XFILE= analyzed response file |
If XFILE=filename is specified in the control file, a file is output which enables a detailed analysis of individual response anomalies.
XFILE=? opens a Browse window
This file contains 4 heading lines (unless HLINES=N or ROW1HEADING=N) followed by one line for each person-by-item response used in the estimation. Each line contains:
Field number |
Description |
Number Format |
Abbreviation |
1 |
Person number |
7.0 |
PERSON |
2 |
Item number |
7.0 |
ITEM |
3 |
4.0 |
OBS |
|
4 |
Response value (after scoring and recounting. This usually only happens for rating scales with unobserved intermediate categories and STKEEP=NO.) |
4.0 |
ORD |
5 |
Expected response value. For dichotomous items, probability of success. This is computed from the measures without correction by STBIAS=. If you want to see the values with STBIAS=YES, then: 1. Perform the analysis with STBIAS=YES 2. Output IFILE=if.txt PFILE=PF.txt SFILE=sf.txt 3. Perform the analysis again with STBIAS=NO IAFILE=if.txt PAFILE=pf.txt SAFILE=sf.txt |
7.3 |
EXPECT |
6 |
Modeled Variance of observed values around the expected value This is also the statistical information in the observation. √ (modeled variance) is the observation's raw-score standard deviation or S.E. |
7.3 |
VARIAN |
7 |
Standardized residual: (Observed - Expected)/Square root (Variance). This approximates a unit-normal deviate. Values outside ±2 are unexpected. |
7.3 |
ZSCORE |
8 |
Score residual: (Observed - Expected) |
7.3 |
RESIDL |
9 |
Person measure in USCALE= units |
7.2* |
PERMEA |
10 |
Item measure in USCALE= units |
7.2* |
ITMMEA |
11 |
Measure difference (Person measure - Item measure) in USCALE= units |
7.2* |
MEASDF |
12 |
Log-Probability of observed response. These can be summed for the Log-Likelihood Chi-Square, Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC), etc. |
7.3 |
L-PROB |
13 |
Predicted person measure from this response alone in USCALE= units. for many purposes, this approximates a linearization of the individual original ordinal observation, but, because the top and bottom categories of a rating scale are ininitely wide, averaging these only approximates the person measure estimated from all the person's observations. |
7.2* |
PPMEAS |
14 |
Predicted item measure from this response alone in USCALE= units |
7.2* |
PIMEAS |
15 |
Modeled Kurtosis of the observed values around their expectation |
7.3 |
KURT |
16 |
More Probable of the Responses to be observed |
4.0 |
MPR |
17 |
Response weight (person weight x item weight) |
7.3 |
WEIGHT |
18 |
Response status: 0 = Not scored, 1 = Standard, 2 = in Extreme person score, 3 = in Extreme item score, 4 = in Extreme person and item scores |
2.0 |
ST |
19 |
Response code in data file |
A |
CODE |
20 |
Person label |
A |
PLABEL |
21 |
Item label |
A |
ILABEL |
7.3 means "7 columns with 3 decimal places". 2* means Decimal Places set by the XFILE= dialog box from the Output Files menu or UDECIM= if not set.. A means alphanumeric character field |
Fields can be selected interactively and the default field selection changed at XFILE= dialog box.
If CSV=Y, the values are separated by commas. When CSV=T, the commas are replaced by tab characters. For "non-numeric values in quotation marks", specify QUOTED=Y.
This file enables a detailed analysis of individual response anomalies. The response residual can be analyzed in three forms:
1) in response-level score units, from [(observed value - expected value)].
2) in logits, from [(observed value - expected value)/variance].
3) in standard units, [(observed value - expected value)/(square root of variance)].
Predicted person measure: Imagine that this observation was the only observation made for the person ... this value is the measure we would predict for that person given the item measure.
Predicted item measure: Imagine that this observation is the only observation made for this item ... this value is the measure we would predict for that item given the person measure.
The formulas are the same as for a response string of more than 1 observation. For dichotomies, see www.rasch.org/rmt/rmt102t.htm and for polytomies www.rasch.org/rmt/rmt122q.htm
Example 1: I need the the S.D. of the response values for each item
Output the Xfile= to Excel. Sort by Item number. Then compute the S.D. for each item separately or use the Excel SUBTOTAL function:
1.On the Data tab, in the Outline group, click Subtotal. The Subtotal dialog box is displayed.
2.In the At each change in box, click the nested subtotal column, item number. ...
3.In the Use function box, click the summary function that you want to use to calculate the subtotals, StDevP. ...
4.Clear the Replace current subtotals check box.
5.Click OK
Example 2: You wish to compute differential item functioning, DIF, for a specific classification group of people:
If Table 30 is not suitable, here is a simple approximation:
Since one item does not have enough information to measure a person, for item bias we have to do it on the basis of a classification group of people.
From the XFILE,
add the "score residuals" (not standardized) for everyone in classification "A" on a particular item.
Add the "modeled variance" for everyone in the classification.
Divide the residual sum by the variance sum. This gives an estimate of the DIF for classification "A" relative to the grand mean measure.
Do the same for classification "B" on the same item.
To contrast classification "A" with classification "B" then
DIF size "AB" =DIF estimate for "A" - DIF estimate for "B"
A significance t-test is t =DIF size "AB" / square root ( 1/variance sum for classification A + 1/variance sum for classification B))
Example 3: You want to convert lucky guesses into missing data for some items.
Click on Output Files menu
Click on XFILE=
In the XFILE= Fields dialog box, type the Item numbers you want.
Click on OK
Output to Excel (Temporary file)
Sort the Excel file on Residual
The largest positive residuals are the lucky guesses
Rectangular- Copy the person entry numbers and the item entry numbers into your Winsteps control file:
person item . ; the . is to indicate "make this response missing data"
person item . ; the . is to indicate "make this response missing data"
person item . ; the . is to indicate "make this response missing data"
*
Save your Winsteps control file.
In the next analysis these unexpected correct responses will be scored as missing. The raw scores will also change.
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