While this is extremely important work, the focus of that series of articles was on quantitative assessment of differential item functioning. The only "qualitative" part appears to be "generation of DIF hypotheses."
The article I cited when introducing this thread reported more extensive qualitative evaluation of the understanding of PROMIS items. A more recent example of the use of qualitative methods is: www.ncbi.nlm.nih.gov/pubmed/27599978
Patient. 2016 Sep 7. [Epub ahead of print]
Readability and Comprehension of the Geriatric Depression Scale and PROMIS® Physical Function Items in Older African Americans and Latinos.
Paz SH1, Jones L2,3, Calderón JL4, Hays RD4,5.
BACKGROUND:
Depression and physical function are particularly important health domains for the elderly. The Geriatric Depression Scale (GDS) and the Patient-Reported Outcomes Measurement Information System (PROMIS®) physical function item bank are two surveys commonly used to measure these domains. It is unclear if these two instruments adequately measure these aspects of health in minority elderly.
OBJECTIVE:
The aim of this study was to estimate the readability of the GDS and PROMIS® physical function items and to assess their comprehensibility using a sample of African American and Latino elderly.
METHODS:
Readability was estimated using the Flesch-Kincaid and Flesch Reading Ease (FRE) formulae for English versions, and a Spanish adaptation of the FRE formula for the Spanish versions. Comprehension of the GDS and PROMIS® items by minority elderly was evaluated with 30 cognitive interviews.
RESULTS:
Readability estimates of a number of items in English and Spanish of the GDS and PROMIS® physical functioning items exceed the U.S. recommended 5th-grade threshold for vulnerable populations, or were rated as 'fairly difficult', 'difficult', or 'very difficult' to read. Cognitive interviews revealed that many participants felt that more than the two (yes/no) GDS response options were needed to answer the questions. Wording of several PROMIS® items was considered confusing, and interpreting responses was problematic because they were based on using physical aids.
CONCLUSIONS:
Problems with item wording and response options of the GDS and PROMIS® physical function items may reduce reliability and validity of measurement when used with minority elderly.
