Conflict Intervention as Crime Prevention

Report Author

Justin R. Corbett

Linkedin | Google | ORCID

Table of Contents

Mechanism of Impact

Constructive Propagation

Constructive propagation is the process by which attitudes and behaviors supportive of positive approaches toward conflict cascade through social networks. This propagation occurs through interpersonal transmissions in much the same way viral contagions spread, meaning individuals can effectively catch colds of cooperativeness (or criminality) as a result of innate psychosocial processes.359

The nature and manipulability of constructive propagation’s own mechanisms are the focus of intense ongoing research. Already, however, studies suggest service providers could begin leveraging known qualities of microsociological and network dynamics to distribute constructive conflict engagement beliefs and skill sets into the broader public. They can begin to optimize their outreach to develop network interventions360 informed by the friendship paradox,361 hyper-sensitivity in disadvantaged neighborhoods,362 network size and strength,363 nodal and network absorptive capacities,364 signal redundancy,365 and solidarity preferencing and innate mimicry.366

Even without already leveraging such insights, the constructive propagation of service provider-cultivated behaviors and skill sets is already having an effect. Rough modeling of the providers’ interventions suggests they are contributing a networking effect of 5.5x.367 This means that for every direct service recipient, 5.5 total individuals will have observable, constructive changes in the way they approach conflict, consequently minimizing the criminogenic risks of all influenced. Given this rate of propagation from 2016 service recipients, providers will likely have constructively influenced as many as 190,000 individuals before the regulatory force of these ripples observably dissipates.

Footnotes

359. A growing number of interpersonal and social phenomena are acknowledged to demonstrate contagion characteristics wherein individual behaviors, emotional states, ideas, and even complex physical conditions can spread from person to person in much the same way and with even greater efficiency than viruses are able to affect families, schools, or communities (Fowler, J.H., Christakis, N.A., 2010). Attitudinal and behavioral approaches to conflict is but one such phenomenon. Study of these transmissions constitutes a form of network epidemiology and has revealed effects can range from acute to chronic, be near or distributed, and of nominal or life-altering consequence.

360. Valente, T.W. (2012) describes network interventions as “purposeful efforts to use social networks or social network data to generate social influence, accelerate behavior change, improve performance, and/or achieve desirable outcomes among individuals, communities, organizations, or populations.”

361. There are numerous theoretical systems available to maximize propagation distribution and persistence. Almost all of them, however, are complex and resource heavy. A clever hack that leverages a paradox of human social networks is to use the friendship nominating technique. The friendship paradox, in essence, states that on average, your friends have more friends than you do. In fact, in terms of their network features, your friends are also better informed (closeness), better intermediaries (betweenness), and more powerful (eigenvector) than you (Grund, T.U., 2014) Capitalizing on this by requesting randomly selected individuals identify another friend to receive the organizations’ information, services, or training is the most efficient strategy for propagating information distribution and attitudinal and behavioral shifts. Specific to criminal contexts, identifying co-offenders of a random selection of offenders to receive intervention services would statistically improve the chances of serving more connected, influential criminals from which the constructive benefits of the intervention are more likely to further propagate (Grund, T.U., 2014). For more on the friendship paradox and its successful applications, see: Feld, S. (1991); Kim, D.A., et al. (2015).

362. Social networks in disadvantaged neighborhoods are particularly susceptible to the social contagion of cooperativeness given sufficient encouragement. This is because social networks in disadvantaged neighborhoods are often deceptively dynamic, that is, while their compositions appear fluid (e.g. individuals possess both agency and opportunity to regularly manage the size and membership of their social network), in reality, they are more limited and static than networks centered in less disadvantaged contexts. Unlike more fully fluid networks, however, these relatively equilibrated, viscous networks become more prone to the effects of social contagion of cooperation. Importantly, this contagion only takes full effect after repeated confirmatory interactions (Jordan, J.J., et al., 2013), suggesting: (1) the uptake of cooperative behaviors is likely to be slow, particularly through population-level interventions; (2) expectations for and resources committed to such intervention need to be allocated appropriately; and (3) properly supported, cooperative- focused interventions can influence network- and neighborhood-level conflict and criminal characteristics.

363. Propagation of constructive cognitive behavioral approaches to conflict is dependent upon the number and strength of relationships an individual maintains (McFadyen, M.A., Cannella, A.A., 2004). As to the number and strength of relationships, it is unsurprising that those with few and weak social networks are unlikely to substantively contribute to approach propagation. Somewhat surprisingly, though, is the decreasing returns on propagation from those with greater and stronger social networks. This is because those intimately connected to existing networks are often strongly associated with and benefit from key dynamics within that network. Their close identification with existing norms creates powerful personal incentives that favor network stasis and resist substantive change regardless the objective benefit any such change may have for the broader network. As an alternative, targeting those with moderately higher numbers (+1.0 SD) of and moderately stronger (+.05) network connections will net higher knowledge creation (McFadyen, M.A., Cannella, A.A., 2004) and theoretically generate the most efficient propagation of that new knowledge.

364. The absorptive capacity of a network is determined along four dimensions: acquisition (the capacity to recognize, understand the importance of, and acquire external knowledge); assimilation (the capacity to integrate external knowledge using routines and processes that allow it to understand, analyze, process, and interpret information obtained from external sources); transformation (the capability to develop and refine routines that facilitate combining existing knowledge and the newly acquired and assimilated knowledge); and exploitation (the capacity to competitively use new external knowledge to achieve goals). For a more thorough review of absorptive capacities, see: Noblet, J., Simon, E., Parent, R. (2011); and Zahra, S.A., George, G. (2002).

365. The presence of reinforcing signals increases both the adoption of and commitment to contagious behaviors, particularly among reticent adopters. Centola, D. (2010) notes: “social reinforcement from multiple...buddies made participants much more willing to adopt the behavior... Participants were significantly more likely to adopt after receiving a second signal than after receiving only a single signal...Receiving a third signal also significantly increased the likelihood of adoption, but with a smaller marginal-effect size.” Centola also notes that these repeated signals affected adopters subsequent level of engagement with the targeted behavior, which increased 1.67x for those who received a second signal, and a further 1.32x for those who received three signals. For service providers, this suggests those least likely to utilize cooperative, non-criminal responses to conflict (i.e. those with pre-existing or at high risk of developing criminal careers) will likely ultimately convert to such an approach given a large and diverse enough sampling of their network precedes them in that conversion. Once converted, they are more likely to actively retain that new behavior than those whose adoption process was comparatively faster and less reinforced.

366. Collins, R. (2004).

367. As it is important to estimate the downstream effects of constructive conflict interventions and skills trainings, and as no known field-tested model current exists, a hypothetical model is proposed here based on the following assumptions. (1) Service recipients will only adopt attitudinal and behavioral changes if they perceive those alternatives as procedurally just. While reports vary on the rate of perceived procedural justice for alternative dispute resolution and restorative justice services, Miller, S.L., Hefner, M.K. (2015) estimate the rate at 80% or greater. (2) Diffusion of adopted attitudes/behaviors is most likely to occur among those who have regular, close contact with the original service recipient (network seed). The probability of secondary adoption is even greater among those regular, close contacts with whom the seed otherwise discusses important matters (2.32 individuals, on average, at the originating and each successive network level; Rözer, J., Mollenhorst, G., Poortman, A., 2016). Though attitudinal/behavioral adoption has been shown to persist across three degrees of separation from the network seed (Fowler, J.H., Christakis, N.A., 2010), its influence dissipates progressively at each affected degree (reported at 1o: 100%, 2o: 25%, 3o: 6.25%; Pinheiro, F.L., et al., 2014). Presuming perceived procedural justice is influential at each degree of propagation, the aggregated network effect is thus equal to 5.5x (i.e. 5.5 total individuals, on average, are constructively and conspicuously affected as a result of each positively influenced service recipient).

Different services will produce – potentially significant – variation in this effect. For example, 32-hour mediation trainings are likely to produce far longer lasting and influential effects than short workshop presentations. Similarly, a multiple-session Peacemaking process is likely to produce a far more substantial affect than a comparatively shorter coaching session. In this regard, this conservative projection almost certainly underestimates the networking effect of service providers’ interventions, as almost every intervention they offer is substantially more involved than the interventions used in the source materials referenced to inform its calculation.