|
This file contains information on victims and circumstances of the 34,584 homicides investigated
by law enforcement agencies in California from 1990 to 1999. This information comes from the
Department of Justice (DOJ) Homicide File. The thing that makes the Linked Homicide File different
from the DOJ Homicide File is that the linked file contains additional information from DHS Vital
Statistics Death Record File on 32,163 of the 34,584 (93.0%) records. The two files were linked
using information common to both data sets (victim’s name and sex, date of injury and death,
county of homicide and death).
By linking these two files, we combine the strengths of law enforcement reporting and medical
reporting in one data set. The DOJ Homicide File contains more information on the suspect and
incident, whereas the Death Record File contains more information on the victim. The Linked Homicide
File is intended for researchers to study homicide and provide evidence and support for the
development of strategies for reducing this problem in California.
Additional Help
If these pages don't answer your question, or if you need help obtaining or
interpreting linked homicide data, e-mail EPIC.
Frequently Asked Questions about Linked Homicide Data (FAQ)
What is a linked file?
A linked file is one that is created from 2 or more separate data sets. These files have common
fields used to determine if a case is a match. A linked file might be set to retain only the cases
that linked or it might retain all cases from both files, whether they linked or not. Alternatively,
one could use one file as the base and then merge only those records that link from the second file.
This is what we did.
Why link?
Linkage allows the user to maximize the information in a data set by combining 2 or more separate
files. For instance, one file may contain a certain type of data while the other file contains
another type of information. Combining these two sources gives the user both these types of
information for the same record.
What data sources were used to create the Linked
Homicide File?
We linked selected death records from the California Department of Health Services with the
complete homicide file from the California Department of Justice (DOJ). Death records are compiled
from death certificates completed by either the attending physician at the time of death or, in the
case of sudden and unexpected deaths such as homicides, by the medical examiner or coroner in the
county in which the death occurred. The homicide file consists of Supplemental Homicide Reports,
completed by the investigating law enforcement agency when they determine that the manner of death
was homicide, as well as other information from Uniform Crime Reports (UCR) and news clippings.
How were the data sources linked?
We chose to use law enforcement data as the base because our experience suggests that this file
may be more comprehensive in ascertaining homicides. We are primarily interested in data relevant to
understanding and preventing fatal violence, such as the relationship between perpetrator and
victim, type of weapon used, and circumstances of the fatal assault. The homicide file contains more
detailed information about the homicide than does the death record file. Death records were linked
to the homicide file using Integrity TM probabilistic matching
software. Matching fields included victim’s social security number, first name, middle name, last
name, age, county of residence, and date of injury or incident.
Why don’t all records contain information from Vital
Statistics death records?
Not all records from the DOJ homicide file were able to be linked with death records. We linked
32,163 (93.0%) of the 34,584 homicide file records with matching death records. Thus, 2,421 records
in the Linked Homicide File only contain information from the DOJ homicide file.
Does the Linked Homicide File contain all homicides in
California?
Not necessarily. The file contains all homicides reported by law enforcement agencies in
California from 1990 through 1999. Homicides reported on Vital Statistics death records are included
in the file only when they link with a case in the DOJ homicide file. The number of homicides
reported by law enforcement agencies is typically slightly greater than the number of homicides
reported by coroners or medical examiners on death records. This variance is partly because death
records are recorded for California residents, whether they died in California or elsewhere, while
Supplemental Homicide Reports are filed for all homicides occurring in California, regardless of
resident status of the victim. Also, different agencies may have different criteria for determining
whether the death is in fact a homicide. Furthermore, a law enforcement agency may determine that a
homicide occurred and report it, but if human remains are not recovered there will be no death
certificate.
This file is useful for studying the circumstances and risk factors of homicide, not for
determining the official numbers of homicide in California. If reporting total numbers of homicide
in the file please keep in mind that this is based on law enforcement reporting and may conflict
with health reports. Also, do not count all homicides according to the linked death records because
not all death record homicides linked and consequently not all death record homicides are included
in the file.
Since there are more cases from the DOJ Homicide File than
from death records, does that mean law enforcement reporting is more complete?
It depends. A coroner’s determination of “homicide” is not identical to a prosecutor’s,
and is not always rendered at the same time. Vital Statistics homicides may include some “unintentional”
deaths “at the hands of another” that may not be considered a crime by the criminal justice
system, and would not have a SHR submitted. However, police may report a homicide and submit a SHR
while the coroner’s determination is still pending. A subsequent determination other than “homicide”
would not necessarily result in the removal of that SHR from the file.
Why did you use the DOJ Homicide File as the base rather
than death records?
We chose to use law enforcement data as the base because we were primarily interested in data
relevant to understanding and preventing fatal violence, such as the relationship between
perpetrator and victim, type of weapon used, and circumstances of the fatal assault. The homicide
file contains more detailed information about the homicide than does the death record file.
Why isn't the number of homicides in the Linked Homicide
File consistent with other data sources?
As described above, the Linked Homicide File is based on Supplemental Homicide Reports. Some
homicide statistics are based on death records. The two types of records differ in a couple of ways.
First, law enforcement agencies submit Supplemental Homicide Reports for any homicide occurring in
the state of California whether the victim was a resident of the state or not. Death records are
usually reported by the victim’s state of residence. So, absent from this file but probably
present in another would be California residents killed in another state or country. Another reason
for a slight discrepancy between the two data sources may be differences in the law enforcement and
death records definitions of homicide. In most cases these definitions are concordant, but not
always.
Which data source is more reliable? The demographics of
the victim are very consistent between the two data sources but may be more accurate in the death
records if the attending physician knew the victim well or the next of kin was asked to help fill
out the death certificate. In the homicide file the information is obtained in the course of the
investigation, not necessarily from people who knew the victim well. However, when looking at
information common to both sources, keep in mind that more records contain information from the
homicide file than from death records because of the way the file was linked.
Were the suspects in these cases convicted?
Offender data are based on police identification of the suspect as the offender. No follow-up is
made to determine if a conviction was obtained unless conviction information is included in the
original crime report. Suspects are not necessarily ranked according to their involvement in the
homicide. For instance, suspect #1 is not necessarily the primary suspect and Suspect #2 is not
necessarily the secondary suspect.
What are the limitations of the Linked Homicide File?
Although the Linked Homicide File provides more information than the individual files used to
create it, there are limitations to keep in mind when using it. First, not all homicide files were
linked to death records so when using fields originally from the death record file we can only look
at 32,163 records instead of the full 34,584. Second, although the demographics of the victims match
very well from each source there are some differences that the user should consider. As a general
rule, it may be better to use homicide file demographics, possibly sacrificing quality for quantity.
|
Publications and Directories
