historical violence database
We have posted below the codebooks and templates for our worksheets on violent crimes (including homicides), violent deaths (excluding homicides), violent assaults (excluding homicides), and collective violence. We have also posted samples of preliminary worksheets and spreadsheets on homicides in Vermont, 1760-1815, and on firearms accidents in New Hampshire and Vermont, 1783-1824. We hope that these samples will give researchers an idea of how the Historical Violence Database is organized and of the kinds of data we hope to include. Please contact us with your criticisms or suggestions for improvement. We would like to make our database as useful and accessible as possible. We appreciate your help.
It is our goal to have a standardized worksheet in Microsoft WORD on each violent crime, violent death, or incident of collective violence, drawing on photographs of the original evidence or on research notes made available by scholars. These worksheets, which organize and interpret the information on each violent event, are available in four formats: for homicides, other violent deaths (suicides, accidents), non-lethal assaults (sexual assaults, attempted murders, arson, etc.), and collective violence.
The four standard worksheets have a similar layout. The most important page is the second--the "evidence" page. It contains the citations on each case, the notes that scholars have taken on particular documents, and (if available) links to JPG or PDF image-files of the original documents. The standard evidence page distinguishes among three sources of information on violent acts (legal records, newspapers, and other) and two sources on suspects and victims (census and genealogical). Contributors should feel free, however, to add or drop categories as necessary, so that their evidence pages are compatible with the sources available for the places and periods they study. It makes no sense to have a "newspaper" category for the medieval period, but it might make a great deal of sense to divide medieval legal records into several categories, such as inquests, case rolls, warrants, etc. The important thing is to make room for all kinds of evidence, so no information is lost. The goal of the project is to encourage the sharing and preservation of data, not to impose a single format for data collection.
Evidence pages will vary in length, according to the amount of evidence on each case. Sometimes, the only record of a violent assault, death or riot is an entry in a docket book. At other times, surviving records include an autopsy report, testimony from the inquests or preliminary hearing, photographs, maps of crime scenes, etc.
Images files of manuscript records, newspaper articles, and other primary sources are welcome. They can be inserted into the worksheets or linked and stored as separate image files. But research notes are also welcome, as are transcriptions of handwritten documents and searchable texts of printed materials. All can be included on the evidence page.
Once the evidence page is complete, the personal information on each victim and suspect is transcribed onto the third page of each worksheet (if inferred, the information is in brackets). The details of each violent event--date, location, weapon, etc.--are transcribed onto the first page of the worksheet, together with the details of any legal proceedings that ensued. The first page also includes the researcher's synopsis and analysis of the event and its legal consequences, and the researcher's comments on the reliability or ambiguity of the evidence.
Once the worksheets are complete, the data on their first and third pages are entered into EXCEL spreadsheets, so they can be analyzed statistically. These spreadsheets, like those used by law enforcement and public agencies today, will be either victim-based or suspect-based. The HVD's spreadsheets will easier to use than law enforcement spreadsheets, however, because they will be coded differently. Most spreadsheets from public agencies code non-numerical data numerically. The FBI's Supplemental Homicide Reports, for instance, use "20" to refer to a murder committed with a knife or cutting instrument (ax, ice pick, etc.) and "30" to refer to a murder committed with a blunt object (hammer, club, etc.). The spreadsheets in the HVD will spell out the name of the weapon, whether a "dirk knife" or a "rock." It will then assign the weapon to a non-numerical category ("sharp," "blunt," etc.) in a second, coded variable. Files with non-numerical codes require more storage space than numerically-coded spreadsheets, but they are more accessible to non-statisticians and less prone to data entry errors, and they give quantitative historians a chance to create their own categories from the raw data. Whether they love statistics or not, historians of violence are better served by having non-numerical data in non-numerical form.
The HVD spreadsheets will include the standard variables available in the FBI's Supplemental Homicide Reports and the NCHS's Vital Statistics. As funds become available, they will also include all the variables in the National Violent Death Reporting System. But the HVD spreadsheets will always make room for additional variables of interest to historians.
Worksheet Codebooks, Templates, and Samples
- Codebook for worksheets on homicides (December 2007 version)
- Template for worksheets on homicide (December 2007 version)
- Codebook for spreadsheets on homicide (May 2002 version)
- Sample of worksheets on homicide:
- Vermont homicides, 1760-1815 (December 2007 version)
- Sample of spreadsheets on homicide:
- Vermont homicides, 1760-1815 (May 2002 Version -- EXCEL)
- Codebook for worksheets on violent deaths (December 2007 version)
- Template for worksheets on violent deaths (December 2007 version)
- Samples of worksheets on violent deaths:
- Firearms accidents in New Hampshire, 1783-1824 (December 2007 version)
- Firearms accidents in Vermont, 1783-1824 (December 2007 version)
- Sample of spreadsheets on violent deaths:
- Firearms accidents in New Hampshire and Vermont, 1783-1824 (May 2002 Version -- EXCEL)
- Codebook for worksheets on violent assaults (December 2007 version)
- Template for worksheets on violent assaults (December 2007 version)