Compilation And Verification Of Risk Assessment Scale For Criminal Offense Violence In Prison

In recent years,with the revision of China’s laws and regulations,such as the Criminal Law Amendment(8)and the Criminal Procedure Law,the structure of detainees in prisons in China has undergone great changes.Death sentences and heavy prisoners in prisons have been greatly restricted in commutation of sentences,and the minimum period after commutation has also risen to 25 years.At the same time,criminals with shorter residual sentences are also handed over to the prison for custody.This has led to an increase in the number of personnel in the prison and a significant change in the structure of the detainees.In addition,because short-term criminals have the characteristics of ugly management,disobedience and discipline,and criminals in prison are more likely to have psychological problems.This will increase the risk of violent or direct physical conflict among criminals in prison.This undoubtedly brings great difficulties and challenges to the management of the prison,which makes it necessary for the prison to strengthen the supervision of the criminals and classify and manage the personnel to prevent some dangerous events.Therefore,the prediction of the risk of criminals in prison is also a demand.In addition to predicting the individual’s risk,the risk assessment tool can also help the individual’s treatment plan.At present,the tools for violent risk assessment have been developed so far,and there are some predictive tools with high reliability and validity,such as HCR-20 and VRS.However,at present,most of the tools available in China are imported and revised foreign evaluation tools.Although there are many domestic scholars’ own evaluation tools,there are few tools for the risk of violence in criminal prisons.Therefore,this study specifically developed a survey scale for the risk of violence in prisons.In the early stage,through reading the relevant literature and interviewing the criminals in the later period,the factors included in the research scale were summarized,mainly static factors,dynamic risk factors and protective factors.There are 177 items in the initial measurement form,and the initial effective questionnaire is 248.Through project analysis and exploratory factor analysis for each subscale.The resulting scale consists of seven subscales,namely the static factor subscale,the protective factor subscale,and the five subscales under dynamic risk factors(social cognition and coping style,emotional characteristics,personality and temperament characteristics,family environment,prison incompatibility subscales),a total of 124 items.The structural factors model is used to analyze the theoretical factors of the proactive and reactive aggression.The results show that the two models have good fitness.Then,the cluster analysis of 248 subjects was carried out,and it was found that the best results were obtained when they were divided into 4 categories.Category 1 is a high-risk group(extreme group),which is basically at a high level for each factor;Category 2 is classified as a reactive aggression,with anxiety,anger,and psychopathology as core traits,and a considerable level of response.Mode aggression,authoritarian parenting style,hostile attribution bias,criminal attitude,and high sensory seeking;3 categories are not obvious groups,basically low level in each factor;4 categories are classified as the proactive aggression.,Narcissism is a core trait,and it also has a fairly low level of self-control,criminal attitude,ruthlessness and high sense of seeking.Because of the clustering analysis,the reactive aggression includes criminal attitude and high sensory seeking factor.Therefore,the structural equation model analysis of the reactive aggression is carried out,and the fitting degree is good.At the official testing stage,the self-made scale was tested simultaneously with the China Revised Attack Questionnaire(BWAQ)and the Active Responsive Attack Questionnaire(RPQ).The prisoners of 5 different district prisons were selected,and the valid questionnaires were 1041,and the data were subjected to confirmatory factor analysis and reliability and validity analysis.The results show that the self-scaled table has good test-retest reliability,and the six subscales have good structural validity.The BWAQ and RPQ were used as the calibration questionnaires.The results showed that each subscale was significantly correlated with the BWAQ total score,and the initiative in the self-programmed scale.The reactivity and the factors were significantly correlated with the total score of the active attack and the reactive attack in the RPQ.The cluster analysis of the subjects who were officially tested also showed the same results as the initial cluster analysis,indicating that the differentiation of the initiative and the reactive attack of this study has good validity.Finally,gain validity analysis of each subscale found that each subscale has an additional impact on the aggression behavior after controlling other scales.All of these indicate that the Proactive and Reactive Violence Risk Assessment Scale has good reliability and validity.