To access and utilize the script, open the "needham-gun.law.R" file, which contains the code for my findings. To access the dataset, please utilize the GSS Data Explorer exclusively using variables and data from 2022.
To access and read the paper created from the data and supplied code, open the file named, "Symbols of Safety: The Impact of Gun ownership, Regional cultural legacies, Political identity, and Trust in Government on Support for permit requirements.docx".
This study utilized data from the 2022 General Social Survey (GSS), a nationally representative survey designed to track social trends and public opinion in the U.S. The GSS surveys adults aged 18 and older utilizing a full-probability sampling design, ensuring that the dataset reflects the U.S. population. In 2022, the GSS employed a combination of face-to-face interviews and probability sampling methods to gather responses. Probability sampling methods, such as stratified random sampling, were used to ensure that every non-institutionalized adult in the U.S. had a known and non-zero chance of being selected. This approach helps minimize bias and ensures the sample represents diverse demographic groups, including variations in age, gender, race, and geographic location. The initial sample included 4,149 respondents; however, my dataset consists of only 1,274 cases due to various limitations in the data collection process.
The reduction in sample size is due to several factors. First, missing data played a role, as some respondents did not answer key questions, making their cases unusable. This could have occurred for several reasons, such as a lack of knowledge on the topic, discomfort with the question, becoming tired due to over-questioning, or simple oversight. Regardless, these cases need to be excluded to ensure accuracy and consistency in the dataset. Second, the GSS uses a ballot rotation system, in which respondents are assigned to one of three ballots—Ballot A, B, or C. In 2022, only Ballots A and C included the variables I examined, which further reduced the sample size. Lastly, some individuals selected “Don’t Know” or refused to answer, leading to their exclusion from the final dataset. As such, these factors explain the discrepancy between the total GSS sample and the number of observations in my dataset.
This study examines the relationship between gun ownership and attitudes toward gun control, specifically whether individuals support or oppose requiring a police permit to purchase a firearm (specifically under the category for Support Gun Permits). The dependent variable, support for gun permits, is operationalized using responses to the question: “Would you favor or oppose a law which would require a person to obtain a police permit before he or she could buy a gun?” which used the variable, gunlaw. Respondents could answer either “Favor” or “Oppose.” A majority—73%—favor such a law, while 27% oppose it, indicating strong overall support for requiring police permits. The variable utilized from the GSS was gunlaw.
To assess symbolic attachments to firearms, I examined Gun Ownership, based on whether respondents reported having guns or revolvers in their home, using the GSS variable owngun. About 33% answered “Yes,” while 66% answered “No.” I created a dummy variable, coding gun ownership as 1 and non-ownership as 0, to simplify analysis in regression models.
To explore the impact of regional cultural identities, I used the variable region based on where respondents lived at age 16 (reg16) to measure Region. Categories were recoded and thus collapsed to match major U.S. regions: 10% of respondents were classified as Foreign, 16% as Northeast, 22% as Midwest, 32% as South, and 21% as West. This variable reflects early-life regional influences on gun policy attitudes.
Partisan identities were measured using the variable Political Identity, and used the variable partyid from the GSS. This variable was recoded, and respondents were grouped into four categories: Democrat (34%), Independent (19%), Republican (43%), and Other (4%) based on their answers to the question, “Generally speaking, do you usually think of yourself as a Republican, Democrat, Independent, or what?” (the variable being named partyid in the GSS). These reflect respondents’ general partisan self-identification and were used to assess how political affiliation relates to support for permits.
Trust in government was measured through views on Government Spending on Crime via the variable natcrime on the GSS. Respondents were asked whether the government is spending “Too Little” (73%), “About Right” (19%), or “Too Much” (8%) on halting crime. No recoding was needed. Finally, Afraid to Walk at Night in Neighborhood was used to assess perceived personal safety via the variable fear on the GSS. Respondents were asked if there was any area within a mile of their home where they would be afraid to walk alone at night. About 41% said “Yes” and 59% said “No.” This variable was used as a proxy for fear of crime, which may influence support for gun regulation.