Theoretical Foundation
A social-science foundation, not a safety startup.
Every feature in Gorebet traces to a research tradition. We document the connection so you can hold us to it.
Social Cohesion
OperationalCollective Efficacy
Sampson, Raudenbush & Earls, Science, 1997
Robert Sampson's longitudinal research across Chicago neighborhoods established that communities where residents trust each other and act on shared expectations — not just police density — exhibit measurably lower violent crime rates. The two dimensions are mutual trust among neighbors and shared willingness to intervene for the common good.
How Gorebet applies this
Gorebet's neighborhood groups and community reaction layer directly operationalize the behavioral infrastructure of collective efficacy. When community members report an incident and others react to it, they are enacting the behavioral proxy Sampson measured. The neighborhood group membership count and reaction rate are Gorebet's primary collective efficacy proxies — behavioral traces, not survey-validated scores.
Environmental Opportunity
OperationalRoutine Activity Theory
Cohen & Felson, American Sociological Review, 1979
Lawrence Cohen and Marcus Felson showed that crime is less about pathological individuals than about convergence: motivated offenders, suitable targets, and the absence of watchful presence meeting in the same place at the same time. Crime prevention operates by disrupting this convergence — increasing guardianship, reducing target attractiveness, or modifying routines.
How Gorebet applies this
Gorebet's incident map raises effective guardianship by making spatial patterns visible. The notification system activates distributed guardianship — more eyes at the right place and time. The Protected Asset framework directly addresses target reduction by flagging at-risk assets to their operators.
Space-Time Paths
RoadmapTime-Geography
Hägerstrand, Papers in Regional Science, 1970
Torsten Hägerstrand established that human activity is structured by time-space constraints. Events in space are inseparable from events in time — patterns of risk, gathering, and safety are constituted by daily rhythms, not fixed geography.
How Gorebet applies this
Every Gorebet report carries a precise timestamp and location. This geo-temporal data is suited to Hägerstrand-style analysis of when incidents cluster at which places. The NeighborhoodSafetyEngine's temporal decay function — more recent incidents carry higher weight — is a direct application of time-geography thinking.
Micro-Place Concentration
RoadmapEnvironmental Criminology & Crime-and-Place
Brantingham & Brantingham (1981); Sherman & Weisburd (1995); Weisburd et al. (2004)
Most crime concentrates at specific micro-places — street segments, addresses, activity nodes — that persist over time. Lawrence Sherman and David Weisburd demonstrated that in any city, roughly half of all crimes occur at just a few percent of addresses.
How Gorebet applies this
Gorebet's map and clustering engine make micro-place patterns visible, giving community members, service operators, and public agencies the spatial intelligence to act where it matters most. Street-segment-level clustering is a planned enhancement beyond the current woreda-level scoring.
Handled with Caution
OperationalThe Disorder Hypothesis
Wilson & Kelling (1982); critiqued in Harcourt (2001)
Wilson and Kelling's 'broken windows' hypothesis proposed that visible disorder signals low social control and invites further disorder. The empirical literature on whether disorder causes crime — as opposed to being correlated with concentrated disadvantage — is contested.
How Gorebet applies this
Gorebet supports reporting of utility failures, infrastructure issues, and public-space concerns — categories closer to maintenance requests than punitive social control. We explicitly do not recommend that incident data drive enforcement decisions without human review, institutional governance, and bias auditing.
Social Diffusion
OperationalSocial Network Theory
Granovetter, American Journal of Sociology, 1973
Mark Granovetter's research established that information spreads most effectively through weak ties — the acquaintances who bridge otherwise disconnected clusters. Adoption of a new civic behavior, like reporting, can cascade through neighborhoods if seed users activate bridges.
How Gorebet applies this
Gorebet's neighborhood membership creates an explicit social graph. The feed recommendation engine surfaces incidents from users' neighborhoods, creating information cascades along this graph. Identifying bridge users — those who connect otherwise disconnected clusters — is a planned research extension for supporting community outreach.
Equity & Coverage
RoadmapSpatial Inequality & Report Deserts
Sampson (2012); Lum & Isaac (2016)
Concentrated disadvantage — poverty, residential instability — shapes neighborhood-level outcomes and is spatially clustered. Measurement systems built on existing reports can amplify inequalities if low-reporting areas are treated as low-incident areas. Kristian Lum and William Isaac demonstrated this feedback loop in predictive policing systems.
How Gorebet applies this
In woredas with low app adoption, Gorebet scores underrepresent actual incident rates — creating 'report deserts.' The NeighborhoodSafetyEngine's Bayesian shrinkage component partially corrects for this by pulling low-count estimates toward the city mean rather than treating silence as safety. The annual fairness audit will compare score distributions across woreda income proxies, language groups, and gender.
Research Methods
OperationalComputational Sociology Methods
Salganik (2018); Barocas, Hardt & Narayanan (2023)
Computational sociology applies the rigor of quantitative social science — hypothesis testing, uncertainty quantification, causal identification, fairness auditing — to large-scale behavioral traces that digital platforms generate.
How Gorebet applies this
Every Gorebet score, map layer, and aggregate is accompanied by a methods note: what data feeds it, what assumptions it makes, where it is reliable, and where it may be wrong. We commit to publishing our aggregation thresholds, suppression rules, and an annual fairness audit.
Full bibliography
- Barocas, S., Hardt, M., & Narayanan, A. (2023). Fairness and Machine Learning. MIT Press. fairmlbook.org
- Brantingham, P.L., & Brantingham, P.J. (1981). Environmental Criminology. Sage.
- Cohen, L.E., & Felson, M. (1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 44(4), 588–608.
- Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology, 78(6), 1360–1380.
- Hägerstrand, T. (1970). What about people in regional science? Papers in Regional Science, 24(1), 6–21.
- Harcourt, B.E. (2001). Illusion of Order: The False Promise of Broken Windows Policing. Harvard University Press.
- Lum, K., & Isaac, W. (2016). To predict and serve? Significance, 13(5), 14–19.
- Salganik, M.J. (2018). Bit by Bit: Social Research in the Digital Age. Princeton University Press. bitbybitbook.com
- Sampson, R.J. (2012). Great American City: Chicago and the Enduring Neighborhood Effect. University of Chicago Press.
- Sampson, R.J., Raudenbush, S.W., & Earls, F. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277(5328), 918–924. doi:10.1126/science.277.5328.918
- Sherman, L.W., & Weisburd, D. (1995). General deterrent effects of police patrol in crime 'hot spots.' Justice Quarterly, 12(4), 625–648.
- Weisburd, D., Bushway, S., Lum, C., & Yang, S.M. (2004). Trajectories of crime at places. Criminology, 42(2), 283–321.
- Wilson, J.Q., & Kelling, G.L. (1982). Broken windows: The police and neighborhood safety. The Atlantic Monthly.