This is also for my work, so I don't have to look for syntax examples in my folders each time I try to do Rasch analysis.

1. Factor analysis/Correlation analysis just to see how items may line up.
2. Create a data file using SAS
3. Write a Winstep file (I pasted the data directly in this program, which I don't always do.)
4. Result data sets (just to show you how they look like)
   • person measures data (986 KB)
   • item difficulty measures data (2 KB)
5. Read output file, so we can get SAS data sets off the results.

• Table Pickers for WINSTEPS
  • Table Picker SAS syntax
  • Example output file (Table picker reads this into SAS)
  • RESULT (This text file is produced.)
  • Control file used to to produce an output file above.
• Table 8 pickers (for FACETS.  May work for Winsteps, too)

The meaning of variable names in person measure or item measure files:

• Entry: Entry number (sequence)
• Measure: The measure
• STTS: if 1, estimated in an authentic way (not imputed) DEFINITION OF STTS can be found on page 148 of http://www.winsteps.com/aftp/winsteps.pdf.  2=Anchored (fixed) calibiration; 1=Estimated calibration; 0=xtreme minimum, -1=Extreme Maximum, -2=No responses available for calibiration; -3=deleted by user.
• COUNT: the number of possible categories
• SCORE: the sum of actual values
• ERROR: standard error
• INMSQ: infit statistics, ideal value is 1
• INZSTD: standardized version of INMSQ,  red flag if the absolute value of this is larger than 2
• OUTMS: outfit statistics, ideal value is 1
• OUTZSTD: standardized version of OUTMS,  red flag if the absolute value of this is larger than 2
• DISPL: I have to check what this is
• PTME: Biserrial correlation, correlation between the measure and the measure minus an item in question
• WEIGHT: weight (ignore, it is all 1 in most application)
• G
• M
• NAME: IDs

• Online Manual
• Mike's online forum on Winsteps or Facets