Proefschrift

4 95 REPLICATION AND COMPARISON NETWORKS without recidivism estimated on data of both independent samples. Within the spinglass community networks, general social rejection/loneliness and capacity for relationship stability formed a distinct community in both samples, while a community formed by poor cognitive problem-solving and impulsive acts was most apparent in the BC sample. The walktrap community networks estimated on the DSP sample showed a second community of general social rejection/loneliness, capacity for relationship stability, and impulsive acts and a third community of six remaining dynamic risk factors. The walktrap algorithm based on the BC sample did show a second community of the remaining nine dynamic risk factors. 4.3.1.2 Networks including sexual recidivism Interrelationships between dynamic risk factors within networks including sexual recidivism were all positive for both the DSP and BC sample (Figure 4.2a and 4.2c). Within both networks, general social rejection/loneliness, lack of concern for others, impulsive acts and poor cognitive problem-solving were among the four dynamic risk factors with the highest estimated strength centralities. In contrast, capacity for relationship stability and significant social influences were relatively weak connected with other factors in both the original (DSP) and the replication (BC) sample sexual recidivism (Figure 4.2c and 4.2d). In both independent samples, the spinglass and the walktrap algorithms revealed a distinct community consisting of sexual preoccupation, sex as coping, deviant sexual interests, and emotional identification with children in the networks with sexual recidivism. The spinglass algorithm showed three additional, somewhat different, communities in the original (DSP) and replication (BC) sample. For the DSP sample, the second community was formed by general social rejection/loneliness, capacity for relationship stability, and negative emotionality/hostility, the third consisted of significant social influences, impulsive acts, poor cognitive problem-solving, and sexual recidivism, the fourth community contained hostility toward women, lack of concern for others, and cooperation with supervision. However, spinglass community analyses of the BC sample showed a second community of general social rejection/loneliness and capacity for relationship stability. Negative emotionality/ hostility, hostility toward women, lack of concern for others, cooperation with supervision, and significant social influences formed the third community. The fourth community consisted of impulsive acts, poor cognitive problem-solving, and sexual recidivism. The walktrap community networks estimated on both the original (DSP) and replication (BC) sample showed a second community of the remaining ten dynamic risk factors.

RkJQdWJsaXNoZXIy MjY0ODMw