Scope
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
Scope can be subdivided into the types of events that are captured and
the populations covered. The scope of the NCVS, for example, is restricted
to nonfatal incidents and to the characteristics of crime victims rather than
offenders. Vital statistics and hospital-based information on firearm violence
is also limited to the victims. The UCR, by contrast, captures information
on both crime victims and offenders, but they are limited to offenses
that are known to and recorded by law enforcement agencies. The NCVS
includes data on both crimes reported to the police and those that victims
do not report. Household-based surveys such as the NCVS and the GSS are
limited to the population of persons with stable residences, thereby omitting
transients and other persons at high risk for firearm violence. Such
persons are included in the ADAM program, which collects information on
persons who come into contact with the criminal justice system.
Geographic coverage is another dimension of scope. The GSS, for example,
is representative of the United States and the nine census regions,
but it is too sparse geographically to support conclusions at finer levels of
geographical aggregation. This lack of individual-level data from small
geographical areas is a significant shortcoming in the firearms data. Presumably,
we would like to be able to make statements about, for example,
the probability that an individual commits suicide conditional on owning a
gun (or having one available) and other covariates. This cannot be done if
the smallest geographical unit that the data resolve is a multistate region.
Similar statements can be made about other forms of gun violence.
Perhaps no better illustration of the patchwork character of information
on firearms violence in the United States exists than the multiple and
nonoverlapping or partially overlapping coverage of the data sets. That
should come as little surprise, inasmuch as many of the data sets were
expressly intended to provide information about crime, violence, or injury
that was not available from other sources. The major impetus for the
development of the NCVS, for example, was to gather information on
crime incidents that do not come to the attention of law enforcement agencies.
The collection of information on violence from hospitals and emergency
departments is intended to reveal types of violence, such as partner
abuse, thought to be underreported in crime data sources.
The patchwork of existing data sources, in other words, has been created
with the best of intentions and has shed light on aspects of violence,
including the role of firearms, that otherwise would have remained hidden
from view, such as the burden on hospital emergency departments of firearm
injuries (Zawitz and Strom, 2000). However, insufficient attention has
been devoted to linkages across data in population coverage and the types
of firearm violence covered. Can data from the UCR, the NCVS, and emergency
departments be effectively linked to draw inferences about the firearms
violence in the population? As with data standardization, continuing
assessments of remaining gaps in the scope of firearms data should be part
of an ongoing program of methodological research on firearm violence.