I'm trying to do some percentage based comparisons across different groups in a criminal sentencing data set (http://dl.dropbox.com/u/1156404/wightCrimRecords.csv)
I have a data in an array of the form:
$Female x Burglary Criminal Damage and Arson 0.004950495 0.017326733 Driving Offences Murder 0.371287129 0.000000000 $Male x Burglary Criminal Damage and Arson 0.013001083 0.058504875 Driving Offences Murder 0.303358613 0.000000000 $`Not Stated` x Burglary Criminal Damage and Arson 0.0000000 0.0000000 Driving Offences Murder 0.1111111 0.0000000
This was derived from the original data as follows:
What I would like to do is generate a single data frame that contains a gender column, a frequency column, and rows corresponding to Burglary, Murder etc.
I can extract a single datatable from the multidimensional array, eg using:
which generates a separate row for each offence and columns referring to offence type and frequency, but can't see how to generate a single datatable.
PS I was also wondering whether it's possible to pull out even more structured data, that for example counts the percentages of offence type sex and age group, so for example I could lookup up what percentage of convictions for males in the 35+ age range are related to murder.