U.S. Department of Justice Office of Justice Programs Bureau of Justice Statistics Crime Data Crime Firearms Stolen during Household Burglaries and Other Property Crimes, 2005–2010 Lynn Langton, BJS Statistician October 2012, NCJ 239436 ----------------------------------------------------- This file is text only without graphics and many of the tables. A Zip archive of the tables in this report in spreadsheet format (.csv) and the full report including tables and graphics in .pdf format are available on BJS website at:http://bjs.gov/index.cfm?ty=pbdetail&iid=4534 -------------------------------------------------------- Victimizations involving the theft of firearms declined from 283,600 in 1994 to 145,300 in 2010 (figure 1). Overall, about 1.4 million guns, or an annual average of 232,400, were stolen during burglaries and other property crimes in the six-year period from 2005 through 2010. Of these stolen firearms, at least 80% (186,800) had not been recovered at the time of the National Crime Victimization Survey (NCVS) interview. The data in this report were drawn from the Bureau of Justice Statistics’ (BJS) NCVS, which annually collects information on nonfatal victimizations reported and not reported to the police against persons age 12 or older from a nationally representative sample of U.S. households. The NCVS collects data on criminal incidents for which theft or attempted theft is either a component of the victimization (i.e., robbery, personal larceny, burglary, motor vehicle theft, and other property theft) or could occur in connection with the victimization (i.e., rape or sexual assault). This report examines the theft of firearms in criminal victimizations, focusing on the rate, number, amount of loss, and recovery of guns taken during burglaries and other property crimes, which include motor vehicle theft and other theft. It presents information on how firearms may end up in the hands of persons to whom they do not belong. Trend estimates are based on two-year rolling averages centered on the most recent year (figure 1). For example, estimates reported for 2010 represent the average estimates for 2009 and 2010. This method improves the reliability and stability of estimate comparisons over time. For all tables in this report, aggregate data for the time from 2005 through 2010 are the focus. ************************************ Firearms were stolen in 2% of violent and 1% of property crimes involving theft from 2005 through 2010 ************************************ On average, firearms were stolen in an annual average of about 4% of the 2.4 million burglaries occurring each year, in 2% of the 529,200 robberies, and in less than 1% of the 13.6 million other crimes involving theft from 2005 through 2010 (table 1). Burglaries accounted for 58% of the 153,900 victimizations each year in which a gun was stolen, whereas robberies accounted for about 7% of the victimizations involving a gun theft. About 0.4% of thefts involved the theft of a gun, yet thefts accounted for about a third (33%) of the victimizations in which a gun was stolen. Overall, about 93% of gun thefts occurred during property crimes. Therefore, remainder of this report focuses on property crime. Between 1994 and 2010, no statistically significant change was observed in the percentage of completed burglaries or other property crimes that involved the theft of at least one firearm (figure 2). This may suggest that the overall decline in the number of victimizations involving gun theft was not due to a decline in the number of privately owned guns that could be stolen. ************************************ Handguns were the most common type of firearm stolen ************************************ At least one handgun was stolen in 63% of burglaries involving gun theft (table 2). In 19% of burglaries, a handgun was stolen along with another type of firearm. About 39% of burglaries involving gun theft resulted in the theft of multiple guns, compared to about 15% of other property crimes involving gun theft. An average of about three guns were stolen during burglaries involving the theft of multiple guns, and about two guns were stolen during other property crimes involving multiple gun thefts. Due to the greater percentage of victimizations involving more than one gun, as well as the greater number of victimizations involving stolen firearms each year, burglaries accounted for nearly three times as many stolen guns than did other property crimes. In the six-year period from 2005 through 2010, an average of about 172,000 guns were stolen during burglaries and 60,300 guns were stolen during other property crimes each year. A total of 1.4 million guns were stolen during the six-year period. ************************************ Property crimes involving only stolen firearms resulted in an average annual loss of $27 million ************************************ Each year from 2005 through 2010, households that experienced gun theft from burglaries or other property crimes lost a total of about $600 million on average from these crimes. The majority of the loss was from other items stolen along with firearms. Households that experienced the theft of a firearm and other items had a mean loss of $7,600 in burglaries and $4,700 in other property crimes (table 3). The mean loss when only one gun and nothing else was stolen was between $400 and $500 per incident. Among households that experienced burglaries and other property crimes in which a gun was the only type of item stolen, the total loss was about $27 million per year. ************************************ Nearly 90% of burglaries involving stolen firearms were reported to the police ************************************ From 2005 through 2010, 86% of burglaries and 75% of other property crimes involving a stolen firearm were reported to police (table 4). Households were more likely to report to the police burglaries or other property crimes involving a stolen gun than property crimes in which other items with comparable value were stolen (approximately $500 on average). Among burglary victimizations, a greater percentage of households reported the incident to the police when a handgun (91%), another type of firearm (76%), or both (94%) were stolen than when other items valued from $500 to $999 were stolen (62%). A slightly greater percentage of households reported the theft of one firearm and no other items stolen to the police (77%) (not shown in table) than other stolen items valued from $500 to $999 (62%). ************************************ About 4 of 5 firearms stolen during household property crimes were not recovered ************************************ In 83% of burglaries and 85% of other property crimes that involved a stolen firearm, none of the stolen guns had been recovered at the time of the NCVS interview (table 5). Assuming these guns were not recovered later, this amounts to an annual average of at least 135,000 unrecovered guns from burglaries and 51,800 unrecovered guns from other property crimes. Although the victimizations involving stolen firearms could have occurred from one day to up to six months before the NCVS interview, the amount of time that had elapsed made no significant difference in the percentage of households for which guns had not been recovered at the time of the interview (not shown in table). No items had been recovered at the time of the NCVS interview in a lower percentage of burglaries involving the theft of a firearm (83%) than in burglaries involving the theft of items other than guns (89%). Among other property crimes, there was no difference in the percentage of incidents from which no items were recovered regardless of whether the victimization involved theft of a gun or other items. In about 85% of other property crime victimizations, no items were recovered. ************************************ The majority of property crimes involving stolen firearms occurred in the South ************************************ An estimated 22% of burglaries involving a stolen firearm occurred in households comprised of one male adult with no children. In contrast, these households accounted for 13% of all households (table 6). Households comprised of one female adult with no children experienced 8% of the burglary victimizations in which a gun was stolen, while accounting for 16% of households nationwide. A greater percentage of households with two or more adult residents with children experienced gun theft during burglaries or other property crime victimizations than households with one male or female adult resident with children. Households with a white non-Hispanic head of household accounted for the majority of burglaries and other property crimes in which a gun was stolen. White non- Hispanics also accounted for the majority of U.S. households (71%). Households in the South were more likely than households in other regions to have experienced gun theft during burglaries or other property crimes. Households in the South accounted for 37% of all households in the U.S., but 56% of all burglaries and 59% of other property crimes involving the theft of a firearm. Similarly, a disproportionate percentage of households in rural areas experienced burglaries involving the theft of a gun (34%), compared to the overall percentage of households in rural areas (17%). ***************** Methodology ***************** ---------------- Survey coverage ---------------- The National Crime Victimization Survey (NCVS) is an annual data collection conducted by the U.S. Census Bureau for the Bureau of Justice Statistics (BJS). The NCVS is a self-report survey in which interviewed persons are asked about the number and characteristics of victimizations experienced during the prior six months. The NCVS collects information on nonfatal personal crimes (rape or sexual assault, robbery, aggravated and simple assault, and personal larceny) and property crimes (burglary, motor vehicle theft, and other theft) both reported and not reported to police. The NCVS is administered to persons age 12 or older from a nationally representative sample of households in the United States. The NCVS defines a household as a group of members who all reside at a sampled address. Persons are considered household members when the sampled address is their usual place of residence at the time of the interview and when they have no usual place of residence elsewhere. Once selected, households remain in the sample for three years, and eligible persons in these households are interviewed every six months for a total of seven interviews. New households rotate into the sample on an ongoing basis to replace outgoing households that have been in the sample for the three-year period. The sample includes persons living in group quarters, such as dormitories, rooming houses, and religious group dwellings, and excludes persons living in military barracks and institutional settings, such as correctional or hospital facilities, and the homeless. (For more detail, see the Survey Methodology in Criminal Victimization in the United States, 2008, NCJ 231173, BJS website, May 2011.) In 2010, about 41,000 households and 73,300 individuals age 12 or older were interviewed for the NCVS. Each household was interviewed twice during the year. The response rate was 92.3% of households and 87.5% of eligible individuals. From 2005 through 2010, the primary reference period for this data brief, a total of 835,000 persons from about 472,000 households were interviewed. This equates to an annual average of 139,000 persons age 12 or older in 79,000 households interviewed each year from 2005 through 2010. Victimizations that occurred outside of the U.S. were excluded from this report. From 2005 through 2010, about 1% of the total unweighted victimizations involving theft of a firearm occurred outside the U.S. and were excluded from the analyses. Also excluded were the smaller number of attempted burglaries and other thefts in which the perpetrator tried to steal a firearm. The NCVS is unable to measure whether safes, locks, alarms, or other target hardening devices were in place in the home. The use of anti-theft measures may vary by population demographics. ************************************ Weighting adjustments for estimating household victimization ************************************ Estimates in this report use data from the 1993 to 2010 NCVS data files. These files are weighted to produce annual estimates of victimization for persons age 12 or older living in U.S. households. Because the NCVS relies on a sample rather than a census of the entire U.S. population, weights are designed to inflate sample point estimates to known population totals and to compensate for survey nonresponse and other aspects of the sample design. The NCVS data files include both person and household weights. Person weights provide an estimate of the population represented by each person in the sample. Household weights provide an estimate of the total U.S. household population. Both household and person weights, after proper adjustment, are also used to form the denominator in calculations of crime rates. Victimization weights used in this analysis account for the number of persons present during an incident and for repeat victims of series incidents. The weight counts series incidents as the actual number of incidents reported by the victim, up to a maximum of 10 incidents. Series victimizations are similar in type but occur with such frequency that a victim is unable to recall the details of each individual event. Survey procedures allow NCVS interviewers to identify and classify these similar victimizations as series victimizations and to collect detailed information on only the most recent incident in the series. In 2010, about 3% of all victimizations were series incidents. Weighting series incidents as the number of incidents up to a maximum of 10 incidents produces more reliable estimates of crime levels, while the cap at 10 minimizes the effect of extreme outliers on the rates. Additional information on the series enumeration is detailed in the report Methods for Counting High Frequency Repeat Victimizations in the National Crime Victimization Survey, NCJ 237308, BJS website, April 2012. Trend estimates provided are based on two-year rolling averages centered on the most recent year. For example, estimates reported for 2010 represent the average estimate from 2009 through 2010. This method is used to smooth trend lines and improve the reliability of estimates by increasing the sample sizes for each annual average estimate. --------------------------- Standard error computations --------------------------- When national estimates are derived from a sample, as is the case with the NCVS, caution must be taken when comparing one estimate to another estimate or when comparing estimates over time. Although one estimate may be larger than another, estimates based on a sample have some degree of sampling error. The sampling error of an estimate depends on several factors, including the amount of variation in the responses, the size of the sample, and the size of the subgroup for which the estimate is computed. When the sampling error around the estimates is taken into consideration, the estimates that appear different may, in fact, not be statistically different. One measure of the sampling error associated with an estimate is the standard error. The standard error can vary from one estimate to the next. In general, for a given metric, an estimate with a smaller standard error provides a more reliable approximation of the true value than an estimate with a larger standard error. Estimates with relatively large standard errors are associated with less precision and reliability and should be interpreted with caution. In order to generate standard errors around estimates from the NCVS, the Census Bureau produces generalized variance function (GVF) parameters for BJS. The GVFs take into account aspects of the NCVS complex sample design and represent the curve fitted to a selection of individual standard errors based on the Jackknife Repeated Replication technique. The GVF parameters were used to generate standard errors for each point estimate (such as counts, percentages, and rates) in the report. In this report, BJS conducted tests to determine whether differences in estimated numbers and percentages were statistically significant once sampling error was taken into account. Using statistical programs developed specifically for the NCVS, all comparisons in the text were tested for significance. The primary test procedure used was Student’s t-statistic, which tests the difference between two sample estimates. To ensure that the observed differences between estimates were larger than might be expected due to sampling variation, the significance level was set at the 95% confidence level. Data users can use the estimates and the standard errors of the estimates provided in this report to generate a confidence interval around the estimate as a measure of the margin of error. The following example illustrates how standard errors can be used to generate confidence intervals: According to the NCVS, from 2005 through 2010, 86% of burglaries involving the theft of a firearm were reported to police (see table 4). Using the GVFs, BJS determined that the estimate has a standard error of 2.6% (see appendix table 5). A confidence interval around the estimate was generated by multiplying the standard errors by ±1.96 (the t-score of a normal, two-tailed distribution that excludes 2.5% at either end of the distribution). Thus, the confidence interval around the 86% estimate is equal to 86% ± 2.6% X 1.96 (or 80.9% to 91.1%). In other words, if different samples using the same procedures were taken from the U.S. population during the period from 2005 through 2010, 95% of the time the percentage of burglaries involving gun theft that were reported to police would fall between 80.9% and 91.1%. In this report, BJS also calculated a coefficient of variation (CV) for all estimates, representing the ratio of the standard error to the estimate. CVs provide a measure of reliability and a means to compare the precision of estimates across measures with differing levels or metrics. In cases where the CV was greater than 50%, or the unweighted sample had 10 or fewer cases, the estimate was noted with a “!” symbol (interpret data with caution; estimate is based on 10 or fewer sample cases, or the coefficient of variation exceeds 50%). Many of the variables examined in this report may be related to one another and to other variables not included in the analyses. Complex relationships among variables were not fully explored in this report and warrant more extensive analysis. Readers are cautioned not to draw causal inferences based on the results presented. ************************************ Appendix: External measures of gun ownership, gun stock, and gun theft ************************************ Gun theft in this report varies by demographic group. This variation is driven, in part, by the prevalence of gun ownership; however, the NCVS does not collect information on gun ownership. This appendix describes external measures of gun ownership, stock, and theft. One of the limitations of this report is that NCVS data are not aligned with these external measures to the extent that rates of firearm theft can be easily or reliably computed from a denominator of households with guns or the total number of guns in the United States. For example, although the NCVS shows a decline in the percentage of households experiencing firearm theft from 1994 to 2010, and the General Social Survey and Gallup poll both show declines in the percentage of households owning firearms during the same period, when these data were combined to generate trends in the rate of firearm theft among gun-owning households, no differences were detected in the rate in 1994 compared to 2010 (figure 3). This may be indicative of stability in the rate of gun theft among gun- owning households over time or it may be a function of the lack of precision due to the large standard errors associated with generating estimates from surveys with different sampling methodologies. Similarly, while the UCR theft measure provides context for these findings, due to methodological differences in the collection of data on firearm theft, direct comparison between NCVS and UCR measures of gun theft are not feasible. ------------------------- Household gun ownership ------------------------- Although a number of surveys have collected data on household gun ownership at the state level or at particular points in time,***Footnote 1 1See Cook, P.J., & Ludwig, J. (1997). Guns in America: Results of a Comprehensive National Survey on Firearms Ownership and Use. Washington, DC: Police Foundation; Centers for Disease Control and Prevention (2001– 04). Behavioral Risk Factor Surveillance System Survey Data. Atlanta, GA: Centers for Disease Control and Prevention.***there are two main sources of national data on long-term trends in household gun ownership: the General Social Survey (GSS) and the Gallup poll. ---------------------- General Social Survey ---------------------- The National Opinion Research Center (NORC) has administered the GSS since 1972 to collect data on the demographics, behaviors, and attitudes of the U.S. population. In 1993 and every other year from 1994 through 2010, the GSS included the question, “Do you happen to have in your home (IF HOUSE: or garage) any guns or revolvers?” According to the GSS, the percentage of households that reported having a gun in the home declined from 46% in 1993 to 32% in 2010.***Footnote 2 Smith, T.W., Marsden, P., Hout, M., & Kim, J. (2011). General Social Surveys, 1972–2010 [machine-readable data file]. Chicago, IL: National Opinion Research Center.*** Figure 3 uses GSS data to present trends in the rate of burglaries and other property crimes involving firearm theft per 1,000 gun-owning households. To compute the denominator for the rate, the percentage of GSS households that owned guns was applied to the total number of NCVS households for the year. The rate was then computed using two-year rolling averages. The GSS and NCVS standard errors were pooled to compute a standard error around the rate. Although the GSS shows a significant decline in the percentage of households that owned guns from 1994 through 2010, the differences in the rate of burglaries and other property crimes involving firearm theft per 1,000 households across the period did not test at the p<.1 level using the pooled standard errors. The GSS data were based on a full probability sample of persons age 18 or older living in noninstitutionalized arrangements in the U.S. Until 2004, the survey was administered to English-speaking persons only. Beginning in 2006, Spanish-speaking respondents were eligible. From 1993 to 2010, data on household gun ownership were collected from an average of about 1,500 respondents. For more information on the sampling and weighting of GSS data, see the GSS Codebook at http://publicdata.norc.org/GSS/DOCUMENTS/BOOK/GSS_Codebook.pdf. ------------- Gallup poll ------------- Gallup produces a national public opinion poll that dates back to 1935. Gallup frequently conducts polls of persons in U.S. households to measure opinions on gun possession rights and gun laws as well as household gun ownership. From 1993 to 2010, Gallup conducted 13 polls in which respondents were asked, “Do you have a gun in your home? Do you have a gun anywhere else on your property such as in your garage, barn, shed, or in your car or truck?” From December 1993 to October 2010, Gallup polls showed a decline in the percentage of households with guns on the property, from 54% to 41%.***Footnote 3 Gallup poll, Do you have a gun in your home? Do you have a gun anywhere else on your property such as in your garage, barn, shed or in your car or truck? [COMBINED RESPONSES] 1991–2011. More information is available at www.gallup.com/poll/1645/Guns.aspx.*** Figure 3 also uses Gallup data to show trends in the rate of burglaries and other property crimes involving firearm theft per 1,000 gun-owning households. The process of computing the denominator for the rate was the same as used with the GSS; the percentage of Gallup households that owned guns was applied to the total number of NCVS households for the year. The rate was then computed using two-year rolling averages. Gallup reports that survey results are accurate within a margin of error of ±4%, so a conservative standard error of 2.04 was applied to all estimates. The Gallup and NCVS standard errors were pooled to compute a standard error around the rate. As with the rate computed using the GSS percentages from 1994 through 2010, the differences in the rate of burglaries and other property crimes involving firearm theft per 1,000 Gallup households that owned guns did not test at the p<.1 level using the pooled standard errors. The Gallup data were based on adults age 18 or older with a landline or cellular telephone who were selected for the poll through a process of random-digit dialing. When a sampled household was contacted on a landline phone, Gallup pollsters requested an interview with the person age 18 or older who had the most recent birthday. Any person reached on a cellular phone was interviewed directly. For a standard survey, Gallup used a sample of between 1,000 and 1,500 persons. For more information on the Gallup polling methodology, see “How are polls conducted?” at http://www.gallup.com/poll/101872/How-does-Gallup-polling- work.aspx. ----------- Gun stock ----------- Bureau of Alcohol, Tobacco, Firearms and Explosives The Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) collects data on the number of firearms that are manufactured in, imported to, and exported from the United States each year. Measures of U.S. gun stock are sometimes computed by adding the number of manufactured guns to the number of imported guns and subtracting exported firearms to get a total for each year, and then summing across years to get a count of the total number of guns in circulation. In 2010, about 5.5 million guns were manufactured, about 2.8 million were imported, and about 240,000 were exported, giving a total of about 8.1 million new guns added to the existing gun stock in 2010.***Footnote 4 United States Department of Justice, Bureau of Alcohol, Tobacco, Firearms, and Explosives. (2012). Firearms Commerce in the United States: Annual Statistical Update. Available at http://www.atf.gov/publications/firearms/050412-firearms- commerce-in-the-us-annual-statistical-update-2012.pdf.*** Guns that were destroyed or otherwise removed from circulation were not taken into consideration in this count. Moreover, although the ATF counts exclude firearms manufactured for the U.S. military, they include firearms purchased by law enforcement agencies. Because of these limitations in using the ATF data to estimate the number of privately owned guns, the report does not include a rate of the number of stolen guns per 1,000 guns owned. ----------- Gun theft ----------- Federal Bureau of Investigation Through the Uniform Crime Reporting (UCR) Program, the Federal Bureau of Investigation (FBI) collects limited data on firearm theft and recovery. A supplemental UCR reporting form, which is optional for UCR participating agencies, collected monthly data from state and local law enforcement agencies on the aggregate dollar value of items stolen, by type of stolen item. The supplemental form also collected aggregate data on the value of items recovered by police. In addition to collecting information on items stolen and recovered from household burglaries and other property crimes, the supplemental form collected information on burglaries and other property crimes involving commercial establishments. From 2005 through 2010, out of the approximate 17,800 agencies that submitted UCR data annually, about 13,000 agencies submitted supplemental data that were eligible for inclusion in Crime in the United States.***Footnote 5 5For more information on the FBI’s Uniform Crime Reporting Program, see http://www.fbi.gov/about-us/cjis/ucr. UCR supplemental data on the value of firearms stolen and recovered for each year are found in Table 24, Property Stolen and Recovered.*** From 2005 through 2010, the FBI agencies that submitted supplemental data reported that an average of $122 million worth of firearms was stolen each year. During the same period, an average of $11 million worth of stolen firearms (or about 8.7% of the value of total stolen firearms) was recovered each year.***Footnote 6 FBI, Uniform Crime Reporting Program, Crime in the United States, 2005–2010.*** Because the NCVS estimates of monetary loss include the monetary value of any other items stolen along with firearms, the UCR and NCVS estimates of the monetary value of stolen firearms cannot be directly compared. The UCR data cannot be used to generate an estimate of the number of incidents that involved a gun theft because the value of different types of firearms varies and agencies only submit aggregate data on the total value of all firearms stolen during the reporting period. ******************************************************* The Bureau of Justice Statistics is the statistical agency of the U.S. Department of Justice. James P. Lynch is the director. This report was written by BJS statistician Lynn Langton. Matthew Durose verified the report. Morgan Young, Jill Thomas, and Brian Higgins (Lockheed Martin) edited the report, and Tina Dorsey and Morgan Young produced the report, under the supervision of Doris J. James. November 2012, NCJ 239436 ****************************************************** ****************************************************** Office of Justice Programs * Innovation * Partnerships * Safer Neighborhoods * http://www.ojp.usdoj.gov ****************************************************** ______________________ 10/26/12/JER/9:20pm ______________________