With the estimated number of people diagnosed with Major Depressive
Disorder (MDD) reaching 264 million by 2017 (ADAA, 2020), it is essential
for clinicians to accurately diagnose MDD. The most commonly used scale in
MDD diagnostics is Beck Depression Inventory Second Edition (BDI-II), which
adheres to the definition from Diagnostic and Statistical Manual of Mental
Disorders, 5th Edition (DSM-5). However, some specific questions in BDI-II
are controversial in accurately characterizing MDD. To assess the
questionnaire, statistical analysis on BDI-II was conducted using the data
set collected from the Dominican Republic by Garcia-Batista et al. (2018),
using classical test theory, factor analysis, and item response theory. Of
21 questions, seven were identified as detrimental to the statistical
properties of the scale, and were removed to create a final scale. This
scale increased reliability of the parts by 0.043, average factor loading
by 0.05, and unidimensionality by 0.036, indicating inefficiency in the
prior questions. This result is consistent with the study by Zimmerman et
al. (2006), which provides critical evidence to support the need for an
accurate diagnosis of MDD.