UPMEAN = the mean or center of the person abilities = (computed) |
Assigns your chosen numerical value to the average of the non-extreme abilities for persons (UEXTREME=No) or all persons (UEXTREME=Yes), i.e., this provides a norm-referenced user-scaling.
The last UIMEAN= or UPMEAN= values in the control file is actioned.
UIMEAN= and UPMEAN= are ignored when there are anchor values, IAFILE= or PAFILE=. Anchor values are treated according to UASCALE=
Example 1: You want to used conventional IRT norm-referenced user-scaling with person mean of 0 and person S.D. of 1.
UPMEAN = 0
USCALE = 1 / (person S.D. in logits) ; find this from Table 3.1
If there are extreme person scores, see User-friendly rescaling
Example 2: I want to compare the mean performance of random samples of examinees from my database. Will UPMEAN= help in this?
UPMEAN=0 sets the local mean of the persons to zero (excluding extreme scores) regardless of the sampling.
If you wish to investigate the behavior of the person mean for different person samples, then
(1) analyze all persons and items: set UPMEAN=0, for convenience, and write IFILE=. For better comparison, set STBIAS=NO.
(2) anchor the items using IAFILE=
(3) analyze samples of persons with the anchored items.
The person means reported in Table 3.1 now show person means (with or without extreme scores) in the one frame of reference across all analyses defined by the anchored items.
For more examples, and how to compute this by hand, see User-friendly rescaling
Example 3: We want the mean of a target group in our analysis to be 50.
1. Put a code in the person label for the target group (or sort them to be adjacent rows) in the data file.
2. Do a free analysis of everything.
3. In the Specification menu, PSELECT= the target group on the code or PDELETE=+ the target range of rows.
4. Output Table 3.1 - this will give you the mean of the target group
5. Winsteps Specification menu: UPMEAN = - mean of target group
6. Output Table 3.1 - this will give you the mean of the target group which should be 0.
7. If not mean=0, then UPMEAN = - mean of target group from step 4, adjusted by value in step 6.
8. Repeat 5. and 6. if needed.
9. Write down the UPMEAN= value for later, if needed. You can see it in the Control Variables list.
10. Reinstate everyone: PSELECT= or PDELETE=
11. Run your reports.
Item Easiness = Reversed Item Difficulties
The reversal formula for the items on the maps in Tables 1.10, 1.11, 1.12, 1.13, 12.12, and 16.13 is:
Item Easiness = 2*Mean Person Ability - Item Difficulty
For a full set of item easiness statistics, set the Mean Person Ability as the reference point, then reverse the user-scaling . Use the USCALE= Calculator or:
For the person ability statistics:
UEXTREME = Yes
UPMEAN = 0 or preferred reference value
USCALE = current value of USCALE=
For the matching item easiness statistics:
UEXTREME = Yes
UPMEAN = 0 or preferred reference value
USCALE = - current value of USCALE=
Help for Winsteps Rasch Measurement and Rasch Analysis Software: www.winsteps.com. Author: John Michael Linacre
Facets Rasch measurement software.
Buy for $149. & site licenses.
Freeware student/evaluation Minifac download Winsteps Rasch measurement software. Buy for $149. & site licenses. Freeware student/evaluation Ministep download |
---|
Forum: | Rasch Measurement Forum to discuss any Rasch-related topic |
---|
Questions, Suggestions? Want to update Winsteps or Facets? Please email Mike Linacre, author of Winsteps mike@winsteps.com |
---|
State-of-the-art : single-user and site licenses : free student/evaluation versions : download immediately : instructional PDFs : user forum : assistance by email : bugs fixed fast : free update eligibility : backwards compatible : money back if not satisfied Rasch, Winsteps, Facets online Tutorials |
---|
Our current URL is www.winsteps.com
Winsteps® is a registered trademark