We are developing a scale with Likert items. There is an item which loaded very well on factor 1 in PCA. After determining the factor structure, I thought of observing the gender DIF patterns and found through ordinal regression that this item has significant DIF when grouped by gender. What shall I do now? Shall I remove this item from the scale and re-run the PCA or is it okay to retain an item with DIF if it is reckoned to be important for the construct theoretically?
HiMr. Chandu;
Differential item functioning is just a statistical indicator about item bias. DIF detected item have not always biased. An expert must check the item. If expert decides and explain that item has not bias between genders, you can always use item.
Good luck.
An item with significant DIF in favor or against a focal or reference group is an indication of consistent error in measurement. This means that there is an extraneous factor systematically favoring one group of examinees over the other, other than what the item was designed to measure.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression