1 Represents from the Inception of the Brady Act on March 1, 1994 to 1999.
2 For period January 1 to November 29, 1998.
3 For period November 30 to December 31, 1999. Counts are from the National Instant CriminalBackground Check System and may include multiple transactions for the same application.

Source: U.S. Bureau of Justice Statistics,Background Checks for Firearm Transfers, 1999, series NCJ 180882, June 2000.

http://www.ojp.usdoj.gov/bjs/guns.htm

*Methodology

Definitions

State instant approval (instant check) systems require a sellerto transmit a purchaser's application to a checking agency bytelephone or computer; the agency is required to respondimmediately or as soon as possible without delay.

Purchase permit systems require a prospective firearm purchaserto obtain, after background check, a government-issued document(called a permit, license, identification card, etc.) that mustbe presented to a seller to receive a firearm.

Exempt carry permit is a State carry permit (issued after abackground check) that exempts the holder from a check at thetime of purchase under an ATF ruling or State law.

Other type of approval systems require a seller to transmit apurchaser's application to a checking agency by mail, telephone,or computer; the agency is not required to respond immediatelybut must respond before the end of the statutory time limit.

Application for firearm transfer is information submitted by aperson to a State or local checking agency to purchase a firearmor obtain a permit that can be used for a purchase; includesinformation submitted directly to a checking agency or forwardedby a prospective seller.

Transactions are inquiries to the Federal NICS system and mayinclude more than one inquiry per application.

Rejection occurs when an applicant is prohibited from receivinga firearm or a permit that can be used to receive a firearm, dueto the finding of a disqualifying factor during a backgroundcheck.

Data collection procedures

The Regional Justice Information Service (REJIS), through acooperative agreement with BJS under the Firearm InquiryStatistics (FIST) program, collected the data.

The agencies supplied data on either paper or diskette. Severaldifferent forms were provided to meet the varying officeprocedures of the agencies. In addition REJIS wrote specialsoftware distributed free of charge to requesting agencies. This software was designed to simplify the record tabulatingfunctions of the agency. It also helped to reduce the burden ofkeeping the statistical data because a capability of thesoftware was to automatically report the data needed for thestudy. In all cases the data that the agency sent to REJIScontained only statistical information and would not allow theidentification of an individual. The software also assistsagencies in purging records after the delay time specified bylaw.

FIST data are collected directly from State agencies conductingbackground checks. Two approaches were used for the collectionof data from local checking agencies. The first involvedcontinuation of an ongoing survey. The ongoing survey includedagencies that have provided information consistently for theFIST survey since 1995. The second consisted of telephone andmail contact to obtain data from all other local checkingagencies that collected and would share information on firearmapplications.

The following presents the approach used to supplement theongoing survey among checking agencies. Telephone and mailcontact was attempted in 1999 with all local checking agenciesnot previously invited to participate in the ongoing survey.

Information collected included the following: firearmapplications made to the agency, firearm applications rejectedby the agency, and the reasons for rejection. Information onappeals of rejected applications was not included since localchecking agencies may not handle appeals through the entireprocess and may have only limited information on outcomes fromsuch appeals.

Determining populations

To estimate the application and rejection rates within a givenarea, the appropriate agency population was needed and wasdetermined as follows:

The stratification classification of the county was based on thesize of the largest city within the county.

If cities within a county were conducting their own backgroundchecks, their populations were subtracted from the countypopulation.

If a municipal agency provided services for other selectedmunicipalities, then populations for those municipalities wereadded to the populations of the reporting municipality.

If an agency participating in the study relied upon otherjurisdictions to conduct background checks, they were replacedby those other jurisdictions (for example, a town being replacedby a county).

State and local checking agencies were stratified by size of thepopulation served: State agencies that served an entire Statepopulation; local agencies that served a population greater than100,000; local agencies that served a population between 10,000and 100,000; and local agencies that served a population of lessthan 10,000. Population size was based on 1998 Census Bureauinformation. The population categories were chosen to beconsistent with those commonly used by the FBI when conductingsimilar studies.

All agencies serving a population greater than 100,000 wereasked to contribute data in 1999, either by continuing to reportin the ongoing survey or by providing a report in thesupplemental survey. The number of agencies in both the ongoingsurvey and the supplemental survey are shown by populationcategory in the table below.

Number of agencies Population served

Under 10,000 26210,000 to 100,000 211Over 100,000 37Statewide agencies (POCs and thoseissuing permits) 25

Total agencies 535

These agencies together served 192,394,762 people, or 71.18% ofthe more than 270 million people estimated to live in the U.S.in 1998. The addition of agencies did not skew the distributionof agencies toward any particular State or region.

In some States one statewide agency conducts background checksfor purchase and another agency (or division within an agency)issues ATF-approved permits. Although both agencies conductedbackground checks, care was taken not to count State populationstwice in the estimation process. This situation of dualagencies conducting background checks did not occur among localagencies.

Estimation procedures

Based on data provided by both sets of agencies, Nationalestimates were developed using population weighting factors. When an agency did not provide data for all months, a simplelinear extrapolation or interpolation was used to generate a12-month total.

It is important to emphasize that the mix of State and localagencies conducting background checks changed during thetransition from the interim to the permanent provisions of theBrady Act. Consequently, while data collection proceduresremained similar to those used in 1998, the agencies providingdata changed.

For example, when the permanent provisions of the Brady Actbecame effective, six agencies that had previously conductedstatewide checks turned their background checks over to the FBI.However, three States (Pennsylvania, Tennessee, and Vermont)that had not performed their background checks during theinterim period began statewide systems.

The distribution of local checking agencies also changed afterthe permanent provisions of the Brady Act became effective. Many States delegated their responsibilities to the FBI, solocal agencies in those States no longer conducted backgroundchecks. Further, it was deemed necessary to collect data fromlocal agencies (and some statewide agencies) that issue certaincarry permits approved by ATF or State law as an alternative toa presale background check.

In 11 States local agencies still conducted checks required byState law. Local agencies that had been reporting these data toFIST were asked to continue. However, not enough local agenciesremained in the ongoing survey to calculate a sufficientlyaccurate national estimate. Additional local checking agenciesconducting background checks for permits had to be included inthe estimation process. For these reasons, FIST supplementedparticipants in the ongoing survey with data from other checkingagencies.

Agencies with a rejection rate over four standard deviationsabove the average standard rejection rate were classified asoutliers and their data were not used for projection ofestimates. In addition, rejection rates that could not bedetermined with sufficient accuracy were not used.

The accuracy of the estimates presented in this report dependson two types of errors: nonsampling and sampling. In thisstudy, nonsampling error may occur from the following: nonresponse; differences in the ways checking agencies process,code, store, and retrieve their information; differences ininterpretation of the survey questions; and activities thatdelay personnel from doing paperwork.

*