INTRODUCTION Negative urgency is a dimension of impulsiveness characterized by the tendency to act rashly under situations of negative emotionality. Previous research shows that negative urgency scores are associated with increased frequency of binge eating in clinical or community samples. Negative urgency has been assessed in the literature by one of the six subscales of the UPPS (Urgency, Premeditation, Perseverance, and Sensation Seeking scales), but this subscale is frequently used individually. Unfortunately, there is a dearth of research supporting the psychometric properties Negative urgency as a stand-alone subscale. Objective/Sample: To assess the validity evidence of the Negative Urgency subscale of the UPPS (UPPS-NU) in different disordered eating samples. METHODS Sample: 1) Community (n=341); 2) adolescents with obesity (n=210); 3) bariatric surgery patients (n=291); 4) eating disorders patients (anorexia and bulimia nervosa, n=143). Procedure: Participants responded to a set of self-report measures in the context of other parent studies including the UPPS-NU and measures assessing grazing (Rep(eat)-Q), disordered eating (TFEQ-21), eating disorder psychopathology (ED-15), loss of control eating (LOCES), depression (EADS[DASS]), emotion regulation (DERS). Statistics: Confirmatory factor analysis (CFA). Pearson correlation; Cronbach’s Alpha (?) and Omega (?). Statistics were conducted using R and SPSS. RESULTS Validity evidence based on the internal structure Dimensionality Dimensionality The original model (eleven items) showed poor fit (?2(54)=1613.637; CFI=0.966; NFI=0.965; TLI=0.959; RMSEA=0.168; P(rmsea≤0.05)<0.001; RMSEA 90% CI]0.161;0.175[; SRMR=0.107). The factor loadings and modification indices were investigated. Three items were removed: item 11 presented low factor loading (?item11=0.085), and items 2 and 5 due to having several high modification indices values. The reduced model (Figure 1) presented acceptable fit to the data (?2(26)=285.568; CFI=0.993; NFI=0.992; TLI=0.99; RMSEA=0.099; P(rmsea≤0.05)<0.001; RMSEA 90% CI ]0.089;0.109[; SRMR=0.061). Reliability The reliability of the scores in terms of internal consistency presented good values (?=0.916;?=0.916) Validity evidence based on the relation to other variables The validity evidence based on the relation to other (external) variables revealed promising findings (Figure 2; Figure 3; Figure 4; Figure 5). The magnitudes and directions of the bivariate associations between the Negative urgency and other variables occurred as expected. The findings suggest good convergent evidence, strengthening the nomological network. DISCUSSION AND CONCLUSION The results of this investigation suggest that caution should be made when using the original 11 items UPPS negative urgency subscale. A reduced scale with 9 items presented good factor structure and external validity evidence for different populations with disordered eating.

--