. Additionally, if you
## recorded the registration numbers (written on the back of the ibuttons) you
## could use this to get them from the datafiles themselves

table(get.registration.numbers("example.data/"))

## correct the dates and make dataframe 
## Now that the data is checked, this function takes the list of ibuttons and
## combines them together.  It also reformats the "Date.Time" variable, into a
## format that R recognizes as a date and time.
ibutton.dataframe <- ibuttons.to.data.frame(ibutton.data)
head(ibutton.dataframe)
summary(ibutton.dataframe)

## Here are some ways to graph this data:
with(subset(ibutton.dataframe,ibutton=="10a"),plot(Date.Time,Value,type="n"))
for(i in levels(ibutton.dataframe$ibutton)){
  x <- which(ibutton.dataframe$ibutton==i)
  lines(ibutton.dataframe$Date.Time[x],ibutton.dataframe$Value[x])
}