Evaluation Of The Intervention Program Based On Network Analysis

In recent years,there has been a growing awareness of the importance of mental health to individuals,families and society,which has led to a geometric growth of psychological intervention programs and higher requirements on the evaluation methods.The current evaluation method for intervention programs mainly compares whether there is a significant difference in the improvement of the target variable between the intervention group and the waiting group from the sum scores level,which ignores that the intervention may have different effects on different dimensions of the target variable.Moreover,the current evaluation methods also rarely consider how the intervention effect is achieved and whether there are individual differences in the intervention effect.In this study,the network analysis method was applied to evaluate the intervention effect of self-compassion intervention program from the perspectives of group,time and individual,so as to demonstrate the advantages of network analysis in evaluating the effect of intervention programs.At the group level,the impact of intervention on the dimensions of self-compassion and the relationship between the dimensions was discussed,so as to evaluate the effectiveness of the intervention program;At the time level,the dynamic process of intervention was discussed and the effectiveness of the weekly intervention was evaluated;At the individual level,this study discussed the change patterns of self-compassion dimensions which affected by the intervention and explored the individual differences in intervention effects.In this study,a total of 119 college students were recruited and randomly assigned to the intervention group and the waiting group,based on the pre-post and daily follow-up controlled design.Participants in the intervention group were provided with a fourweek self-compassion stress management program(online version)and daily selfcompassion measure,while the participants in the waiting group only received the daily self-compassion measure.The results of this study were as follows:(1)Non-network analysis: After four weeks of intervention,participants in the intervention group showed significant improvement in self-compassion,each dimension of self-compassion,self-esteem,well-being,depression,anxiety and stress levels compared with those in the waiting group.(2)Group level analysis: In the network on the differences of self-compassion dimension,the intervention directly affected the dimensions of self-kindness,selfjudgment,isolation,mindfulness and over-identification,which basically covered the three components of self-compassion,indicating that the intervention program could comprehensively promote the participants' level of self-compassion.In the temporal network,there were more feedback loops between the nodes in the intervention group,but there were only self-loops effect of the nodes in the waiting group,which reflected that the intervention strengthened the correlation among the self-compassion dimensions,and all dimensions influenced and promoted each other,so as to improve the self-compassion level.This process reflected the dynamic process of selfcompassion as a collaborative interactive system.In contemporaneous network,the nodes of the intervention group network were more correlated,while those of the waiting group network were less correlated,which reflected that at the within-subject level,the two groups had a large difference in the correlation of each dimension of selfcompassion.However,whether the difference was attributed to the intervention effect still needed to be further explored.In between-subject network,the connections of nodes were very similar between the two groups,emphasizing the importance to distinguish between intra-individual and inter-individual levels when examining the effects of interventions.(3)Time level analysis: the results of weekly self-compassion dimension network showed that the intervention effects were basically consistent with the intervention content.The first week of body scanning training effectively improved the participants' levels of mindfulness and sense of common humanity.The second week of selfcompassion breathing training significantly reduced the levels of self-judgment and over-identification as well as promoted the level of mindfulness.The third week of training to show kindness to the loved ones could effectively improve the level of selfkindness,and at the same time improved the sense of common humanity and reduced the sense of isolation.The fourth week of loving-kindness meditation could effectively reduce participants' sense of isolation and self-judgment.(4)Individual level: According to the analysis of individual self-compassion network,effective individuals accounted for 60%.There were some similarities in the change pattern of self-compassion dimensions which were affected by the intervention.However,there were no difference in the level of the all pre-test variables as well as the demographic variable between the intervention effective and invalid individuals.To sum up,the current study comprehensively evaluated the effects of selfcompassion intervention at the group,time and individual levels,demonstrating the potential of network analysis in the field of intervention evaluation research.Finally,in order to facilitate the researchers' subsequent exploration,R codes of the relevant analysis were provided.