Abstract
Objectives
This study aimed to compare determinants of firearm purchasing related to the pandemic.
Study design
This was a cross-sectional survey.
Methods
A total of 3853 online panel participants completed a survey between December 22, 2020, and January 2, 2021, to approximate a nationally representative sample of US adults (aged ≥18 years). Four firearm ownership groups were created: non-owners, a proxy for first-time COVID-19 owners, prepandemic owners with COVID-19 purchase, and prepandemic owners without COVID-19 purchase. Explanatory variables were in four domains: demographics, concern about the pandemic, actions taken in response to COVID-19, and emotional response to COVID-19. Multivariate analysis estimated the adjusted odds of the outcomes.
Results
Respondents were categorized as non-owners (n = 2440), pandemic-related purchasers with no other firearms (n = 257), pandemic-related purchasers with other firearms (n = 350), and those who did not purchase in response to the pandemic but have other firearms (n = 806). Multivariable logistic regression found that compared with non-owners, those who had firearms at home with no pandemic-related purchases are more likely to be male, live in rural settings, have higher income, and be Republican.
Conclusions
The results highlight the changing profile of American firearm owners and identify that those who purchased firearms for the first time (in response to the pandemic) should be the focus of tailored public health interventions, including provision of education about recommended firearm storage to reduce firearm violence, particularly because they are more likely to have children at home, and belong to demographic groups that may have less experience with firearm safety.
Keywords: Firearm purchasing, COVID-19 pandemic, Violence
Introduction
Gun ownership in the United States is higher than in any other country with an estimated 1.2 firearms per resident.1 During the COVID-19 pandemic, firearm purchasing is reported to have increased.2 Firearm sales increased to approximately 2.5 million in March 2020, an estimated 85% increase compared with March 2019.3 This increase may be because of fear and uncertainty related to the COVID-19 pandemic and stay-at-home orders enacted in some jurisdictions.4 Similarly, firearm-related injuries and deaths increased during the pandemic, and some of these injuries, for example, domestic violence-related firearm injuries, may be associated with the increase in firearm sales.5 , 6 One study reported that the primary motivations for firearm purchasing during the pandemic included protection against people, fear of an increase in crime, and concerns about supply chain disruptions, healthcare access, and the economy.7 Another study found that compared with individuals not planning to purchase a firearm, those who were planning a purchase in the next 12 months (during the pandemic, July 2020 to June 2021) had higher economic concerns, fears of infection, and stress, as well as less tolerance for uncertainty.8
Before the COVID-19 pandemic, those who purchased firearms were more likely to be male, older, Republican, and live in more rural areas.9, 10, 11, 12, 13, 14, 15, 16 A study conducted in 2014 reported a weak relationship between firearm ownership and fear of victimization.10
We hypothesized that individuals purchasing firearms in the United States during the pandemic may be more likely to engage in pandemic-related behaviors prompted by fear and may have a higher negative emotional response to the COVID-19 pandemic compared with individuals who did not purchase firearms during the pandemic.
Methods
Study design
We approximated a nationally representative sample of US adults (aged ≥18 years) with a stratified non-probability sample to survey 3853 online panel participants between December 22, 2020, and January 2, 2021. Respondents were recruited by Climate Nexus Polling using several market research panels as described elsewhere.17 Participants were recruited using stratified sampling methods.17 Compensation for participants depended on the specific market research panel and respondents' preferences (e.g. cash, gift cards, reward points). Quotas were set to match the US Census Bureau's Voting and Registration Supplement to the Current Population Survey parameters for age, gender, race, educational attainment, census region, and Hispanic ethnicity. Sampling weights were used to account for small deviations from the preselected census parameters.
Outcomes
We created four firearm ownership groups from the questions, “Which, if any, of the following actions have you taken because of the spread of the coronavirus? – Purchased a gun and/or ammunition” and “Do you or anyone in your household own a gun?” as others have.7 Groups were as follows: (1) did not purchase in response to the pandemic and has no other firearms at home (‘non-owners’), (2) purchased in response to the pandemic and has no other firearms at home (a proxy for ‘first-time COVID-19 owners’), (3) purchased in response to the pandemic and has other firearms at home (a proxy for ‘prepandemic owners with COVID-19 purchase’), and (4) did not purchase in response to the pandemic and has firearms at home (a proxy for ‘prepandemic owners without COVID-19 purchase’).
Determinants
Explanatory variables were in four domains: demographics, concern about the pandemic, actions taken in response to COVID-19, and emotional response to COVID-19. Demographic determinants included gender, age, education, race/ethnicity, residence, income, and political affiliation. Other determinants included belief that the government does too much for its citizens, level of trust in the Centers for Disease Control and Prevention (CDC), actions taken in response to the pandemic (staying home, stocking up on items, keeping children at home), and worry about the impact of the pandemic on sickness, housing, crime, and martial law.
A variable that measured emotional response to the pandemic was created using the responses to the question, “How strongly do you feel each of the following emotions when you think about the coronavirus?” The 13 emotions asked about were afraid, helpless, angry, hopeful, disgusted, anxious, depressed, confused, panicked, lonely, overwhelmed, bored, and compassionate. Responses were on a 4-point scale (not at all, a little, moderately, and very). Typologies of emotional response to COVID-19 were identified by latent profile analysis based on participants' responses to the 13 emotions using common methods.25 Four emotional response profiles emerged that primarily reflected differing intensities of negative emotions and were categorized as follows: no response, mild response, moderate response, and strong response.
Statistical analysis
We conducted bivariate analyses between the outcomes and each of the potential determinants and between pairs of determinants to assess their association and multicollinearity. None of the independent variables were highly correlated with each other defined as a correlation >0.7 for our purposes. We included variables associated with the outcomes at a P-value <0.05 in the bivariate analysis or reported in the literature in the final models.
Multivariable logistic regression estimated adjusted odds ratios and 95% confidence intervals between the outcomes and potential determinants. STATA version 16 (StataCorp, LP, College Station, TX) was used.
Five models were constructed to compare populations based on firearm ownership status, as follows: Model 1: prepandemic firearm owners with no COVID-19 purchase and non-owners; Model 2: first-time COVID-19 owners and non-owners; Model 3: first-time COVID-19 owners and prepandemic owners without COVID-19 purchase; Model 4: prepandemic owners with COVID-19 purchase and prepandemic owners without COVID-19 purchase; and Model 5: first-time COVID-19 owners and prepandemic owners with COVID-19 purchase.
This project was considered exempt by the George Mason University Institutional Review Board (IRB 1684418-3).
Results
The survey participation rate was 68.5%, which is considered acceptable by the American Association of Public Opinion Research standards. More than half of the respondents (n = 2440, 63.3%) were non-owners. Of the remaining 36.7% (n = 1413) reporting owning firearms, 57.0% (806/1413) were prepandemic owners who did not purchase additional firearms in response to COVID-19, 24.8% (350/1413) were prepandemic owners who purchased a firearm in response to COVID-19, and 18.2% (257/1413) were first-time owners (who purchased a firearm in response to COVID-19; Table 1 ).
Table 1.
Participant demographics, beliefs, actions, and concerns about COVID-19 by firearm ownership category.
| Variable | Entire study |
No ownership |
First-time (COVID-19) owner |
Prepandemic owners without COVID-19 purchase |
Prepandemic owners with COVID-19 purchase |
|---|---|---|---|---|---|
| n (weighted %) | n (%) | n (%) | n (%) | n (%) | |
| Age | 2440 | 257 | 806 | 350 | |
| 18–29 | 750 (20.7) | 440 (18.0) | 80 (31.1) | 157 (19.5) | 73 (20.9) |
| 30–49 | 1397 (35.9) | 822 (33.7) | 143 (55.6) | 255 (31.6) | 177 (50.6) |
| 50–64 | 899 (23.1) | 612 (25.1) | 23 (9.0) | 208 (25.8) | 56 (16.0) |
| ≥65 | 807 (20.4) | 566 (23.2) | 11 (4.3) | 186 (23.1) | 44 (12.6) |
| Gender | |||||
| Female | 2016 (52.3) | 1365 (55.9) | 105 (40.9) | 407 (50.5) | 139 (39.7) |
| Male | 1837 (47.7) | 1075 (44.1) | 152 (59.1) | 399 (49.5) | 211 (60.3) |
| Race/ethnicity | |||||
| Non-Hispanic White | 2835 (70.0) | 1785 (73.2) | 163 (63.4) | 638 (79.2) | 249 (71.1) |
| Non-Hispanic Black | 466 (10.7) | 301 (12.3) | 43 (16.7) | 79 (9.8) | 43 (12.3) |
| Hispanic | 361 (15.9) | 212 (8.7) | 38 (14.8) | 65 (8.1) | 46 (13.1) |
| Other | 191 (3.6) | 142 (5.8) | 13 (5.1) | 24 (3.0) | 12 (3.4) |
| Education | |||||
| Less than high school | 241 (6.7) | 164 (6.7) | 21 (8.2) | 39 (4.8) | 17 (4.9) |
| High school graduate | 961 (25.9) | 657 (26.9) | 40 (15.6) | 209 (25.9) | 55 (15.7) |
| Some college | 1345 (36.1) | 834 (34.2) | 61 (23.7) | 311 (38.6) | 139 (39.7) |
| Bachelor's or higher | 1306 (31.3) | 785 (32.2) | 135 (52.5) | 247 (30.7) | 139 (39.7) |
| Income | |||||
| <$20,000 | 808 (21.6) | 575 (23.6) | 55 (21.4) | 125 (15.5) | 53 (15.1) |
| $20,000–$49,999 | 1313 (34.1) | 894 (36.6) | 48 (18.7) | 283 (35.1) | 88 (25.1) |
| $50,000–$99,999 | 1065 (27.4) | 619 (25.4) | 68 (26.5) | 255 (31.6) | 123 (35.1) |
| ≥$100,000 | 667 (16.9) | 352 (14.4) | 86 (33.5) | 143 (17.7) | 86 (24.6) |
| Type of residence | |||||
| Urban | 1219 (32.4) | 759 (31.1) | 137 (53.5) | 203 (25.2) | 120 (34.4) |
| Semiurban | 1748 (45.1) | 1184 (48.5) | 82 (31.9) | 345 (42.8) | 137 (39.1) |
| Rural | 886 (22.4) | 497 (20.4) | 38 (14.8) | 258 (32.0) | 93 (26.6) |
| Employment | |||||
| Self or other employment | 1955 (50.6) | 1107 (45.4) | 194 (75.5) | 404 (50.1) | 250 (71.4) |
| Unemployed | 690 (18.2) | 484 (19.8) | 37 (14.4) | 138 (17.1) | 31 (8.9) |
| Retired | 868 (22.3) | 603 (24.7) | 10 (3.9) | 207 (25.7) | 48 (13.7) |
| Other not working | 340 (9.0) | 246 (10.1) | 16 (6.2) | 57 (7.1) | 21 (6.0) |
| Political affiliation | |||||
| Republican | 1400 (36.5) | 757 (31.0) | 81 (31.5) | 391 (48.5) | 171 (48.9) |
| Democrat | 1807 (46.7) | 1216 (49.8) | 143 (55.6) | 302 (37.5) | 146 (41.7) |
| Independent | 646 (16.9) | 467 (19.1) | 33 (12.8) | 113 (14.0) | 33 (9.4) |
| Believe that government does too much for citizens | |||||
| Strongly disagree | 969 (25.2) | 691 (28.3) | 47 (18.3) | 170 (21.1) | 61 (17.4) |
| Somewhat disagree | 1104 (28.8) | 769 (31.5) | 57 (22.2) | 213 (26.4) | 65 (18.6) |
| Somewhat agree | 1130 (29.2) | 671 (27.5) | 79 (30.7) | 274 (34.0) | 106 (30.3) |
| Strongly agree | 650 (16.9) | 309 (12.7) | 74 (28.8) | 149 (18.5) | 118 (33.7) |
| Level of trust in CDC | |||||
| Strongly distrust | 363 (9.6) | 199 (8.2) | 86 (33.5) | 232 (28.8) | 101 (28.9) |
| Somewhat distrust | 680 (18.2) | 413 (16.9) | 92 (35.8) | 338 (41.9) | 133 (38.0) |
| Somewhat trust | 1621 (41.7) | 1058 (43.4) | 46 (17.9) | 149 (18.5) | 72 (20.6) |
| Strongly trust | 1189 (30.5) | 770 (31.6) | 33 (12.8) | 87 (10.8) | 44 (12.6) |
| Action taken | |||||
| Stay home (vs not) | 2069 (53.4) | 1261 (51.7) | 202 (78.6) | 382 (47.4) | 224 (64.0) |
| Stocked up (vs not) | 2335 (60.4) | 1417 (58.1) | 210 (81.7) | 447 (55.5) | 261 (74.6) |
| Kept children home (vs not) | 1299 (34.2) | 672 (27.5) | 212 (82.5) | 212 (26.3) | 203 (58.0) |
| Worry (vs not very or not at all worry) | |||||
| Sickness | 2636 (68.1) | 1686 (69.1) | 165 (64.2) | 554 (68.7) | 231 (66.0) |
| Losing home | 1233 (33.0) | 697 (28.6) | 151 (58.8) | 229 (28.4) | 156 (44.6) |
| Increased crime | 2193 (57.1) | 1337 (54.8) | 174 (67.7) | 451 (56.0) | 231 (66.0) |
| Martial law | 1709 (45.1) | 1028 (42.1) | 157 (61.1) | 330 (40.9) | 194 (55.4) |
| Emotional response to COVID | |||||
| No response | 1268 (32.6) | 823 (33.7) | 58 (22.6) | 284 (35.2) | 103 (29.4) |
| Mild response | 1289 (33.4) | 861 (35.3) | 73 (28.4) | 256 (31.8) | 99 (28.3) |
| Moderate | 897 (23.6) | 566 (23.2) | 72 (28.0) | 189 (23.5) | 70 (20.0) |
| Strongest negative | 399 (10.3) | 190 (7.8) | 54 (21.0) | 77 (9.6) | 78 (22.3) |
Model 1: prepandemic firearm owners with no COVID-19 purchase and non-owners
When compared with non-owners, prepandemic owners without a COVID-19 purchase are significantly more likely to be male, have higher income, and live in rural areas. They are also more likely to identify as Republican, to strongly believe the government does too much for its citizens, and to report a high emotional response to the pandemic. There are no statistically significant differences between these two groups in actions taken in response to COVID-19, reasons for worry, and trust in the CDC (Table 2 ).
Table 2.
Multivariable logistic regression models assessing relationship between demographics, beliefs, actions, and concern about COVID-19 pandemic between firearm ownership categories.
| Variable | Model 1: prepandemic owners without COVID-19 purchase compared with non-owners (n = 3246) |
Model 2: first-time owner compared with non-owners (n = 2697) |
Model 3: first-time owner compared with prepandemic owners without COVID-19 purchase (n = 1063) |
Model 4: prepandemic owners with COVID-19 purchase compared with prepandemic owners without COVID-19 purchase (n = 1156) |
Model 5: first-time owner compared with prepandemic owners with COVID-19 purchase (n = 607) |
|---|---|---|---|---|---|
| Adjusted OR (95% CI) | Adjusted OR (95% CI) | Adjusted OR (95% CI) | Adjusted OR (95% CI) | Adjusted OR (95% CI) | |
| Age | |||||
| 18–29 | Reference | Reference | Reference | Reference | Reference |
| 30–49 | 0.70 (0.54, 0.90) | 0.62 (0.43, 0.91) | 0.75 (0.47, 1.20) | 1.12 (0.75, 1.68) | 0.58 (0.36, 0.93) |
| 50–64 | 0.74 (0.56, 0.98) | 0.27 (0.15, 0.48) | 0.41 (0.21, 0.81) | 0.59 (0.36, 0.97) | 0.52 (0.26, 1.06) |
| ≥65 | 0.64 (0.45, 0.92) | 0.32 (0.13, 0.79) | 0.43 (0.15, 1.20) | 0.69 (0.36, 1.35) | 0.60 (0.22, 1.66) |
| Gender | |||||
| Female | Reference | Reference | Reference | Reference | Reference |
| Male | 1.24 (1.05, 1.47) | 1.90 (1.38, 2.63) | 1.62 (1.09, 2.41) | 1.42 (1.05, 1.92) | 0.97 (0.65, 1.43) |
| Race/ethnicity | |||||
| Non-Hispanic White | Reference | Reference | Reference | Reference | Reference |
| Non-Hispanic Black | 1.01 (0.75, 1.35) | 1.25 (0.79, 1.97) | 1.50 (0.85, 2.66) | 1.27 (0.77, 2.08) | 0.89 (0.50, 1.58) |
| Hispanic | 0.98 (0.72, 1.35) | 1.14 (0.72, 1.81) | 1.23 (0.68, 2.20) | 1.25 (0.76, 2.04) | 0.66 (0.38, 1.17) |
| Other | 0.57 (0.36, 0.91) | 0.65 (0.33, 1.28) | 1.17 (0.48, 2.87) | 1.09 (0.49, 2.41) | 1.07 (0.42, 2.72) |
| Education | |||||
| Less than high school | Reference | Reference | Reference | Reference | Reference |
| High school graduate | 1.30 (0.87, 1.30) | 0.55 (0.29, 1.06) | 0.89 (0.40, 2.00) | 0.93 (0.45, 1.93) | 0.59 (0.25, 1.39) |
| Some college | 1.43 (0.96, 2.14) | 0.69 (0.0.37, 1.29) | 0.86 (0.39, 1.89) | 1.52 (0.75, 3.06) | 0.40 (0.18, 0.89) |
| Bachelor's or higher | 1.08 (0.71, 1.65) | 1.04 (0.54, 2.01) | 2.00 (0.86, 4.67) | 1.32 (0.63, 2.76) | 0.84 (0.37, 1.93) |
| Income | |||||
| Under $20,000 | Reference | Reference | Reference | Reference | Reference |
| $20,000–$49,999 | 1.37 (1.07, 1.77) | 0.51 (0.32, 0.83) | 0.34 (0.19, 0.60) | 0.67 (0.42, 1.08) | 0.55 (0.30, 1.00) |
| $50,000–$99,999 | 1.90 (1.44, 2.50) | 0.84 (0.51, 1.40) | 0.35 (0.18, 0.66) | 0.79 (0.49, 1.29) | 0.59 (0.31, 1.12) |
| ≥$100,000 | 2.16 (1.56, 2.99) | 1.10 (0.63, 1.90) | 0.42 (0.21, 0.86) | 0.78 (0.44, 1.39) | 0.82 (0.41, 1.63) |
| Type of residence | |||||
| Urban | Reference | Reference | Reference | Reference | Reference |
| Semiurban | 1.00 (0.81, 1.24) | 0.83 (0.59, 1.18) | 0.87 (0.56, 1.36) | 1.14 (0.79, 1.65) | 0.63 (0.41, 0.97) |
| Rural | 1.90 (1.49, 2.42) | 0.94 (0.58, 1.52) | 0.53 (0.30, 0.91) | 1.27 (0.84, 1.93) | 0.56 (0.33, 0.97) |
| Employment | |||||
| Self or other employment | Reference | Reference | Reference | Reference | Reference |
| Unemployed | 0.91 (0.71, 1.16) | 0.75 (0.48, 1.19) | 0.92 (0.53, 1.62) | 0.41 (0.25, 0.66) | 1.68 (0.90, 3.15) |
| Retired | 0.99 (0.74, 1.32) | 0.63 (0.27, 1.46) | 0.51 (0.20, 1.31) | 0.69 (0.40, 1.20) | 0.69 (0.26, 1.84) |
| Other not working | 0.76 (0.54, 1.08) | 1.01 (0.53, 1.92) | 1.68 (0.75, 3.80) | 0.95 (0.50, 1.82) | 1.58 (0.68, 3.66) |
| Political affiliation | |||||
| Republican | Reference | Reference | Reference | Reference | Reference |
| Democrat | 0.56 (0.46, 0.69) | 0.77 (0.53, 1.11) | 1.29 (0.84, 2.00) | 0.82 (0.57, 1.17) | 1.62 (1.05, 2.48) |
| Independent | 0.53 (0.41, 0.68) | 0.61 (0.37, 1.00) | 1.08 (0.59, 1.96) | 0.71 (0.44, 1.15) | 1.89 (1.01, 3.55) |
| Believe that government does too much for citizens | |||||
| Strongly disagree | Reference | Reference | Reference | Reference | Reference |
| Somewhat disagree | 0.97 (0.77, 1.24) | 1.08 (0.68, 1.72) | 1.01 (0.57, 1.80) | 0.90 (0.57, 1.44) | 1.32 (0.73, 2.41) |
| Somewhat agree | 1.30 (1.02, 1.65) | 1.18 (0.76, 1.83) | 1.24 (0.72, 2.15) | 1.21 (0.79, 1.87) | 1.07 (0.61, 1.88) |
| Strongly agree | 1.38 (1.04, 1.84) | 2.17 (1.35, 3.48) | 1.50 (0.83, 2.71) | 2.16 (1.37, 3.41) | 0.89 (0.50, 1.59) |
| Level of trust in CDC | |||||
| Strongly distrust | Reference | Reference | Reference | Reference | Reference |
| Somewhat distrust | 0.84 (0.60, 1.17) | 0.56 (0.31, 1.00) | 0.94 (0.46, 1.91) | 1.00 (0.59, 1.70) | 0.79 (0.40, 1.54) |
| Somewhat trust | 0.83 (0.61, 1.13) | 0.40 (0.24, 0.69) | 0.64 (0.33, 1.23) | 0.77 (0.47, 1.27) | 0.80 (0.43, 1.51) |
| Strongly trust | 0.87 (0.63, 1.21) | 0.40 (0.23, 0.70) | 0.47 (0.24, 0.92) | 0.54 (0.32, 0.93) | 0.88 (0.44, 1.77) |
| Action taken | |||||
| Stay home (vs not) | 0.91 (0.75, 1.09) | 1.86 (1.28, 2.71) | 2.59 (1.68, 4.00) | 1.62 (1.18, 2.22) | 1.43 (0.91, 2.23) |
| Stocked up (vs not) | 0.93 (0.78, 1.12) | 1.89 (1.27, 2.80) | 2.18 (1.37, 3.46) | 1.95 (1.41, 2.71) | 1.08 (0.67, 1.76) |
| Kept children home (vs not) | 0.92 (0.75, 1.13) | 5.83 (3.93, 8.65) | 6.72 (4.29, 10.55) | 2.56 (1.84, 3.56) | 2.28 (1.41, 3.71) |
| Worry (vs not very or not at all worry) | |||||
| Sickness | 1.15 (0.94, 1.41) | 0.49 (0.34, 0.71) | 0.45 (0.28, 0.70) | 0.72 (0.51, 1.02) | 0.81 (0.52, 1.27) |
| Losing home | 1.03 (0.84, 1.28) | 2.41 (1.70, 3.43) | 2.67 (1.70, 4.21) | 1.38 (0.96, 1.97) | 1.63 (1.03, 2.58) |
| Increased crime | 1.12 (0.93, 1.36) | 1.18 (0.80, 1.73) | 1.46 (0.92, 2.33) | 1.42 (1.01, 2.00) | 0.86 (0.53, 1.40) |
| Martial law | 0.94 (0.77, 1.14) | 1.16 (0.80, 1.67) | 1.09 (0.68, 1.74) | 1.11 (0.79, 1.57) | 0.93 (0.58, 1.50) |
| Emotional response to COVID | |||||
| No response | Reference | Reference | Reference | Reference | Reference |
| Mild response | 0.95 (0.76, 1.17) | 0.84 (0.55, 1.30) | 0.82 (0.48, 1.42) | 0.86 (0.58, 1.27) | 1.06 (0.61, 1.84) |
| Moderate | 1.11 (0.86, 1.43) | 0.93 (0.59, 1.48) | 0.63 (0.35, 1.11) | 0.60 (0.38, 0.94) | 1.13 (0.63, 2.01) |
| Strongest negative | 1.51 (1.07, 2.14) | 1.41 (0.83, 2.41) | 0.78 (0.39, 1.54) | 1.06 (0.63, 1.80) | 0.60 (0.32, 1.12) |
CI, confidence interval; OR, odds ratio.
Model 2: first-time COVID-19 owners and non-owners
Relative to non-owners, first-time firearm owners are significantly more likely to be younger and male, more likely to believe the government does too much for its citizens, and less likely to trust the CDC. They also have greater odds of staying home, stocking up on items, and keeping their children at home. In addition, they are also more likely to be worried about losing their home but less worried about their family members or themselves getting sick from COVID-19. There are no significant differences in these two groups by political affiliation, urban/rural residence, or educational status (Table 2).
Model 3: first-time COVID-19 owner and prepandemic owners without COVID-19 purchase
First-time owners are more likely to be female and have a higher income and also have greater odds of staying home, stocking up on items, and keeping their children at home. They are more worried about losing their home but less worried about their family members or themselves getting sick from COVID-19 and have significantly lower odds of reporting that they strongly trust the CDC. There are no statistically significant differences by political affiliation (Table 2).
Model 4: prepandemic owners with COVID-19 purchase and prepandemic owners without COVID-19 purchase
Relative to prepandemic owners without a COVID-19 purchase, prepandemic owners with a COVID-19 purchase are younger and less likely to be unemployed. There are no differences between the two by political affiliation, educational status, or income. They are also significantly more likely to report staying at home, stocking up on items, keeping their kids at home, and worrying about crime increasing and have significantly lower odds of trusting the CDC (Table 2).
Model 5: first-time COVID-19 owner and prepandemic owners with COVID-19 purchase
More first-time owners are Democrats compared with prepandemic owners with a COVID-19 purchase. More live in urban areas and are significantly more likely to report keeping their kids at home and worrying about losing their home (Table 2).
Discussion
We found that those who purchased firearms in response to the pandemic are different in terms of demographics compared with prepandemic purchasers. Before the pandemic, American firearm owners were more likely to be male and aged ≥65 years, live in a rural setting, and be Republican.18 We found similar associations with prior owners with no new pandemic purchase more likely to be male, live in rural settings, be Republican, and have higher income. In contrast, those with a pandemic-related purchase are younger and have lower income. Others have similarly reported that younger individuals (mean age range 34.7–40 years of age) are more likely to report plans to purchase a firearm in response to the pandemic19 , 20 and that pandemic purchasers are more likely to live in urban areas.20 Prior owners with a new COVID-19 purchase are similar in political affiliation compared with prior owners without a COVID-19 purchase; however, new firearm owners are more likely to be Democrat, which has not previously been reported.
One study of firearm purchasing during the pandemic found that firearm purchasers who had previously owned guns were similar to other firearm ownership groups in regards to key demographics, including age and whether they lived in urban or rural areas.21 However, another study that was similar to ours in regards to sampling frame and sample size (n = 2709) reported that those purchasing guns during the pandemic were more likely to be younger, male, less educated, recently unemployed, Republican, and residents of southern states.22 Importantly, that study's results for sex, unemployment, education, which region people lived in, and political leaning were similar to the results of studies conducted before the pandemic that looked at the demographics of firearm owners; however, the null patterns for race/ethnicity, household income, marital status, and the presence of children differed from that reported during prepandemic times.9, 10, 11, 12, 13, 14, 15, 16 The latter is similar to what we found, specifically, no differences by sex, race, or education were found in our adjusted models.
Pandemic-related behaviors and worries vary by firearm ownership. Prior owners with a pandemic purchase are more likely than those without one to report worry about crime increasing, similar to what has been reported previously.7 , 20 Our findings parallel those from another study finding that pandemic firearm purchasers are more worried about the economy than non-purchasers.19 We detected no association between pandemic firearm purchasing and worry about COVID-19 illness, in contrast to others reporting individuals with a pandemic-related purchase had increased fear about the danger of COVID-19 and fear of “contamination.”19 Prior owners with no pandemic purchase are more likely to report a negative emotional response to the pandemic compared with non-owners, consistent with one other study that found that individuals who already own firearms (prepandemic) had an increased fear of COVID-19 and increase the level of threat perception.19
Those with a pandemic purchase (regardless of prior ownership) have a greater odds of reporting stockpiling behavior, and this was also reported by others.7 We also found keeping children at home and staying home in response to the pandemic are associated with pandemic purchasing among both prior owners and new owners, and similar associations were reported by others.7 , 20 This is of concern because having children and firearms present in the same household is linked to significantly greater risk of firearm-related death.23 Another study reported that 39% of those purchasing firearms for the first time during the pandemic reported suboptimal firearm storage.7 Health education interventions that address recommended firearm storage may be particularly helpful for new firearm owners and for firearm owners who live with children and others.
A number of limitations are inherent in quota-based non-probability methods. First, sampling is based on the individual's propensity to respond. Second, a theoretical basis for generalizing to a source population similar to what is found for a probability-based survey does not exist. However, these methods are common in policy-related research and public polling because non-response (and non-response bias) in probability-based sampling has become extremely high. In fact, when best practices are used, non-probability–based survey results can outperform those from probability-based surveys.24 The emotional response variable was created using the question, “when you think of the pandemic” broadly, and did not assess mental health or cognitive (e.g. threat assessment) components that could provide a more holistic view of how emotion “shows up” through actions. Firearm purchasing categories were created using two questions similar to how others have categorized this,7 and misclassification could have occurred.
Nevertheless, this study highlights differences between those who purchase firearms in response to the pandemic by prior firearm ownership. Firearm purchasing increased significantly during the pandemic. Understanding the changing demographic of who is purchasing firearms can prepare public health, law enforcement, and other responders for a possible surge in firearm-related societal impacts.
Author statements
Acknowledgments
The authors wish to thank our colleagues at Climate Nexus for their support collecting data, including James Wyatt, Molly Fisch-Friedman, and Reece Rushing.
Ethical approval
This project was considered exempt by the George Mason University Institutional Review Board (IRB 1684418-3).
Funding
N/A.
Competing interests
None reported.
Author contributions
A.A.R. had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis and contributed to concept and design and statistical analysis. All authors contributed to acquisition, analysis, or interpretation of data, drafting of the article, and critical revision of the article for important intellectual content.
References
- 1.Karp A. 2018. Estimating global civilian-held firearms numbers. [Google Scholar]
- 2.Miller M., Zhang W., Azrael D. Firearm purchasing during the COVID-19 pandemic: results from the 2021 National Firearms Survey. Ann Intern Med. 2022;175(2):219–225. doi: 10.7326/M21-3423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Brauer J. Small Arms Analytics & Forecasting; 2020. U.S. firearm sales: March 2020 unit sales show anticipated Covid-19-related boom. [Google Scholar]
- 4.Mannix R., Lee L.K., Fleegler E.W. Coronavirus disease 2019 (COVID-19) and firearms in the United States: will an epidemic of suicide follow? Ann Intern Med. 2020;173(3):228–229. doi: 10.7326/M20-1678. [DOI] [PubMed] [Google Scholar]
- 5.Ahmad F., Cisewski J. National Center for Health Statistics, National Vital Statistics System, Vital Statistics Rapid Release Program; 2021. Quarterly provisional estimates for selected indicators of mortality, 2019-Quarter 1, 2021.https://www.cdc.gov/nchs/nvss/vsrr/mortality-dashboard.htm# [Google Scholar]
- 6.Schleimer J.P., McCort C.D., Shev A.B., Pear V.A., Tomisch E., De Biasi A., et al. Firearm purchasing and firearm violence during the coronavirus pandemic in the United States: a crosssectional study. Inj Epidemiol. 2021;8(1):43. doi: 10.1186/s40621-021-00339-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lyons V.H., Haviland M.J., Azrael D., Adhia A., Bellenger M.A., Ellyson A., et al. Firearm purchasing and storage during the COVID-19 pandemic. Inj Prev. 2021;27(1):87–92. doi: 10.1136/injuryprev-2020-043872. [DOI] [PubMed] [Google Scholar]
- 8.Anestis M.D., Bond A.E., Daruwala S.E., Bandel S.L., Bryan C.J. Suicidal ideation among individuals who have purchased firearms during COVID-19. Am J Prev Med. 2021;60(3):311–317. doi: 10.1016/j.amepre.2020.10.013. [DOI] [PubMed] [Google Scholar]
- 9.Azrael D., Hepburn L., Hemenway D., Miller M. The stock and flow of US firearms: results from the 2015 National Firearms Survey. RSF Russell Sage Found J Soc Sci. 2017;3(5):38–57. [Google Scholar]
- 10.Dowd-Arrow B., Hill T.D., Burdette A.M. Gun ownership and fear. SSM-Popul Health. 2019;8 doi: 10.1016/j.ssmph.2019.100463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Goss K.A. The socialization of conflict and its limits: gender and gun politics in America. Soc Sci Q. 2017;98(2):455–470. [Google Scholar]
- 12.Hepburn L., Miller M., Azrael D., Hemenway D. The US gun stock: results from the 2004 national firearms survey. Inj Prev. 2007;13(1):15–19. doi: 10.1136/ip.2006.013607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kelley M.S., Ellison C.G. Who might buy a gun? Results from the Guns in American Life Survey. Sociol Inq. 2021;91(2):455–482. [Google Scholar]
- 14.Legault R.L. LFB Scholarly Pub.; 2008. Trends in American gun ownership. [Google Scholar]
- 15.Mencken F.C., Froese P. Gun culture in action. Soc Probl. 2019;66(1):3–27. [Google Scholar]
- 16.Smith T.W., Son J. Trends in gun ownership in the United States, 1972–2014. Chic IL NORC. 2015:1–9. [Google Scholar]
- 17.Goldberg M.H., Gustafson A., Maibach E.W., Ballew M.T., Bergquist P., Kotcher J., et al. Mask-wearing increased after a government recommendation: a natural experiment in the U.S. During the COVID-19 pandemic. Front Commun. 2020;5:44. doi: 10.3389/fcomm.2020.00044. [DOI] [Google Scholar]
- 18.Parker K., Menasce Horowitz J., Igielnik R., Oliphant J.B., Brown A. Pew Research Center; 2017. America's complex relationship with guns.https://www.pewresearch.org/social-trends/2017/06/22/the-demographics-of-gun-ownership/ [Google Scholar]
- 19.Anestis M.D., Bryan C.J. Threat perceptions and the intention to acquire firearms. J Psychiatr Res. 2021;133:113–118. doi: 10.1016/j.jpsychires.2020.12.033. [DOI] [PubMed] [Google Scholar]
- 20.Khubchandani J., Price J.H. Public perspectives on firearm sales in the United States during the COVID-19 pandemic. J Am Coll Emerg Physicians Open. 2021;2(1) doi: 10.1002/emp2.12293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Perlis R.H., Simonson M.D., Green J., Lin J., Safarpour A., Trujillo K.L., et al. Prevalence of firearm ownership among individuals with major depressive symptoms. JAMA Netw Open. 2022;5(3):e223245. doi: 10.1001/jamanetworkopen.2022.3245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Hill T.D., Wen M., Ellison C.G., Wu G., Dowd-Arrow B., Su D. Modeling recent gun purchases: a social epidemiology of the pandemic arms race. Prev Med Rep. 2021;24 doi: 10.1016/j.pmedr.2021.101634. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Goldstick J.E., Carter P.M., Cunningham R.M. Current epidemiological trends in firearm mortality in the United States. JAMA Psychiatry Chic Ill. 2021;78(3):241–242. doi: 10.1001/jamapsychiatry.2020.2986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Brown G., Low N., Franklyn R., Drever E., Mottram C. Quota Sampling - Government Statistical Service UK; 2017. GSR quota sampling guidance; p. 19.https://gss.civilservice.gov.uk/policy-store/gsr-quota-sampling-guidance/ [Google Scholar]
- 25.Nylund K.L., Asparouhov T., Muthén B.O. Deciding on the number of classes in latent class analysis and growth mixture modeling: a Monte Carlo Simulation Study. Struct Equ Model Multidiscip J. 2007;14(4):535–569. doi: 10.1080/10705510701575396. [DOI] [Google Scholar]