This project addressed
residential burglaries in San Diego, California. We
correlated crime statistics with different types of households, and the
outcome maps were created through using data from: American
Factfinder, the US Census
Bureau, and the San Diego
Police Department. The
variables considered were: the
total population (per census tract), the median household income, the
percentage of owner occupants and renter occupants per census tract,
and the percentage of single unit homes and multiple unit homes. Additionally,
we created regression graphs which presented us with R² values,
which revealed correlation coefficients
that indicated the strength of the relationship between the above
mentioned variables and the number of burglaries. The
number of renter occupants, and the number of multiple unit homes were
found to be the strongest indicators in predicting the number of
residential burglaries. Because of their moderate R² values of
approximately 0.4, neither of these factors can be used as absolute
measurements in determining the number of residential burglaries.
However, strong deviations from expected values were used to identify
areas that consistently had higher numbers of burglaries and could thus
benefit from crime prevention programs, and preferable residential
areas where relatively low levels of burglaries were reported.
|
