Initiatives
GILEE
ILEE
PSP
SAB
SAP

Sources
CP
CPAW
HS-CP
LE-Products
VIDEO
WEB
 

 

      Improving Crime Data

Best viewed with Internet Explorer - 6.0.

 Download Explorer

Can be viewed with Netscape 7.0 and above.

Download Netscape

 

The tables links below are best viewed with Acrobat Reader 6.0.  

Download Acrobat Reader 

horizontal rule

Back to: Press Release

 

Table-1: 2008 - Listing by Unadjusted/Adjusted Scores  Table-1A: 2009 - Listing by Unadjusted/Adjusted Scores

Table-2: 2008 - Listing by Cities  Table-2A: 2009 - Listing by Cities

 

Reports Page

 

ICD Page   Improving Crime Data Project

 

January 21, 2010

Technical Details: How the City Homicide Rates Were Adjusted

The statistical model used to estimate the 2008 and six-month 2009 adjusted homicide rates for the sample of large US cities was specified as follows:

Homrate = a + b1(Poverty) + b2(MdInc) +  b3(Unem) + b4(Black) +  b5(FemHead), where

bullet

Homrate = Homicides per 100,000 city residents;

bullet

Poverty = Percentage of families with incomes below the poverty line;

bullet

MdInc = Median household income;

bullet

Unem = Percentage of persons age 16 and older unemployed;

bullet

Black = Percentage of the population black; and

bullet

FemHead = Percentage of households headed by a female with children under the age of 18.

The homicide data are from the FBI’s 2008 and preliminary 2009 Uniform Crime Reports.  The six-month 2009 homicide data were not available for three cities (Albuquerque, Austin, and Indianapolis).  The socio-economic indicators are from the American Community Survey of the US Census Bureau.

The model was estimated using ordinary least squares regression on the 2006-2008 average homicide rates and socio-economic indicators for the 63 cities. The parameter estimates from this model were then applied to the 2008 and half-year 2009 homicide rates. The model explains about 73% of the variation in homicide rates across the cities.

The residuals from this model (the observed homicide rates minus the homicide rates predicted by the model) represent that component of each city’s homicide rate that is not explained by the variables in the model. The adjusted homicide rankings are based on the standardized residuals from the model.  

See Table-1 for a list of the 63 cities in 2008 and Table-1A for the 60 cities in 2009 arrayed by the unadjusted and by the adjusted rankings; see Table-2 for an alphabetical listing of the 2008 cities and Table-2A for the 2009 cities.

horizontal rule