Today, we are going to talk about:

• Good data entry practices
• How to avoid common formatting mistakes
• Dates
• Basic quality control
• Exporting data from spreadsheets

We are going to follow these lessons developed by data carpentry: http://www.datacarpentry.org/spreadsheet-ecology-lesson/00-intro.html

Follow the link below.

bit.ly/1oVvND1

View raw

Extract the zip file

Benefits

• Needed for large datasets too big for excel e.g Sequencing data
• Can automate the quality control procedure.
• Reproducible.

Downsides

• Pre-formatting required
• Excel sometimes better for simple wrangling

You will learn more about these possibilities this afternoon!

  # read in the data file
  setwd("~/Biochem_R_training/data/")
  dat = read.csv("survey_data_spreadsheet_messy_fixed.csv",header=T)
  # Summarise some columns
  summary(dat$Sex)

##     F  M 
##  6 48 38

  summary(dat$Weight)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##    7.00   32.50   42.00   65.54  113.50  232.00      12

  # Hmmm let's see what plot does with weight, not that useful.
  plot(dat$Weight)

  # A histogram is what we want
  hist(dat$Weight, breaks=10)

  # Let's see the weight by species.
  # you can clearly see that DS is much heavier on average.
  plot(dat$Weight ~ dat$Species)

  # What about the relationship between plot and weight
  plot(dat$Weight ~ dat$Plot)

  # We can make R give us what we want.
  boxplot(dat$Weight ~ dat$Plot)