{"title":"El Análisis de Supervivencia como Técnica para la Evaluación de la Validez Predictiva en la Psicología Jurídica","authors":"Julio Isaac Vega-Cauich","doi":"10.5093/APJ2018A11","DOIUrl":null,"url":null,"abstract":"Being able to determine that an instrument predicts behavior such as violence or recidivism is one of the main objectives of legal psychology. Predictive validity, as it is known, allows evaluating this characteristic. This type of validity has been evaluated through various techniques, such as contingency , sensitivity, specificity, andROC curves. However, these techniques do not address a major statistical problem: data censorship that occurs in prospective studies, since in many cases the subjects will not relapse during the follow-up performed. Survival analysis is a technique that allows taking this information into consideration and modeling the time until a certain event happens. Through different approaches, such as its descriptive techniques or Cox proportional hazards models, this technique allows us to evaluate the predictive power of tests, as well as of interventions.","PeriodicalId":44109,"journal":{"name":"Anuario De Psicologia Juridica","volume":"1 1","pages":""},"PeriodicalIF":0.9000,"publicationDate":"2019-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Anuario De Psicologia Juridica","FirstCategoryId":"102","ListUrlMain":"https://doi.org/10.5093/APJ2018A11","RegionNum":4,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"LAW","Score":null,"Total":0}
引用次数: 1
Abstract
Being able to determine that an instrument predicts behavior such as violence or recidivism is one of the main objectives of legal psychology. Predictive validity, as it is known, allows evaluating this characteristic. This type of validity has been evaluated through various techniques, such as contingency , sensitivity, specificity, andROC curves. However, these techniques do not address a major statistical problem: data censorship that occurs in prospective studies, since in many cases the subjects will not relapse during the follow-up performed. Survival analysis is a technique that allows taking this information into consideration and modeling the time until a certain event happens. Through different approaches, such as its descriptive techniques or Cox proportional hazards models, this technique allows us to evaluate the predictive power of tests, as well as of interventions.