Illinois Goal Assessment Program

1
Illinois Goal Assessment Program
2
Ugly Data

• Http//research.umbc.edu/coates/ec611/612_sample.
  txt
• , decimals, dashes, semicolons
• 709 variables, 3835 observations
• Try reading the data into excel
• How many variables?
• Open wordpad, paste and save as a txt file

3
Stata data dictionary

• Http//research.umbc.edu/coates/ec611/612_sample.
  dct
• dictionary using raw data file name
• _column() tells Stata where the variable begins
• storage (memory management)
• variable name
• format and length
• label

4
Stata data dictionary

• Infile using dictionary file name
• infile using 612_sample.dct
• recall that dictionary calls the raw data set

5
IGAP Layout

• Columns
• column numbers
• number of columns
• variable description

6
IGAP Layout

• variable format
• alphanumeric (strings)
• fortran (total columns, how many after decimal)
• comma (total columns, comma separates thousands)
• (dollar sign, comma separates thousands)

7
Dictionary again

• Handout - find school enrollment
• comma5, sixth column a
• example 2,345
• example 286
• Look at the dictionary
• What did I do to get the variable?

8
IGAP enrollment variable

• Senrollt, senrollo, denrollo, denrollt
• gen senrollsenrollt1000 senrollo
• What happens?
• Replace senrollt0 if senrollt.
• gen senrollsenrollt1000 senrollo
• What happens?
• Drop senroll then try again
• Create denroll

9
IGAP dictionary

• Write a dictionary to read the data set RC97
• Get full record layout from computer lab
• Download rc97.zip
• http//research.umbc.edu/coates/ec611/rc97.zip