Confiabilidad y validez de un instrumento que mide siete dimensiones de la percepción de seguridad en estudiantes de una universidad pública

Introduction: Public security management involves the implementation of public policies that justify the guidance of the State in the prevention of crime and the administration of justice. However, citizen distrust of government action is evidenced by a growing insecurity perception reported by the literature in seven dimensions: territorial, national, public, human, citizen, private and Internet user. Objective: To establish reliability and validity of an instrument that measures the perception of territorial, national, public, human, citizen, private, and Internet user security. Method: Non-experimental, cross-sectional, exploratory study with a non-probabilistic selection of 320 students from a public university. Results: Reliability of the overall scale (alpha = 0.793), and territorial (alpha = 0.792), national (alpha = 0.709), public (alpha = 0.785), human (alpha = 0.782), citizen (alpha = 0.792), private (alpha = 0.794), and Internet user (alpha = 0.731) subscales, show sufficient internal consistency. The territorial security factor accounted for 22% of total variance. Based on adjustment and residual parameters ⌠χ2 = 135.34 (32 gl) p = 0.054; GFI = 0.995; CFI = 0.990; RMSEA = 0.003⌡, the null hypothesis of significant relationship among theoretical dimensions of security with respect to factors weighted was accepted. Conclusions: Inclusion and measurement of a dimension of self-control perception that would negatively and significantly correlate with the perception of territorial security would explain the factorial structure of the scale. Such model would be estimated by a confirmatory factorial analysis with unweighted least squares.