Improving Anchoring Vignette Methodology in Health Surveys with Image Vignettes.

Area of Science:

• Health Survey Methodology
• Psychometrics
• Cross-Cultural Research

Background:

• The anchoring vignette method is widely used in health surveys to enhance cross-group comparisons and adjust for differential item functioning (DIF).
• However, traditional text-based vignettes present challenges, including increased survey time, respondent burden, and frequent violations of underlying assumptions.

Purpose of the Study:

• To introduce and evaluate image anchoring vignettes as an innovative alternative to standard text vignettes.
• To assess the performance of image vignettes compared to text vignettes regarding response time, assumption satisfaction, and DIF-adjusting quality.

Main Methods:

• A cross-cultural experimental study was conducted comparing image and text vignettes.
• Confirmatory factor analysis was used to evaluate vignette performance, assumption satisfaction, and DIF-adjustment quality.

Main Results:

• Respondents demonstrated better differentiation of vignette intensity levels with image vignettes compared to text vignettes.
• The response consistency assumption was better satisfied under the image vignette condition.
• Image vignettes significantly reduced survey time without compromising DIF-adjustment quality.

Conclusions:

• Image anchoring vignettes offer a promising approach to improve the efficiency of the anchoring vignette method by reducing survey time.
• While image vignettes enhance respondent differentiation and assumption satisfaction, achieving vignette equivalence across diverse groups remains a challenge for both image and text formats.