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

Homrate = Homicides per 100,000 city residents;
Poverty = Percentage of families with incomes below the poverty line;
MdInc = Median household income;
Unem = Percentage of persons age 16 and older unemployed;
Black = Percentage of the population black; and
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.

Technical Details: How the City Homicide Rates Were Adjusted
The statistical model used to estimate adjusted homicide rates for the 63 large US cities was specified as follows:
Homrate = a + b1(Poverty) + b2(MdInc) + b3(MaleUnem) + b4(Black) + b5(FemHead), where

Homrate = Homicides per 100,000 city residents;
Poverty = Percentage of families with incomes below the poverty line;
MdInc = Median household income;
MaleUnem = Percentage of males age 20-64 unemployed;
Black = Percentage of the population black; and
FemHead = Percentage of families with children under the age of 18 headed by a female.

The homicide data are from the FBI’s 2007 Uniform Crime Report. 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 2005-2007 average homicide rates and socio-economic indicators for the 63 cities. The parameter estimates from this model were then applied to the 2007 homicide rates. The model explains about 71% 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.

Technical Details: How the City Homicide Rates Were Adjusted
The statistical model used to estimate adjusted homicide rates for the 65 US cities with populations of about 250,000 or more was specified as follows:
Homrate = a + b1(Disadvan), where

Homrate = Homicides per 100,000 city residents; and
Disadvan = A factor representing the level of social and economic disadvantage that combines four highly intercorrelated variables (factor loadings in parentheses): the poverty rate (.880), unemployment rate (.717), % black (.889), % female-headed families w/own children under 18 (.938).

The homicide data are from the FBI’s 2006 Uniform Crime Report. The indicators of socioeconomic disadvantage are from the American Community Survey of the US Census Bureau.
The model was estimated using ordinary least squares regression on the 2004-2006 average homicide rates and disadvantage scores for the 65 cities. The parameter estimates from this model were then applied to the 2006 homicide rates. The model explains about 67% of the variation in homicide rates across the cities. The estimation results are as follows:

Homrate = 13.796 + 8.409(Disadvan)

F1, 63 = 126.03; p < .001; R2 = .667; N = 65

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.

Technical Details: How the City Homicide Rates Were Adjusted
The statistical model used to estimate adjusted homicide rates for the 65 US cities with populations greater than 250,000 was specified as follows:
Homrate = a + b1(Disadvan) + b2(Pop) + b3(Same Res) + b4(Divrate), where

Homrate = Homicides per 100,000 city residents (natural log)
Disadvan = A factor representing the level of social and economic disadvantage that combines five highly intercorrelated variables (factor loadings in parentheses): the poverty rate (.934), male unemployment rate (.888), % black (.839), % female-headed families w/own children under 18 (.928), and median family income (-.862)
Pop = City population in 2000 Census (natural log)
Same Res = % persons living in same residence 5 or more years
Divrate = % persons age 15 and over divorced

The data are from the FBI’s 2005 Uniform Crime Report and the 2000 census.

The model was estimated using ordinary least squares on the 2003-2004 average homicide rates for the 67 cities. The parameter estimates from this model were then applied to the 2005 homicide rates. The model explains about 67% of the variation in homicide rates across the cities. The estimation results are as follows:

Homrate = .138 + .597(Disadvan) + .152(Pop) – .011(Same Res) + .062(Divrate)

F4, 60 = 31.639; p < .001; R2 = .671; N = 65

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.

Technical Details: How the City Homicide Rates Were Adjusted
The statistical model used to estimate adjusted homicide rates for the 67 US cities with populations greater than 250,000 was specified as follows:
Homrate = a + b1(Disadvan) + b2(Pop) + b3(Same Res) + b4(Divrate), where

Homrate = Homicides per 100,000 city residents (natural log)
Disadvan = A factor representing the level of social and economic disadvantage that combines five highly intercorrelated variables (factor loadings in parentheses): the poverty rate (.934), male unemployment rate (.888), % black (.839), % female-headed families w/own children under 18 (.928), and median family income (-.862)
Pop = City population in 2000 Census (natural log)
Same Res = % persons living in same residence 5 or more years
Divrate = % persons age 15 and over divorced

The data are from the FBI’s 2004 Uniform Crime Report and the 2000 census.

The model was estimated using ordinary least squares on the 2002-2003 average homicide rates for the 67 cities. The parameter estimates from this model were then applied to the 2004 homicide rates. The model explains about 70% of the variation in homicide rates across the cities. The estimation results are as follows:

Homrate = .266 + .626(Disadvan) + .159(Pop) – .016(Same Res) + .061(Divrate)

F4, 62 = 35.522; p < .001; R2 = .696; N = 67

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.

Technical Details: How the City Homicide Rates Were Adjusted
The statistical model used to estimate adjusted homicide rates for the 67 US cities with populations greater than 250,000 was specified as follows:
Homrate = a + b1(Disadvan) + b2(Pop) + b3(Samres) + b4(divrate), where

Homrate = Homicides per 100,000 city residents (natural log)
Disadvan = A factor representing the level of social and economic disadvantage that
combines five highly intercorrelated variables (factor loadings in parentheses): the poverty rate (.934), male unemployment rate (.888), % black (.839), % female-headed families w/own children under 18 (.928), and median family income (-.862)
Pop = City population in 2000 Census (natural log)
Samres = % persons living in same residence 5 or more years
Divrate = % persons age 15 and over divorced

The homicide and population data for 2002 are from the FBI’s Crime in the United States 2002 Uniform Crime Report. The homicide data for 2003 are from the FBI’s 2003 Preliminary Report. The calculation of each city’s homicide rate (rate = homicides/population) used the same population figures for 2002 and 2003.

These are based on the census estimates of each city’s 2002 population used by the UCR. All other data are from the 2000 census.

The model was estimated using ordinary least squares on the 2000-2001 average homicide rates for the 67 cities. The parameter estimates from this model were then applied to the 2002 homicide rates. The model explains two-thirds of the variation in homicide rates across the cities. The estimation results are as follows:

Homrate2002 = .061 + .644(Disadvan) + .192(Pop) – .019(Samres) + .048(Divrate)

F4, 62 = 35.407; p < .001; R2 = .696; N = 67

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. These scores can be compared directly with the standard scores computed from the unadjusted rates ((y – m) / s).

The model was then also estimated on the 2001-2002 average homicide rates for the 67 cities. The model explains over 70% of the variation in homicide rates across the cities. The parameter estimates from this model were then applied to the 2003 homicide rates. The estimation results are as follows:

Homrate2003 = .485 + .679(Disadvan) + .174(Pop) – .023(Samres) + .047(Divrate)

F4, 62; = 39.237; p < .001; R2 = .717; N = 67

The estimations of this model on the 2000-2001 and 2001-2002 data display strong consistency and strong robustness. Given the strength of the model and the consistency of the parameters, then one can attribute the changes in the adjusted rankings between 2002 and 2003 to changes in the number of homicides.
See Table-1 for a list of the 67 cities arrayed by their homicide rate, standard score of the rate, and the standardized residual from the model; see Table-2 for a list of the cities arrayed by the unadjusted and by the adjusted rankings; see Table-3 for an alphabetical listing of the cities.