摘要：The main purposes of evaluation of intervention programmes are to check their effects and to improve them. Scientific checking of program effects in penology is usually performed using quasi-experimental design with two groups (treatment and control) in two time points: before (pretest) and after program application in treatment group (posttest). On the basis of obtained difference in criterion variable (in penology the most often recidivism) between treatment and control group in posttest the conclusion about program efficiency is made. The difference is in the most cases presented as the shift in common standard deviation (Cohen d) or as the correlation coefficient. Meta-analysis is often used in intervention programmes evaluation. Evaluations performed using treatment and control groups show only mean effects, but neglect individual differences. According to the principle of treatment individualisation, evaluation of programme effects in practical work with offenders should be individualized in the way that in defined time points planned and obtained results are compared, and when necessary some other modalities of program or a new program might be used. This could be done by cybernetic model.
关键词：penology; intervention programme; evaluation; cybernetic model
