Supporting Data for 2017 DPIC Study of Murder Rates and Killings of Police

How DPIC conducted the analysis


Our Goal

We wanted to find out what homicide numbers and trends would tell us about whether the death penalty deterred murders and, more specifically, what kind of impact—if any—it had in deterring killings of law enforcement in the line of duty.

How We Classified the States

We broke the states down into three categories that we thought would tell us something different about whether the death penalty was a deterrent and created four comparison groups:

  1. United States: the country as a whole.
  2. Death-Penalty States: those states that have had the death penalty continuously from 2000 onward (death penalty states).[1]
  3. Non-Death-Penalty States: those states that have never had the death penalty at any time in the 21st century.
  4. Transitional States: those states that had the death penalty as of 2000 but, as a result of legislative or judicial abolition, no longer do.

Hypotheses

Our hypotheses were that, if the death penalty were a deterrent, then—all other things being equal:

  • Murder rates should be lower in states with the death penalty than in states that did not authorize capital punishment.
  • When homicide rates rise nationally, they should rise less in death-penalty states than in non-death-penalty states.
  • When homicide rates fall nationally, they should fall more in states that have the death penalty than in states that don’t. 
  • Murder rates should rise more (or fall less) after states abolish the death penalty in comparison to both death-penalty states and states that had long before abolished the death penalty.
  • The pattern of disproportionately large increase and disproportionately small decrease in the transitional states should be consistent across the states that abolish the death penalty.

With respect to whether the death penalty protects law enforcement, we hypothesized that everything that should be true about the death penalty in general should also be true with respect to the killings of police officers. if the death penalty were necessary to protect law enforcement, then—all other things being equal:

  • Tthe rates at which law enforcement personnel were murdered should be lower in states with the death penalty than in states that did not authorize capital punishment.
  • When killings of police officers rise nationally, they should rise less in death-penalty states than in non-death-penalty states.
  • When rates of killing officers fall nationally, they should fall more in states that have the death penalty than in states that don’t.
  • In transitional states that abolish the death penalty, there should be a noticeable change in the rates at which police were killed, and the rates of police killings should not only rise following abolition of the death penalty, but should rise disproportionately to the rates at which law enforcement personnel are killed both in long-time death-penalty states and in states that had long ago abolished the death penalty.

In addition, if the death penalty was more necessary to protect police officers than other potential victims or had any special deterrent value when it came to homicides of law enforcement personnel:

  • The percentages of all murders that involve law enforcement victims should be lower in death penalty states than in states that do not have the death penalty.
  • If police are especially vulnerable without the death penalty, murders of officers should rise as a percentage of all homicides when states abolish the death penalty.
  • If there are increases in murders of police officers, the pattern of increases should be consistent across the states that have abolished the death penalty.

The Data We Used

We looked at murder rates nationally from 1987 through 2015. We chose 1987 as the starting date, because that was the earliest date for which we found FBI Uniform Crime Statistics on officers feloniously killed in the line of duty.

  • Our source of information on the number of murders nationwide and in each state was the FBI Uniform Crime Reports, Annual Murder Data from 1987 to 2015.
  • Our source of information on the numbers of murders of law enforcement nationwide and in each state was the FBI Uniform Crime Reports, Law Enforcement Officers Killed & Assaulted, Officers Feloniously Killed Annual Data (LEOKA reports), 1987 through 2015.
  • Our source of information on population nationwide and in each state was the FBI Uniform Crime Reports, Annual State Population Data from 1987 through 2015.

How We Calculated Rates

We calculated the yearly murder rate per 100,000 population for each state by dividing the number of murders in that state during that year by the population of that state that year and multiplying by 100,000.

We calculated the murder rates for 100,000 population each year for each of our categories of states—Death-Penalty States ("DPS"), Non-Death-Penalty States ("NDPS"), and Transitional States by adding the number of murders for all of the states in each category that occurred during a given year and dividing it by the combined population of all of the states in that category that year, then multiplying the result by 100,000.

For the average homicide rates for each state and each category, we totaled all of the murders in the state or category of states over the 29-year period and divided by 29 to obtain the average number of murders. We then totaled the population numbers for each of the 29 years to obtain the average population. We then divided the average number of murders by the average population and multiplied by 100,000 to determine the average murder rate per 100,000 population for the state or category of states.

We followed the same procedure in calculating the rates at which law enforcement officers were feloniously killed in the line of duty for each state and each category of states. Because the numbers of murders in which law enforcement officers were victims was so small, we would have introduced mathematic errors had we rounded their totals and calculated them based upon murders per 100,000 population. To avoid rounding errors, we calculated the numbers as murders per 10,000,000 population. Then, because people find it easier to conceive of comparisons at the 1 million level rather than at the 10 million level, we moved the decimal point one spot to the left so the numbers reflect the number of murders of law enforcement per 1,000,000 population.


[1] Nebraska arguably could fall in the Transitional State category, but because its legislative abolition of the death penalty was suspended and then overturned by voter referenda, it never actually abolished capital punishment in the state. We have therefore included it as a Death-Penalty State.


The Numbers


The numbers for each state and the calculations for each category of state are contained in a series of Microsoft Excel spreadsheets. You can download the spreadsheets containing the data on which the study is based, and our calculations from that data, here: https://deathpenaltyinfo.org/files/spreasheet/AppendixITables.xlsx


Here are the numbers that reflect the numbers over the 29-year-period, 1987 to 2015.  Where we have calculated murder rates, we have shown the raw numbers approximately 14 decimal places to the right so that you can use these numbers for your own calculations without worrying about rounding errors. Or, you can take the data from the spreadsheets themselves. 

Total # of Murders

Total # of Murders, United States, 1987-2015 = 526,064
Total # of Murders, Death Penalty States, 1987-2015 = 374,814
Total # of Murders, Non-Death Penalty States, 1987-2015 = 51,528
Total # of Murders, Transitional States, 1987-2015 = 99,722

Source: FBI Uniform Crime Reports, Annual Murder Data
 

Total # of Police Officers Killed
Total # of Police Officers Killed, United States, 1987-2015 = 1,599
Total # of Police Officers Killed, Death Penalty States, 1987-2015 = 1,227
Total # of Police Officers Killed, Non-Death Penalty States, 1987-2015 = 171
Total # of Police Officers Killed, Transitional States, 1987-2015 = 201

Source: FBI Uniform Crime Reports, Law Enforcement Officers Killed & Assaulted, Officers Feloniously Killed Annual Data (LEOKA reports, 1987 through 2015)*

 

Average Population from 1987–2015

Average Population, United States, 1987-2015 = 282,401,655.034
Average Population, Death Penalty States, 1987-2015 = 194,480,539.034
Average Population, Non-Death Penalty States, 1987-2015 = 37,108,178.862
Average Population, Transitional States, 1987-2015 = 50,812,937.138

Source: FBI Uniform Crime Reports, Annual State Population Data

 

Murder Rate/100K Population

Murder Rate, United States, 1987-2015 = 6.424 (6.42352394458151)
Murder Rate, Death Penalty States, 1987-2015 = 6.646 (6.64571414385248)
Murder Rate, Non-Death Penalty States, 1987-2015 = 4.788 (4.78823709676339)
Murder Rate, Transitional States, 1987-2015 = 6.767 (6.76735069621765)

Murder Rate, DP States1.03 times higher than national; 1.39 times higher than NDP; 1.02 times lower than Transitional
Murder Rate, NDP States1.34 times lower than national; 1.39 times lower than DP; 1.41 times lower than Transitional
Murder Rate, Transitional States1.05 times higher than national; 1.02 times higher than DP; 1.41 times higher than NDP


Police Officer Murder Rate/1M Population

Police Officer Murder Rate, United States, 1987-2015 = 0.195 (0.195246486879654)
Police Officer Murder Rate, Death Penalty States, 1987-2015 = 0.218 (0.217555674401356)
Police Officer Murder Rate, Non-Death Penalty States, 1987-2015 = 0.159 (0.158901673565157)
Police Officer Murder Rate, Transitional States, 1987-2015 = 0.136 (0.136402949192731)

Police Officer Murder Rate, DP States: 1.11 times higher than national; 1.37 times higher than NDP; 1.59 times higher than Transitional
Police Officer Murder Rate, NDP States1.23 times lower than national; 1.37 times lower than DP; 1.16 times higher than Transitional)
Police Officer Murder Rate, Transitional States: 1.43 times lower than national; 1.59 times lower than DP; 1.16 times lower than NDP

 

Percentage of Homicides With Officers as Victims

Officer Percentage of Homicides, United States, 1987-2015 = 0.304% (0.30395541226923%)
Officer Percentage of Homicides, Death Penalty States, 1987-2015 = 0.327% (0.32736237173638%)
Officer Percentage of Homicides, Non-Death Penalty States, 1987-2015 = 0.332% (0.33185840707965%)
Officer Percentage of Homicides, Transitional States, 1987-2015 = 0.202% (0.20156033773891%)
 

Officer Percentage of Homicides, DP States:  1.08 times higher than national; 1.01 times lower than NDP; 1.62 times higher than Transitional
Officer Percentage of Homicides, NDP States:  1.09 times higher than national; 1.01 times higher than DP; 1.65 times higher than Transitional
Officer Percentage of Homicides, Transitional States1.51 times lower than national; 1.62 times lower than DP; 1.64 times lower than NDP


* See FBI Uniform Crime Reports, Law Enforcement Officers Killed and Assaulted, 2001, n.1 (2001) (“Because a catastrophe such as the September 11 attacks falls far outside the normal course of police experience, the FBI has not included those fatalities in the 2001 rate, trend, or disposition tables for to do so would skew the data and render analyses meaningless.”), https://ucr.fbi.gov/leoka/2001