Integrating standard and network psychometrics to assess the quality of prison life in Serbia

Abstract

In the present research, we analysed the properties of the Measuring the Quality of Prison Life (MQPL) survey in Serbia; it assesses five dimensions of prisoners’ well-being (harmony, professionalism, security, conditions and family contact, and well-being and development) composed of 21 narrow scales. The participants were 650 prisoners serving sentences in five prisons. Reliabilities (measured by both Cronbach's alphas and test–retest correlations) were high for most narrow scales and excellent for global scales. Confirmatory factor analysis showed a satisfactory fit of the model although the MQPL dimensions showed high intercorrelations (indicating elevated informational redundancy). The validity of the scales was established by detecting positive correlations with the WHOQOL-BREF quality of life scale and by capturing the differences between the five prisons in which the data were collected. We used network analysis to detect the most central nodes in MQPL: analysis on the dimension level revealed that harmony, professionalism, and well-being and development had high centrality. Estimating the network on the level of narrow scales demonstrated that prisoners’ well-being, organisation and consistency in prison activities, and help and assistance from prison staff were central aspects of their quality of life. Therefore, the present findings show that MQPL scales have high reliability and validity, the model fits the empirical data, and the central aspects of prisoners’ quality of life are identified; at the same time, we also analysed the limitations of MQPL. In general, the results suggest that MQPL is a valuable tool for assessing the quality of life and social climate in prisons.