How To Use

This website gives police and the public easy-to-use access to two datasets maintained by the Federal Bureau of Investigation: the Uniform Crime Report from 1965 to the present and — more useful for investigators — the Supplementary Homicide Report from 1976 to the present. The Murder Accountability Project, using the Freedom of Information Act, has obtained data on more than 39,000 homicides that were not reported to the Justice Department. This means the information at www.murderdata.org is the most complete data on U.S. homicides available anywhere.

Homicide investigators may find this site useful as a quick and easy means to review historical cases within their own departments or to easily identify cases in other jurisdictions that may be of interest. It is possible to search cases sorted by geography, type of victim, method of killing and timeframe of killing. This site is especially useful in cases in which an offender is suspected of killing more than one victim. Possible additional victims may be identified by checking all available reports made to the FBI or obtained independently by the Murder Accountability Project.

To begin using these data, first go to www.murderdata.org and select the “Clearance Rates” tab. You will see two curves, one red and the other grey. These are the plots for the total number of homicides reported (which you see at the very top of the red line) and the total number of homicides that were cleared, represented by the top of the grey curve. The red zone in these displays represents the difference between the total number of murders and murder clearances — a display of the numbers of homicides that went unsolved.

Generally, the rate at which police departments solve homicides through arrest has been declining. Since 1980, more than 220,000 homicides have not been solved. In the mid 1960s, police routinely reported clearing more than 80 percent of murders. Today, a department is doing well if it solves 65 percent and dozens of major departments are solving less than half. You can select homicide data by state, by county and by individual departments using the dropdown menus.

Arizona's "murder curve"
It is also important to look at the overall shape of the curve. Arizona, as pictured in the chart to the right, has a murder curve quite different from the rest of the nation. For most of the nation, the total number of murders plateaued in the early to mid 1990s when rival gangs vied for control of the illegal  drug trade in their communities. Murders have declined since then for a variety of reasons, including declining incidence of fatal domestic violence, the general aging of the U.S. population and, perhaps, a benefit of the high incarceration rate for drug-related crime. But murders in Arizona have been increasing in recent years, probably because of Mexican-based drug activity. Yet the murder curves in California and Texas are roughly similar to the nation as a whole.

The MurderData site also allows you to examine individual cases or groups of cases that have been reported to the FBI through the Supplementary Homicide Report (SHR), an addendum to the UCR. To access SHR data, select the “Search Cases” tab. 

This valuable dataset includes the age, race, sex, ethnicity of victims as well as the weapon used and the circumstances surrounding the killing. Unfortunately, the murders recorded here are far more complete than the database operated by the FBI’s Violent Criminal Apprehension Program. (The Murder Accountability Project strongly encourages that unsolved homicides be entered into ViCAP by local police.) Unlike ViCAP, victims’ names are not reported to the SHR, which is unfortunate for easy identification of specific cases. But the month, year, victim’s demographic characteristics and the reporting agency are included, usually sufficient information to get an ID from local police or from press accounts.
Results of Seattle's Green River Killer, note bar chart
on the left showing two-thirds of cases were unsolved.
 

Police also report demographic information about offenders when the killer has been arrested. However, this information is only available as of the time the report was made. Arrests made after the SHR report was filed are rarely — or never — updated. For this reason, you should only use the UCR reports to study clearance rates. Between 5 percent and 10 percent of homicides reported as  unsolved (no offender identified) at the time of reporting to the SHR were solved later. 

Yet the SHR is very useful in comparing the approximate rate at which police solve crimes according to different kinds of victims and weapons used. For example, generally, killings of women are more likely to be reported as solved than are the killings of men.

This MurderData site also allows you to consider whether a serial killer may be active within a specific community. The large bar chart on the left side of the screen (under title "Was suspect identified at time of report") shows reports in the greater Seattle area for women who were strangled or who died of unknown causes — cases in which the bodies were so decomposed that the medical examiner was unable to make firm diagnoses as to the cause of death. As you can see, about two-thirds of these cases were unsolved at the time the report was made to the FBI. Most of these killings occurred in the mid 1980s and 1990s. Forty-eight of these women died at the hands of “Green River Killer” Gary Leon Ridgway, one of the most dangerous serial killers in U.S. history.

By careful use of the selection criteria, you can isolate subgroups of cases that are important to you. For example, if you wanted to download everything the FBI was told about Ridgway’s killings, select “Seattle” under the “MSA” selector, “Female” under “Victim’s Sex,” and both “Strangulation” and “Other or Type Unknown” under the “Weapon” selector. By “clicking” on areas within the chart, it is possible to select subgroups within the displayed data. For example, in the case of the Seattle-area killings, click on the red bar on the left hand side and you will have selected only cases that were unsolved at the time the SHR report was made. You also can click on the red elements of any individual year if you want to select only unsolved cases within a single year or group of years.

Once you’ve selected a set of specific cases in which you are interested, it is easy to download everything the FBI was told about these crimes when the SHR report was made. Click the “right” mouse button on the field or fields you’ve selected. A small dialogue box will pop up that looks like the image to the right. To start the download, select the small unnamed image block of tiny lines immediately to the right of the “Keep  Only” and “Exclude” options. This symbol indicates you want to pull down data in a spreadsheet.

What will pop up next will be a box that looks like this:

Select the “Underlying” tab and check the tiny blue box to “Show all columns” to download all data in the 142 cases of unsolved strangulation of women in the Seattle area. You will see a spreadsheet-like display of 32 different fields that look (in part) like this: 



By clicking on the “Download all rows as a text file” which is the blue hypertext line, you will start a download of all 142 cases in a format that can be copied-and-pasted into a spreadsheet. This gives you important information like the month and year of each killing, as well as everything known about the offender and possible circumstances involved in the case. 

This website can be helpful to homicide investigators who wish to test their theories about homicide suspects who may have killed over a long time within one jurisdiction or in several different jurisdictions. 

Let’s try a search for a very famous case: the killing of six-year-old JonBenet Ramsey, found strangled in the basement of her Boulder, Colo., home on December 26, 1996. It is a reasonable question to ask if there were similar killings in Colorado which might suggest the presence of a serial killer preying upon young girls.

Go to the “Search Cases” tab and locate the SHR report about JonBenet by selecting “Colorado” under state, “1996” by adjusting the year sliders, the exact age of “6” under victim’s age, “female” under victim’s sex, and “strangulation” under weapons. Only one case will pop up. To confirm that it’s JonBenet’s, right mouse click on the case, select the download option and call up the details. The “Underlying” data tab will show that this murder occurred in December of 1996 and was reported by the Boulder Municipal Police Department, confirming this is the correct record.

Were there similar killings in Colorado at the time of JonBenet’s murder? Try going back to the “Search Cases” page and widen the search parameters. Let’s assume that a serial killer might target young girls in the 5-10 age range. Add those additional ages to the “victim’s age” pulldown menu. Widen the years selector to search from 1980 to the present. You will see only two records, JonBenet’s and the rape/homicide of a 10-year-old girl in Colorado Springs committed in February of 1985. This case was also reported as unsolved at the time of the report. Investigators may wish to contact Colorado Springs authorities to check for commonalities. The 11-year gap between killings suggests a link is unlikely, perhaps, but not impossible.

Could JonBenet have been killed by a highly mobile offender who operated in other states? Go back to the state selector and choose “all.” What you see will look like this: 

By right clicking on the red bar under the “Was suspect identified at time of report” box, you will see there were 27 strangulations of young girls reported without an identification of the offender during the last 33 years. There were no unsolved cases in 1995 or 1997. JonBenet’s killing was the only such case reported nationwide in 1996. With a few mouse clicks, you can download all of the details of these 27 cases, in case you wish to confer with other departments who’ve experienced similar homicides.

But using this website, you have quickly tested the possibility that JonBenet Ramsey was killed by a serial killer. There was one other unsolved strangulation/rape homicide of a young girl in Colorado. A disproportionate number of such cases nationally occurred in Western states (or 11 of 27).