Linking Prisoner Centralization with Social Connectedness: Five Years of Prisoner Visitation

Background & Purpose: Budget concerns contribute to many states consolidating prison facilities as cost-cutting measures. Decreased visitation becomes a potential side effect as prisoners are moved to increasingly centralized locations; women’s facilities tend toward greater centralization anyway. However, research indicates that success with post-incarceration reentry is strongly influenced by prisoners’ social support systems which may be disrupted without visitation. Study objectives were to identify visitation patterns for men and women related to distances between home communities and prisons over time and movement between facilities. Methods: This secondary analysis utilized state Department of Rehabilitation and Corrections (DRC) data for N=37,752 prisoners (5% women) in one state’s 31 facilities. Variables included prisoner demographics, approved visitors data, prisoner relocations, and information concerning each visit received during the period July 1, 2006 to July 19, 2011. The sample excluded prisoners with no approved visitors, coming from out-of-state, and with no chance of community re-entry (original N=50,054). Analysis of variance and descriptive statistics were computed using SPSS-v.19; an R package for generalized linear mixed models (LME4) assessed adjustment of weekly visit probabilities by predictor variables included in a cross-classified, random effects, discrete time event history model. Data were organized such that each record represented whether or not a visit to prisoner i from potential visitor type j on week w occurred. Predictors included static prisoner characteristics (gender, age, marital status at admission, and sentence length) and visitor relationship (family or friend), as well as time varying characteristics (prisoner security level and distances between reported zipcode of residence to current institution, and indicators for both first and last six months of sentence). Models tested were adjusted for prisoner security levels. Results: Men had significantly more institutional moves, more approved visitors, and more visits than women (p<.001). Fitted parameters were similar for male and female data sets. Among both genders, younger prisoners and prisoners in an intimate relationship were more likely to receive visitors than their older and single counterparts (p<.01). Prisoners were more likely to receive visits from family than from friends (p<.01). Men were less likely to receive a visit during the first and last six months of their sentences (p<.01) compared to the rest of their sentence; women were less likely to receive a visit during their first six months, but were no less likely to receive visits six months before release. Finally, each additional mile a female was placed from her home zipcode was associated with a 1.4% reduction in the likelihood of a visit (p<.01); this reduction was 1.6% for males (p<.01). Conclusions and Implications:  Assuming that reentry outcomes are influenced by supportive social networks (family and friends), increased centralization may impede visitation which, in turn may negatively affect support network strength, particularly among women. Social work interventions to enhance supportive family and friend networks during the months prior to community reentry are important to explore, as are their potential impact on reentry outcomes. Methodological contributions include imposing complex analytic models on prisoner visitation data, given the structural influences on visit probabilities.