Firearm violence is a significant public health issue in the United States, yet comprehensive data on firearm sales and ownership are lacking. Existing data sources, such as the National Instant Criminal Background Check System (NICS), have limitations. NICS data may overestimate or underestimate actual firearm sales due to various factors including permit denials, multiple checks for single purchases, waiting periods, exemptions from background checks, and lack of private seller data. These limitations highlight the need for alternative data sources. This study investigates the potential of using internet search query data as a complementary, readily available, and near real-time source for tracking firearm sales and ownership, enabling analysis at finer geographic and temporal scales. Previous research has shown the utility of search data in public health surveillance for various health issues, including influenza, abortion, smoking, and mental health. Some studies have explored gun-related internet searches in relation to mass shootings and background checks, demonstrating changes in search volume in response to these events. This current study expands on this prior work by analyzing the relationship between firearm-related search volumes and background checks, firearm-related mortality, and state-level firearm policies.

Prior research has utilized internet search data to track various public health trends and policy impacts. Studies have demonstrated the correlation between search patterns and events like influenza outbreaks (with Pearson’s r values ranging from 0.82 to 0.99), abortion rates (Spearman’s ρ of −0.48 and −0.55), and increases in search volume related to smoking and mental health. Previous work on gun-related internet searches has shown a correlation between searches for terms like “bury gun” and firearm background checks (Pearson’s r = 0.84) and changes in search volume following mass shootings, with increases ranging from tens to hundreds of percent. One recent study also noted increased gun-related searches during the COVID-19 pandemic. This current research builds upon these findings by providing a more comprehensive analysis of the relationship between gun-related search volumes and various firearm-related phenomena.

The study used data from multiple sources: the Federal Bureau of Investigation’s National Instant Criminal Background Check System (NICS) for background checks, the U.S. Census for population estimates, the Centers for Disease Control and Prevention for firearm-related death data, and The Giffords Law Center to Prevent Gun Violence for state-level firearm policy data. Google Trends provided internet search volume data. The researchers retrieved data on February 10, 2020, using the most recently available data for each analysis. They conducted correlation analyses (Pearson’s r and Spearman’s ρ) between various search terms (e.g., "gun," "shotgun," "9mm") and gun-related variables, including background checks (stratified by gun type), firearm-related deaths, and state-level firearm policy restrictiveness. Time-lagged correlation analyses were performed to examine the effect of shifting search data relative to background check data. Analyses were conducted at both the national and state levels, examining both temporal and spatial correlations. The researchers used Python programming language (version 3.8) with Pandas (version 1.1.3) and Plotly (version 4.11) for data analysis and visualization.

The study found strong correlations between firearm-related internet search volume and background checks, particularly when stratified by gun type (long guns vs. handguns). For example, the correlation between U.S. long-gun background checks and search volume for "shotgun" in 2019 was Pearson’s r = 0.96 and Spearman’s ρ = 0.94. Similarly, the correlation between U.S. handgun background checks and search volume for "9mm" was Pearson’s r = 0.97 and Spearman’s ρ = 0.94. Spatial correlations were also significant, with stronger correlations observed when stratifying by gun type. The researchers also found a positive association between firearm-related search query volume and firearm-related mortality (e.g., Pearson’s r = 0.87, Spearman’s ρ = 0.90 for "9mm" and firearm deaths in 2017). Finally, they observed a negative association between search volume and the restrictiveness of state-level firearm policies (e.g., Pearson’s r = −0.82, Spearman’s ρ = −0.83 for "9mm" and state policies in 2019). These findings suggest that internet search data could be a valuable tool for monitoring firearm sales, predicting firearm-related deaths, and assessing the impact of firearm policies.

The strong correlation between firearm-related search volume and background checks suggests that search data can serve as a complementary data source for tracking firearm sales, especially when stratified by gun type. The positive association between search volume and firearm-related mortality indicates a potential link between public interest in firearms and firearm-related deaths. The negative association between search volume and policy restrictiveness suggests that less restrictive policies may be associated with increased public interest in firearms. These findings have implications for public health surveillance and policymaking. Internet search data offer several advantages: they are readily available, near real-time, and have higher temporal and spatial resolution than other data sources. They may also capture information not captured by other methods, such as private gun sales. However, limitations exist: internet access varies across demographics, and the intent behind each search is unknown. Future research could explore more granular policy components and investigate the motivations behind firearm-related searches.

This study demonstrates that firearm-related internet search query volume is strongly correlated with background checks, firearm-related mortality rates, and the restrictiveness of state-level firearm policies. These findings suggest that internet search data can serve as a valuable, timely, and complementary resource for tracking firearm sales and ownership, monitoring firearm-related deaths, and assessing the impact of firearm policies. Further research should focus on refining the methodology, exploring the motivations behind searches, and validating these findings with additional data sources.

The study acknowledges several limitations. Internet access varies across demographics, potentially affecting the generalizability of the findings. The specific intent behind each search is unknown; a search for "shotgun," for example, may not always indicate an intent to purchase a firearm. The study relied on publicly available data, and further validation with additional data sources would be beneficial. The use of correlation does not imply causation; the observed relationships may be influenced by other confounding factors.