Previous research has provided insufficient evidence on the factorial validity of the negative cognitive style questionnaires, which is a problem for the validity of the total score's computation. In Study 1, we focused on the relationship between internality and the other dimensions of negative cognitive style. In Study 2, we explored the predictive validity of negative cognitive style for negative interpretation bias. In Study 1, 770 participants completed the Cognitive Style Questionnaire - Short Form (CSQ-SF). In Study 2, from a prescreening data collection (N = 300) we selected participants with low (N = 40) and high (N = 32) cognitive vulnerability to depression who were primed with negative mood induction and who completed a generative interpretation task. In Study 1, a confirmatory factor analysis (CFA) indicated that the best fitting model for the CSQ-SF was a bifactor model without the internality dimension. In Study 2, a CFA replicated the factorial structure of Study 1 and individuals with a high negative cognitive style exhibited a negative interpretation bias after controlling for depressive symptoms. The university-age sample limited the generalizability of our results to different populations, and the lack of longitudinal data prevented us from discussing further implications on the relationship between the negative interpretation bias and negative cognitive style in predicting depression. Together, the results of our two studies support the construct validity of the CSQ-SF and recommend the use of a composite score of negative cognitive style without internality. Copyright © 2019 Elsevier Ltd. All rights reserved.