Systematic Social Observations Of An Offender’s Journey To Crime

Street View

The rise of geographic information systems have contributed significantly to the development of crime pattern
theory and the routine activities perspective. However, 2D maps do not account for observational factors that
can contribute to offending. This study conducts a systematic social observation of the theoretical journey to
crime for a sample of property crime offenders in British Columbia, Canada. Observations are collected in a
novel and cost effective method by using Google Street View. The physical and social characteristics of the
neighborhood and residences were observed for offender’s home location, the crime location, and a sampling
of the street blocks along the offender’s travel path. The observations illuminate the offender’s target selection
process, from both the rational choice perspective and the social disorganization perspective. In addition to
adding an important visual dimension to the geographic computer modeling of crime, the results also
contribute new insights to placed-based crime prevention strategies.

Google Street View is being used as a novel tool for collecting systematic social observations (SSO).

A variety of urban landscape constructs, such as physical disorder and defensible space, are being assessed. Google Street View is a cost effective approach to SSO. In addition, this study is the first systematic observation utilizing Google Street View ‘time machine’ feature. ‘Time machine’ allows observers to view the streets at all periods of Google Street View capture, instead of only the most recent. The observations provides a foundation for live or future observations as data because available, especially for street blocks with high concentrations of property crime.

Observation instrument

Street blocks are observed using a modified version of the SSO i-Tour Instrument (Odgers, Bates, Caspi, Sampson, & Moffit, 2009). The modified instrument places greater emphasis on features of the built environment associated with crime. A public demo of the modified instrument is available here.

A code book for systematic social observation in Google Street View, including details on observations using the new ‘time machine’ will be available in August.

Remote observations will be triangulated and validated with other measures of the urban landscape.

Criminal event and offender data.

The Royal Canadian Mounted Police provided criminal event and offender data for the city of Coquitlam.  Criminal event locations were provided for property crimes from August 2001 to August 2006.  Criminal event locations were aggregated to street segments. A sample of high crime, low crime and no crime street segments were selected for analysis.

Census, land use, and ariel observation data.  

Sociodemographic information supplemented observations. Sociodemographic information was retrieved at dissemination block level using the index of material and social deprivation  (Gamache, Pampalon, & Hame, 2006) and  the Canadian marginalization index (Matheson, Dunn, Smith, Moineddin, & Glazier, 2006). Government identified land use data was also retrieved for the street segments. Ariel observations (AO) of the street segments were conducted using Google Earth Pro. Ariel observations included measures of street segment design, green space canopy, and road network importance.

The project is conducted in collaboration with Dr. Richard Frank at Simon Fraser University.