Context

К оглавлению
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 
102 103 104 105 106 107 108 109 110 111 112 113 114 115 

An often-highlighted limitation of existing data on firearms is the lack

of detail regarding the context and circumstances of firearm violence. The

Supplemental Homicide Report provides limited information on the relationship

between victim and offender and event circumstances (e.g., whether

the homicide is related to an argument or the commission of another felony).

The National Incident-Based Reporting System extends such information

to other crime types, but it covers less than 20 percent of the population

more than 20 years after nationwide implementation began. Youth surveys,

such as Monitoring the Future (MTF) and the Youth Risk Behavior Surveillance

System, collect data on multiple attributes of respondents in addition

to firearm behaviors, but little information on the situations in which youth

carry and use firearms. The MTF survey also includes a longitudinal component

that tracks respondents over time. These panel data might be especially

useful for assessing firearms acquisition and use over time. However,

citing agreements with respondents regarding confidentiality, the University

of Michigan’s Institute for Social Research has not made these data

available to external researchers (see National Research Council, 2001).

The most promising emerging data source with respect to information on

the context and circumstances of firearm violence is the National Violent

Death Reporting System, which will compile individual-level data from

both criminal justice and public health sources on event circumstances, as

well as detailed descriptions of the weapons used in violence. The NVDRS

offers a model of a comprehensive data set that bridges existing data sources

on individuals, events, and weapons.

Standardization

An essential quality of any measurement system is the collection of

standard data elements from reporting units for purposes of reliable classification

and comparison. Good examples of standardized data sets for

measuring firearm violence are the FBI’s UCR program, the National Crime

Victimization Survey, and the mortality files available from the National

Vital Statistics System. Each of these data sets provides detailed formats

and instructions for data collection, coding, and entry to ensure standard

measurement of underlying data elements. For example, the UCR program

regularly compiles information on eight serious “index offenses” (murder

and nonnegligent manslaughter, rape, robbery, aggravated assault, burglary,

larceny, vehicle theft, and arson) and requires local law enforcement

agencies to use the same crime classification when compiling data on these

offenses for reporting to the UCR. The National Vital Statistics System

classifies deaths according to the International Classification of Diseases

codes for cause of death.

Such standard classification and coding schemes, however, are necessary

but not sufficient for ensuring valid and reliable measurement. Ultimately,

all data must rely on the faithfulness of their reporting units in

adhering to the standard protocols, which requires continuous monitoring

of data collection and adequate training of data entry personnel. All of the

federally sponsored data sets that collect information on firearm violence

have procedures in place to maintain standard data collection, although

they vary in the degree of compliance exhibited by reporting units. Generally

speaking, systems with direct control over reporting units are able to

maintain higher levels of standardization. The NCVS, administered by the

Census Bureau in cooperation with the Bureau of Justice Statistics, is a

good example of a data source with direct control over data collection. The

UCR, in contrast, has no direct control over local data collection and must

rely on data checks conducted by state UCR programs, as well as its own

quality controls, to ensure adherence to standard coding and classification

criteria. The National Vital Statistics System mortality series lie somewhere

between the NCVS and the UCR with respect to direct control over local

data collection.

We have limited our discussion thus far to standardization within data

sets. However, because data on firearms violence comes from multiple

sources and will continue to for the foreseeable future, we also must be

concerned with standardization of data elements between data sources.

Ongoing investigations of comparable data elements from different sources

should constitute an essential part of a program of methodological research

on firearm-related violence. Moreover, new and emerging data sets should

be designed to ensure transparent linkages of data elements with existing

data sources.

Two of the most important needs identified in public health and criminological

research on violence and other injuries are for the standardization

of data elements and the availability of detailed characteristics surrounding

each event. Several efforts under way to address these concerns, if successful,

may improve the usefulness and quality of data on firearm-related

deaths and injuries: the National Incident-Based Reporting System, the

National Violent Death Reporting System, the Data Elements for Emergency

Department Systems, and the International Classification of External

Cause of Injury coding system. The NIBRS and the NVDRS have been

discussed; the latter two systems are described below.

Data Elements for Emergency Department Systems: CDC’s National

Center for Injury Prevention and Control is coordinating an effort to develop

uniform specifications for data entered into emergency department

records. These specifications, known as DEEDS, are intended for use in 24-

hour, hospital-based emergency departments throughout the United States.

If the data definitions, coding conventions, and other recommended specifications

were widely adopted, incompatibilities between emergency departments

records would be substantially reduced. DEEDS does not specify an

essential or minimum data set, but is designed to foster greater uniformity

among individual data elements chosen for use. DEEDS also specifies standards

for electronic data interchange so that data can be accessed for research

purposes while maintaining confidentiality of patient records.

DEEDS was first released in 1997 for testing and review. Systematic field

studies, however, are still needed to assess the utility and practicality of the

system.

International Classification of External Causes of Injury: An international

effort, under the auspices of the World Health Organization, is currently

under way to develop a new classification system for coding external

causes of injury in mortality and morbidity systems. This system, known as

the International Classification of External Causes of Injury (ICECI) is

designed to capture details about the place of occurrence, activity at time of

injury, alcohol and drug involvement, objects or substances involved, intent

of injury, and mechanism of injury (e.g., firearms). Specific modules that

focus on injuries related to violence, transportation, sports, and work are

also under development. The first draft was released in 1998; the present

version, ICECI 1.0, was released in 2001. A number of shortened versions

have been tested for use as injury surveillance tools in places with limited

resources for surveillance. CDC has tested its own short version as a means

for capturing external cause of injury information from hospital emergency

departments records in the United States with promising results. The European

Union is also testing portions of ICECI as part of its efforts to create

a minimum data set on injuries. ICECI is designed to replace the International

Classification of Diseases coding system, which is thought to lack the

scope and specificity needed to inform injury research. The present version

of ICECI is undergoing formal review at the World Health Organization.11

11Details about ICECI 1.0, including the data dictionary, are available at http://

www.iceci.org.