CRIME dataset
This dataset is used to motivate the commands from this website. It is copied from the R USArrests dataset (this can be accessed using the command data(USArrests)). More information can be found here.
It is derived from the 1975 World Almanac and Statistical Abstract of the United States, includes 50 observations at the state level on 4 variables (murder arrests per 100,000, assault arrest per 100,000, percent urban population and rape arrests per 100,000.
The datasets can be downloaded below:
- crime.txt (Text format, with header on first line)
- crime.csv (Comma separated)
- crime.dta (Stata format)
- crime.sas7bdat (SAS format)
Alternatively, this dataset can be read directly into R using the command:
ds <- read.csv("http://www.math.smith.edu/tutorial/crime.csv")
This can be read directly into Stata using the command:
use http://www.math.smith.edu/tutorial/crime.dta
To read it in SAS, run the command:
filename myurl url 'http://www.math.smith.edu/tutorial/crime.csv';
proc import datafile=myurl out=ds dbms=dlm;
delimiter=",";
getnames=yes;
run;
Using Stat/Transfer to create files in different formats
Stat/Transfer is a very powerful and simple-to-use utility to convert dataset from one format to another. It is installed on classroom computers at Smith, as well as on the Scinix system (as st).
A tutorial on Stat/Transfer is available from the CCPR website.