DIF= specifies the part of the person label which is to be used for classifying persons in order to identify Differential Item Function (DIF) - uniform or non-uniform - using the column selection rules. See also DIF Table and DIF and DPF considerations.
DIF= *file name
|
file containing details
|
DIF= *
|
in-line list
|
DIF= $S1W1
|
field in person label. DIF fields must be inside the person label defined by NAME1= and NAMELENGTH=
|
DIF= MA2
|
person ability levels for non-uniform DIF: MA2 splits at the middle of the value value.range
|
DIF= $S1W1 + MA2
|
person classes or groups
|
DIF= ?
|
open browser for file
|
DIF= location is usually column number within person label field. DIF=1 means "DIF selection character is first character of person label."
Example 1: Columns 18-20 of the person label (in columns 118-120 of the data record) contain a district code:
NAME1=101 ; person label starts in column 101
NAMELENGTH = 20 ; person label must include all the DIF codes
DIF = $S18W3 ; district starts in column 18 of person label with a width of 3
or
@district = 18W3 ; district starts in column 18 of person label with a width of 3
DIF = @district ; DIF classifier
tfile=*
30 Table 30 for the DIF report (or use Output Tables menu)
*
Example 2: DIF by Gender+Grade: Column 1 of the person labels contains gender (M or F) and Columns 7-8 contain grade level (K-12).
DIF = 1W1 + 7W2
Example 3: I have tab-separated data and my DIF indicator is in a separate field from the Person label.
Solution: for the DIF analysis, do a separate run of Winsteps. At the "Extra Specifications" prompt:
NAME1=(location of DIF indicator)
DIF=$S1W1
Example 4: Columns 18-20 of the person label (in columns 118-120 of the data record) contain a district code. Column 21 of the person label (in column 121 of the data record) has a gender code. Three independent DIF analyses are needed: district, gender, district+gender
NAME1=101 ; person label starts in column 101
NAMELENGTH = 21 ; person label must include all the DIF codes
DIF = *
$S18W3 ; start in person label column 18 with a width of 3 - district
$S21W1 ; start in person label column 21 with a width of 1 - gender
$S18W3+ $S21W1 ; DIF by district+gender
*
tfile=*
30 Table 30 for the DIF report (or use Output Tables menu)
*
Example 5: An investigation of non-uniform DIF with high-low ability classification for the KCT data.
; action the following with the Specification pull-down menu
@SEX = $S9W1 ; the sex of participants is in column 9 of the person label
DIF = @SEX + MA2 ; look for non-uniform DIF (gender + two ability strata): MA2
PSUBTOT = @SEX + MA2 ; summary statistics by gender and ability strata
Tfile=* ; This is more easily actioned through the Output Tables Menu
30 ; Table 30 - DIF report
28 ; Table 28 - Person subtotals for DIF classifications
*
Table 30: DIF specification is: DIF=@SEX+MA2
+-----------------------------------------------------------------------------------------+
| KID DIF DIF KID DIF DIF DIF JOINT TAP |
| CLASS MEASURE S.E. CLASS MEASURE S.E. CONTRAST S.E. t d.f. Number Name |
|-----------------------------------------------------------------------------------------|
| F1 -1.86 1.09 M1 -4.54 1.15 2.68 1.59 1.69 8 6 6= 3-4-1 |
Table 28: Subtotal specification is: PSUBTOTAL=@SEX+MA2
+---------------------------------------------------------+
| KID MEAN S.E. OBSERVED MEDIAN REAL |
| COUNT MEASURE MEAN P.SD SEPARATION CODE |
|---------------------------------------------------------|
| 4 -2.08 .90 .89 .00 F1 | <- Non-extreme
| 6 -2.82 .41 .91 -2.86 .32 M1 |
Example 6: With Example0.txt (the Liking for Science rating scale data) you want to see if any items were biased against names starting with any letter of the alphabet, then:
run example0.txt
request the DIF Table (Table 30) from the Output Tables menu
specify: $S1W1
a DIF table is produced.
The equivalent DIF specification is: DIF=$S1W1
Positive DIF size is higher ACT difficulty measure
+------------------------------------------------------------------------------------+
| KID DIF DIF KID DIF DIF DIF JOINT ACT |
| CLASS MEASURE S.E. CLASS MEASURE S.E. CONTRAST S.E. t d.f. Number Name |
|------------------------------------------------------------------------------------|
| R -.06 .54 W .89> 2.05 -.95 2.12 -.45 8 1 WATCH BIRDS |
| R -.06 .54 L -.65 .75 .59 .92 .64 12 1 WATCH BIRDS |
| R -.06 .54 S -.42 .57 .36 .78 .46 18 1 WATCH BIRDS |
| R -.06 .54 H -1.63 1.13 1.57 1.25 1.26 11 1 WATCH BIRDS |
| R -.06 .54 D .12 .86 -.18 1.01 -.18 11 1 WATCH BIRDS |
.......
Example 7: The DIF demographic column is not part of the person label. Use the FORMAT= instruction to rearrange the record.
Was:
0 1 2 3
123456789012345678901234567890
AxxxxxxxxxxxxxxxxxxxBCyyyyyyyy ; original record layout
ITEM1=2
NI=19 ; the x's
NAME1=21 ; starts with B
To move column 1 after column 21 with XWIDE=1:
xxxxxxxxxxxxxxxxxxxBACyyyyyyyy ; after internal reformatting
FORMAT = (T2,20A,T1,1A,T22,100A)
ITEM1 = 1 ; references position in the reformatted record.
NI = 19 ; the x's
NAME1 = 20 ; starts with B - references position in the reformatted record.
DIF = S2W1 ; code A
The first reformatted record is shown on your screen so you can check that it is how you want it.
Example 8: We only want DIF between person classification groups A and B, not the other person groups. Classification group code is in column 3 of the person label.
In the Specification pull down menu
PSELECT=??{AB}
Output Table menu: Table 30.
