An adequate assessment of depression has been a concern to many researches over the last half-century. These efforts have brought forth a manifold of depression rating scales, of which one of the most commonly used self-assessment scales is the Beck Depression Inventory (BDI). Since its revision by Beck, Steer, and Brown (1996), the item structure of the BDI-II has been examined in many factor analyses, both, exploratory and confirmatory. Yet, a consensus about the underlying factor structure could not be achieved.

The present study was conducted to reexamine the item structure based on robust Nonmetric Multidimensional Scaling (NMDS). The NMDS solution allows for evaluating the diverging results from the factor analyses conducted so far.

The norm sample of the German version of the BDI-II was used (Hautzinger, Keller, & Kühner, 2006), which consisted of N = 266 patients diagnosed with a major depressive disorder.

The NMDS solution emphasized a facet oriented structure and revealed the insufficiency of the factor analyses to adequately capture the data.

Based on the NMDS solution, we propose an item structure consisting of a depressive core syndrome, surrounded by five specific facets (Figure 1). The core syndrome measures the severity of the depression mainly, whereas the five facets pronounce different qualities inherent to depression.

[Figure 1]