The main objective of this research was to investigate the psychometric properties of a Spanish-language version of the Attitudes toward Traffic Safety Scale (ATTS) for the assessment of risky driving attitudes among Spanish-speaking populations. Five hundred and fifty-eight drivers from Argentina participated in the study. Confirmatory factor analysis supported the ATTS three-factor structure: Attitude towards violations and speeding, attitude towards the careless driving of others and attitude towards drinking and driving, χ2(87) = 205.91, p < .001; χ2/df = 2.36; GFI = .94; CFI = .93; TLI = .91; RMSEA = .05; 90% CI [.04, .06]. A model with one higher-order factor (overall attitude toward risky driving) also fits the data. Reliability estimates were acceptable for the total scale (α = .81) and for all subscales (with α values ranging from .74 to .84), and concurrent validity was supported by theoretically expected correlations with self-reported risky driving behavior (r ranging from .27 to .45, ps < .01). Subsequent comparison between zero-order correlation and partial correlation (controlling for Driver Impression Management) between ATTS subscales and self-reported risky driving behavior revealed minor or no effects of social desirability bias. Implications for road safety intervention are discussed.

The aim of this article was to describe the driving behavior profile of drivers aged 18 to 25 years old. Four hundred young adults were interviewed, 320 (80%) of them male and 80 (20%) female. Cluster analysis identified a group characterized by sensation-seeking behavior (Cluster 1), a group that did not show any risky driving behavior (Cluster 2), and a group engaged in transgressive behavior and driving under the influence of alcohol and/or drugs (Cluster 3). Discriminant analysis classified successfully and correctly 81.3% of the young adults into their original profiles. Function 1 distinguished cluster 1 from clusters 2 and 3, on the basis of the following factors: higher frequency of alcohol consumption, intrusive behavior, and motorcycle riding, as well as younger age, more aggressive behavior, and lower education level. Function 2 distinguished cluster 3 from cluster 1 and 2, especially as to higher amounts of alcohol consumption, higher frequency of marijuana use and delinquent behavior, larger number of traffic tickets and motor vehicle accidents, higher paternal education level, which were the variables with discriminant values above .20. Characteristics of vulnerability were identified, especially those related to alcohol consumption, drug use, and externalizing issues.