Public Opinion About Gun Control Post

Abstract

I use data from a national public opinion poll conducted 4 months after the mass shooting of teachers and students in Sandy Hook Elementary School to analyze the content and predictors of public opinion about gun control and gun control politics. I find that a slim majority of Americans favors a semiautomatic weapon ban and proposals to make gun control laws stricter, and a large majority supports a federal background check law. Consistent with previous research, I also find that both instrumental concerns and cultural beliefs are significantly related to people’s opinions about gun control, but the strongest, most consistent predictors of people’s gun control preferences are their political beliefs and affiliations. I conclude by discussing the implications of these findings for the national gun control debate.

Keywords

public opinion gun control semiautomatic weapon background check

On December 14, 2012, Adam Lanza entered Sandy Hook Elementary School in Newtown, Connecticut, with multiple firearms and rounds of ammunition, and he proceeded to shoot and kill 26 people, including 20 students aged seven and younger (Barron, 2012). This tragedy reignited the decades-long debate about gun violence and gun control in the United States. Though the Sandy Hook shooting was merely the latest event in the history of American mass shootings (see Burns & Crawford, 1999; Schildkraut & Hernandez, 2014), the fact that Lanza claimed the lives of young children led President Barack Obama to argue that the country could not afford to fall back into gridlock over the issue of gun control. In a speech, the president said, “We can’t tolerate this anymore. These tragedies must end. And to end them, we must change. Surely we can do better than this” (J. Epstein & Epstein, 2012).

The Democratic and Republican Parties proceeded to react in predictable ways consistent with old issue polarization (Lindaman & Haider-Markel, 2002; Patterson & Eakins, 1998; Vizzard, 2000; Wilson, 2007). The Democratic Caucus in the House of Representatives convened a Gun Violence Prevention Task Force that released a statement of principles and legislative priorities (Wozniak, 2014). These principles included a commitment to reinstate and strengthen prospective, federal bans on assault weapons and bullet magazines; require background checks for every gun sale; strengthen the National Instant Criminal Background Check System; and bolster the nation’s mental health care and violence-prevention systems (among other priorities). Being in the minority, though, the House Democrats possessed virtually no power to push legislation. As such, Senate Democrats took the lead, with senators such as Diane Feinstein introducing assault weapons ban legislation and Chuck Schumer introducing enhanced background check legislation (see Allen, 2012).

The Republicans, in contrast, staunchly opposed the Democrats’ proposals for new gun control laws (Gibson, 2013; Martin & Sherman, 2013; Nocera, 2012; VandeHei & Allen, 2013; Weinger, 2013). For example, Representative Louie Gohmert publicly argued that mass shootings occur because there are too few armed civilians who could kill a shooter before he or she claims too many lives, which means that gun control is the problem, not the solution (Allen, 2012; J. Epstein & Epstein, 2012). Referring to President Obama’s executive actions to strengthen existing gun control laws, Representative Steve Stockman said,

The President’s actions are an existential threat to this nation. The right of the people to keep and bear arms is what has kept this nation free and secure for over 200 years. The very purpose of the Second Amendment is to stop the government from disallowing people the means to defend themselves against tyranny. Any proposal to abuse executive power and infringe upon gun rights must be repelled with the stiffest legislative force possible. (Gibson, 2013)

In total, the Republicans in both the House and the Senate presented a united message; they were willing to consider proposals to improve mental health care and, possibly, background checks to ensure that deranged people could not procure guns, but any new restrictions on gun ownership were strictly off the table (R. J. Epstein & Epstein, 2013; Martin & Sherman, 2013).

The positions of the Democrats and the Republicans created a highly polarized debate with the Democrats investing much of their efforts into legislative proposals that the Republicans said they flatly refused to consider. During this partisan debate, members of both parties frequently cited public opinion to make their case. Democrats argued that a majority of citizens support gun control, while Republicans argued that a majority of citizens oppose gun bans because they support the Second Amendment right to own a gun (R. J. Epstein & Epstein, 2013).

I have three goals in this article. The first is to assess the empirical accuracy of the Democrats’ and Republican’s seemingly-contradictory beliefs about public opinion by analyzing data from a national public opinion poll conducted 4 months after the school shooting during the height of the post–Sandy Hook debate about gun control. The second is to compare the present findings with previous public opinion studies to judge whether or not the Sandy Hook shooting changed people’s policy positions as President Obama argued it should. The third is to move beyond univariate poll analyses by using theory to analyze the predictors of public opinion about gun control.

Trends in Poll Data

Much of our knowledge of public opinion about this policy domain comes from polls conducted by national survey centers, such as the Gallup Corporation and the National Opinion Research Center (NORC) at the University of Chicago. Summarizing data from decades of NORC polls, Smith (2002) concludes, “large majorities [of Americans] back most policies to control the manufacture and sale of guns, increase gun safety, and restrict criminals from acquiring firearms” (p. 155). The public overwhelmingly supports background checks and waiting periods for gun purchases, and a majority of Americans supports a ban on high-capacity ammunition magazines. These poll data suggest that Democratic politicians are correct that the American public supports gun control. However, NORC polls also consistently find that a majority of the public opposes complete bans on the possession of guns, and some polls find that a plurality of citizens believes that allowing citizens to carry concealed firearms makes communities safer. These findings suggest that Republican politicians are correct that the American public opposes policies that would ban gun possession.

What is unambiguous is the finding that public opinion about gun control has been generally quite stable over time (Erskine, 1972). Figure 1 graphs data from Gallup Poll questions about gun control from 1990 through 2014. Unfortunately, not all questions were asked each year, which creates some large gaps in the time series, but we can still observe general trends. The percentage of respondents who stated that “laws covering the sale of firearms” should be “more strict” has slowly but steadily declined from 78% in 1990 to 47% in 2014. There was a spike in support for making firearms laws more strict in December 2012, which suggests that the Sandy Hook shooting may have increased public support for gun control. However, within a year, that spike in support regressed to the steady, average level of support seen throughout the 2000s.

Figure 1.

Gallup poll data on public support for making firearms laws more strict, a ban on handguns, and a ban on semiautomatic weapons, 1990-2014.

The percentage of people who stated that there “should be a law that would ban the possession of handguns, except by the police and other authorized persons” has also been slowly declining over time from 41% in 1990 to 26% in 2014. It is important to note that support for restrictions on handgun ownership, specifically, is about 10% to 20% lower than support for more general restrictions on “firearm sales.”

In contrast to the other two questions, Gallup only sporadically asked people about their support for “a law which would make it illegal to manufacture, sell, or possess semiautomatic guns known as assault rifles,” but the data that do exist suggest a similar trend in opinion. Support for an assault rifle ban dropped from 59% in 2000 to 44% in December, 2012. Importantly, there was only a one percentage point increase in public support for an assault rifle ban between 2011 and 2012, which suggests that the Sandy Hook shooting did not alter mass public opinion about semiautomatic weapons.

With the exception of the one temporary shift in public support for making firearms laws more strict, the general trend in the Gallup data is a very slow but steady shift in public opinion against gun control. Furthermore, the relative flatness of the lines shows that even the many other, high-profile mass shootings in recent American history, such as the 1999 Columbine High School shooting, failed to noticeably alter mass public opinion about gun control (see also Kohut, Doherty, Dimock, & Keeter, 2012; Smith, 2002).

Compared with bodies of scholarship on topics like the causes of crime, the scholarly literature on public opinion about gun control is surprisingly small. However, many of the studies that analyze data other than NORC’s General Social Survey or National Gun Policy Survey largely replicate the findings that Americans support policies like background checks but oppose gun bans (Barry, McGinty, Vernick, & Webster, 2013; Kauder, 1993; Roberts & Stalans, 1997; Semet & Ansolabehere, 2011; Wolpert & Gimpel, 1998; Wright, 1981). Opposition to a handgun ban is widespread and consistent over time. On the other hand, support for an automatic rifle ban is comparatively higher, though some data sets suggest that it fluctuates more over time (Semet & Ansolabehere, 2011; Wolpert & Gimpel, 1998).

In addition to giving us a relatively clear picture of mass opinion about gun ownership and gun control, poll data also consistently show that gun policy opinions are correlated with demographic characteristics. As Smith (2002) states, “Women, residents of large cities and their suburbs, liberals, and Democrats are most likely to support general gun control measures, whereas, men, residents of rural areas, conservatives, and Republicans are least likely to support such measures” (p. 156). Scholarly studies typically replicate the relationship between demographic characteristics and policy opinions observed by Smith in the NORC data (Brennan, Lizotte, & McDowall, 1993; Celinska, 2007; Dowler, 2002; Holbert, Shah, & Kwak, 2004; Kleck, 1996; Robbers, 2005; Semet & Ansolabehere, 2011; Wilson, 2007). Furthermore, some other studies have found a relationship between people’s religious affiliations, gun ownership, and attitudes toward gun control; the common finding is that Protestants are more likely to own guns and less likely to support gun control than Catholics or Jews (Celinska, 2007; Kahan & Braman, 2003), though other studies find no significant effects of religion (Semet & Ansolabehere, 2011).

Theoretical Predictors of People’s Attitudes About Guns and Gun Control

The challenge for scholars, then, has been to craft a theoretical explanation for the fact that public opinion about gun control is divided along demographic and political cleavages. Scholars have put forth two primary theoretical perspectives. The first perspective posits that people’s opinions about gun control are shaped by instrumental concerns. The second perspective posits that public opinion about gun control is inextricably intertwined with public beliefs about cultural values. Each of these two perspectives may be further subdivided. The instrumental perspective includes arguments that gun owners will oppose gun control because they will not vote against their self-interest, as well as arguments that people will support or oppose gun control based upon perceptions of those policies’ ability to prevent crime (or lack thereof). The symbolic values perspective includes discussions of several different facets of the American culture, including individualism versus collectivism, libertarianism, and a moral code rooted in a romanticized, collective memory about pioneering expansion into “the Frontier.” I explain each of these theoretical perspectives in turn.

The instrumental theories are straightforward and relatively self-explanatory. The self-interest perspective, which posits that gun owners will oppose any policies that might threaten their ability to own firearms, has been extensively tested. Gun ownership and/or presence of a gun in the home are among the strongest and most consistent predictors of people’s opposition to various gun control policies, especially any type of gun ownership ban (Brennan et al., 1993; Celinska, 2007; Dowler, 2002; Kleck, 1996; Kleck, Gertz, & Bratton, 2009; Semet & Ansolabehere, 2011; Wolpert & Gimpel, 1998). The self-protection/crime control perspective argues that people will support gun control if they believe that gun control policies will reduce crime, and they will oppose gun control if they perceive those policies to be ineffective or believe that widespread private ownership of firearms is a more effective way to fight crime (Robbers, 2005; Semet & Ansolabehere, 2011; Spitzer, 2011; Tyler & Lavrakas, 1983; Wolpert & Gimpel, 1998). Empirical support for the self-protection/crime control perspective reflects the ambiguous directionality of the theory. Some studies find that fear of crime makes people more supportive of gun control (Dowler, 2002; Heath, Weeks, & Murphy, 1997; Kahan & Braman, 2003; Robbers, 2005), whereas others find that fear makes people more opposed to gun control (Kleck, 1996; Tyler & Lavrakas, 1983). On the other hand, some studies find no significant effect of fear (Semet & Ansolabehere, 2011). The relationship between fear of crime and actually owning a gun is inconsistent; some studies find that people who fear crime are more likely to own a gun than non-fearful people (Holbert et al., 2004), while others find no significant effect of fear on ownership (Cao, Cullen, & Link, 1997; Celinska, 2007).

Many other scholars have theorized that guns are a symbol of deeper cultural values for many Americans, so public opinion about the private ownership or government restriction of guns is really a proxy for people’s opinions about fundamental socio-cultural values. Scholars argue that guns symbolize the American heritage of (a) revolution against government tyranny and (b) frontiersmanship and Manifest Destiny (Kohn, 2004; Spitzer, 2011; Tonso, 1982; Wilson, 2007). As such, the gun control debate directly taps into cultural debates about the power of government versus individual rights and liberty, and concern for the collective good versus individual freedom (Celinska, 2007; Harding, 1998; Kahan & Braman, 2003; Kleck et al., 2009; Semet & Ansolabehere, 2011; Tyler & Lavrakas, 1983; Vizzard, 2000; Wolpert & Gimpel, 1998). Evidence shows that pro-gun politicians and lobby groups, such as the National Rifle Association (NRA), have explicitly appealed to these cultural values when making their case against gun control (Goss, 2004; Harding, 1998; Vizzard, 2000; Wilson, 2007).

Scholars typically argue that people who favor private gun ownership and oppose gun control (especially efforts to ban gun possession) do so because they are skeptical of government power and believe in “rugged individualism.” As an illustrative example, Harding (1998) quotes conservative commentator and onetime presidential candidate Pat Buchanan, who said, “[The Second Amendment] right ain’t about shooting ducks. That right is about a man’s right to defend his wife, his family, his freedom, and his country” (p. 202). Similarly, a former spokesman for the NRA explained that the group’s members believe that the Second Amendment embodies

. . . our freedom to choose, our freedom to decide . . . . It really has nothing to do with guns; it has to do with freedom. It has to do with, again, do you give your freedom to the government or do you keep it within yourself, within your community, within your family? (Breslow, 2015)

After conducting extensive interviews with gun owners and gun enthusiasts, Kohn (2004) concludes that the symbolism of guns encompasses a distinctly moral dimension for American gun owners. They represent romanticized, almost mythic ideals of the “citizen soldiers” and “cowboy lawmen” who opposed tyranny and carved a moral order out of the wild and chaotic frontier. As such, to be a gun owner is to be a steward of the values of “real America” and oppose the moral decay of postmodern society. Kohn (see also Spitzer, 2011; Tonso, 1982) states,

. . . shooters simply do not believe that gun restrictions or prohibitions will disarm the population most capable of violence. Such a situation would seriously threaten or even destroy the integrity of the social body and the body politic. In short, gun control signifies a breakdown of the social and moral order— the very fabric of society itself. On the other hand, gun ownership denotes defending the social and moral order; in short, being tough. Tough individuals create tough communities, which in turn create a tougher nation. (Kohn, 2004, p. 110, emphasis added)

The fact that the values symbolized by guns are correlated with political conservatism in the United States may explain the consistent, empirical relationship between conservative ideology, gun ownership, and opposition to gun control (Cao et al., 1997; Celinska, 2007; Dowler, 2002; Holbert et al., 2004; Kahan & Braman, 2003; Kleck, 1996; Robbers, 2005; Semet & Ansolabehere, 2011). Overall, the cultural values perspective posits that people’s opinions about cultural and political values significantly predict their opinions about gun control above and beyond their demographic characteristics or belief in the utility of guns to prevent crime (Kahan & Braman, 2003; Wolpert & Gimpel, 1998).

It is difficult to summarize the state of our knowledge regarding these different theoretical perspectives. As other scholars have observed (Kleck et al., 2009; Semet & Ansolabehere, 2011), there is little consistency between studies when it comes to the cultural values that scholars test, both in regard to their presence in the analysis and the operational definitions used. Furthermore, scholars have used a wide variety of different questions to measure people’s opinions about different aspects of gun control. Kleck and his colleagues (2009) argue that people typically have only very weak opinions about most gun control proposals, and analyses of weak attitudes are unlikely to uncover strongly-held cultural beliefs; they argue that scholars should focus on testing public opinion about gun ownership bans (especially handgun bans), which evoke the strongest feelings. Support exists in the empirical literature for both the instrumental and cultural values perspectives, though the strength of the support varies depending upon the precise indicators used. This mix of support has led some scholars to believe that the different theoretical perspectives are not mutually exclusive and most likely work in tandem to explain the consistent demographic divisions in support for gun control (Kleck et al., 2009).

A Dearth of Political Factors

However, there is one notable gap in the extant literature. Studies almost never control for political factors other than respondents’ political ideology and/or party affiliation. Given the universally-acknowledged influence of the NRA as an agenda-setter in the gun control debate (Allen, 2012; Parti, 2013), it stands to reason that a person’s affiliation with and/or attitude toward the NRA would be a significant predictor of his or her opinions about gun control policies. We possess surprisingly little empirical survey evidence to confirm whether or not this supposition is accurate.

People study public opinion about gun control because we assume that those opinions determine people’s votes. For example, it has become an accepted mantra among Democratic politicians that the party’s effort to pass the 1994 federal assault weapons ban directly caused many Members of Congress to lose reelection (Tau, 2013; Wozniak, 2014), and it is not difficult to find other stories of politicians who claim that a wrong choice to support gun control cost them their place in elected office (Mathews, 2008). Indeed, former President Bill Clinton said, “All these polls that you see saying the public is for us on all these issues—they are meaningless if they’re not voting issues” (Tau, 2013).

However, the “common sense” belief among elected officials that voters’ positions on gun control will always determine their vote, trumping all other considerations, has never been empirically tested, and political science theory gives us reason to be skeptical of this “single-issue voter” explanation. Scholars of voting behavior largely agree that voters make their choices based upon their belief in a candidate’s ability to do his or her job well and his or her reputation for honesty (Lewis-Beck, Jacoby, Norpoth, & Weisberg, 2008). It is certainly possible that a highly-salient issue about which a voter cares deeply could be the primary determinant of his or her vote choice, but research and theory tells us that we should not take it as a given. Even if a voter disagrees with a politician on gun control, that voter might still vote for the candidate if he or she perceives that candidate to be trustworthy and competent on the whole. The current data set equips me to empirically measure people’s judgments about the degree to which a candidate’s gun control positions would or would not determine their votes, which is a novel contribution to this literature.

Method

Data

The data analyzed in this study come from a public opinion poll jointly conducted by CBS News and The New York Times between April 24 and 28, 2013 (CBS News/New York Times Poll, 2013). The survey questionnaire was administered via telephone to a sample of 965 adults in the United States. The sampling frame was created using both land-line and cell phone numbers. The sampling error of the poll is plus or minus three percentage points.¹ As this is a study of people’s policy opinions, it bears noting that 89% of respondents in this sample say that they are registered to vote, which suggests that this poll captures data from citizens whose opinions are more likely to influence policy outcomes than from citizens who are unengaged in politics.²

These data present many of the typical challenges of working with secondary data. The first challenge is making use of the best available items to measure theoretical constructs when those items were not originally written to specifically measure those constructs. On the other hand, this poll possesses the desirable characteristic of containing multiple measures of different facets of gun control and its related politics. Some studies have demonstrated that the precise determinants of people’s opinions can vary from one gun control question to another (Brennan et al., 1993; Dowler, 2002; Kleck et al., 2009; Semet & Ansolabehere, 2011; Wolpert & Gimpel, 1998), but many studies in this body of literature analyzed single gun control questions or created multiple item indices that do not allow us to identify differential predictors across different types of gun control. In addition, this data set includes the rare measure of respondents’ affiliation with and opinion of the NRA.

The second challenge is missing data. Thirteen of the variables in this analysis suffered from missing cases. The volume of missing data was relatively low for each individual variable, ranging from 0.4% to 6.4% of cases. About 72% of respondents answered all questions, and an additional 24% skipped three or fewer questions; in other words, there were very few cases with large amounts of missing data. Furthermore, an examination of the data revealed that missing cases were relatively evenly spread across variables, which suggests that omissions may be missing completely at random. Unfortunately, the spread of missing data across variables also resulted in a significant drop in sample size due to listwise deletion of missing data in the regression analyses (a loss of about 24% of cases).

Allison (2002) explains that a loss of cases through listwise deletion may not pose a threat to the internal validity of a study. He states, “ . . . if the probability of missing data on any of the independent variables does not depend on the values of the dependent variable, then regression estimates using listwise deletion will be unbiased . . . ” (pp. 6-7). To determine whether or not the missing data were confounded with respondents’ answers on the dependent variable questions, I created dummy variables to identify the missing cases in each independent variable. I then regressed each of these missing case dummies on the four dependent variables.

Out of all the independent variables, only the missing cases in the gun possession variable were significantly related to the dependent variables. This finding raises the possibility that the 5.7% of respondents who declined to state whether or not there is a gun in their home may have significantly different opinions about gun control than the majority of respondents who did answer the question. To address this possibility, I used multiple imputation to predict missing cases in the gun possession variable as a function of participants’ answers to all of the other independent and dependent variable questions in the analysis. I then compared my original regression analyses with the analyses with the imputed data. The substantive findings were unchanged, and as Allison (2002) argues that no method of overcoming missing data is unambiguously superior to listwise deletion, I report the original, non-imputed models here. The results of the imputation analyses were vetted by the peer reviewers and are available upon request.

Dependent Variables—Gun Control and Electoral Politics

General gun control—“In general, do you think gun control laws should be made more strict, less strict, or kept as they are now?” The response options to this question were “more strict,” “less strict,” and “kept as they are.” I retained the nominal coding scheme of this variable.

Semiautomatic weapon ban opposition—“Do you favor or oppose a nationwide ban on semiautomatic weapons including some rifles, pistols, and shotguns that have detachable magazines, allowing them to rapidly fire a high number of rounds?” I coded the responses to this question such that 1 = “oppose ban” and 0 = “favor ban.”

Background check opposition—“Do you favor or oppose a federal law requiring background checks on all potential gun buyers?” I coded the responses such that 1 = “oppose” and 0 = “favor.”

Electoral influence—“Is it possible you would ever vote for a candidate who does not share your views on gun policy, or is this issue so important that you could not vote for a candidate who disagrees with you?” I coded the responses such that 1 = “No, would not vote” and 0 = “Yes, would vote” or “Depends.”

Independent Variables—Instrumental Concerns

Gun possession—“Do you or does any other member of your household own a handgun, rifle, shotgun, or any other kind of firearm?” I coded the responses such that 1 = “Yes, self,” “Yes, other member of household,” or “Yes, self and other” and 0 = “No.”

Concern about crime—This data set does not include a direct measure of fear of crime or criminal victimization. My proxy measure for this theoretical construct is the poll’s “most important problem” question, which was phrased “What do you think is the most important problem facing this country today?” To measure concern about crime, I created a categorical variable that includes people who responded that the country’s most important problem is “crime,” “drugs,” “youth crime/violence/gangs,” or “drugs/youth.”

Independent Variables—Cultural Values

Libertarianism—So, too, does the data set not include a direct measure of adherence to libertarian values or attitudes toward the appropriate scope of government in society. I again use responses to the most important problem measure as a proxy for the construct of libertarianism. I created a categorical variable that includes people who responded that the country’s most important problem is “politicians/government” or “big government/bureaucracy/spending.” I infer that people who classify big government as the country’s “most important problem” hold antithetical attitudes toward government.

Moral order—I include two variables to measure the symbolic tie between gun ownership and the belief that people are responsible for protecting their values and way of life against a chaotic, amoral world. The first variable is concern about societal morals. This is a categorical variable that includes people who responded that the country’s most important problem is “moral values,” “lack of unity in the community,” “religious values,” “family breakdown/single parents,” “moral/family values,” “parenting/lack of discipline,” or “youth values/respect.” The second variable is concern that the country is on the wrong track. This variable was measured with responses to the question, “Do you feel things in this country are generally going in the right direction or do you feel things have pretty seriously gotten off on the wrong track?” I coded responses such that 1 = “wrong track” and 0 = “right direction.” I contend that these measures are consistent with Kohn’s (2004) finding that many gun owners perceive a breakdown in society’s moral order.

Independent Variables—Political Characteristics

NRA opinion—“Is your opinion of the National Rifle Association favorable, not favorable, undecided, or haven’t you heard enough about the National Rifle Association yet to have an opinion?” I coded responses such that 1 = “favorable” and 0 = “not favorable,” “undecided,” or “not heard enough.”

NRA affiliation—“Are you or is any other member of your household a member of the National Rifle Association?” I coded responses such that 1 = “yes, self,” “yes, other member of household,” or “yes, self and other” and 0 = “no.”

Voter registration—“Some people are registered to vote and others are not. Are you registered to vote in the election district where you now live, or aren’t you?” I coded responses such that 1 = “yes” and 0 = “no.”

Partisanship—“Generally speaking, do you usually consider yourself a Republican, a Democrat, an Independent, or what?” I created dummy variables for each affiliation, and I set Democrats as the omitted category in the regression analyses.

Political ideology—“How would you describe your views on most political matters? Generally do you think of yourself as liberal, moderate, or conservative?” To mirror partisanship, I created dummy variables for each classification, and I set liberals as the omitted category.

Control Variables—Demographic Characteristics

First, I control for a respondent’s race (1 = White, 0 = Black or Asian) and ethnicity (1 = Hispanic, 0 = Not Hispanic). Second, I control for a respondent’s region of residence. This variable utilizes the regional classification scheme of the U.S. Census. I created dummy variables for each region, and I set Northeast as the omitted category against North Central, South, and West. Finally, I control for a respondent’s self-identification as an evangelical or born again Christian with a dummy variable that contrasts them against people who do not so identify.

I present the descriptive statistics of all variables in Table 1 along with the predicted direction of the relationship I expect each independent and control variable to have with the dependent variables according to theory.

Table 1.

Descriptive Statistics of Analysis Variables.

Variable	Theorized sign	n	M	SD
Gun control	NA	940	1.78	.92
Automatic gun ban	NA	931	0.45	.50
Background check	NA	952	0.11	.32
Electoral influence	NA	921	0.48	.50
Gun possession	+	910	0.46	.50
Concern about crime	Unclear	965	0.02	.15
Libertarianism	+	965	0.04	.20
Concern about societal morals	+	965	0.03	.17
Country on wrong track	+	905	0.64	.48
NRA opinion	+	937	0.36	.48
NRA affiliation	+	943	0.13	.34
Voter registration	Unclear	961	0.89	.31
Republican	+	903	0.28	.45
Independent	+	903	0.38	.48
Democrat	Omitted	903	0.34	.48
Conservative	+	911	0.42	.49
Moderate	+	911	0.36	.48
Liberal	Omitted	911	0.23	.42
White	+	947	0.82	.39
Hispanic	−	955	0.06	.24
North Central	+	965	0.23	.42
South	+	965	0.34	.47
West	+	965	0.22	.42
Northeast	Omitted	965	0.21	.41
Evangelical	+	965	0.27	.44

Plan of Analysis

I begin by presenting descriptive information about the distribution of respondents’ answers to the gun control questions. I then analyze the relationship between those attitudes and the theorized predictors using multivariate analysis. The categorical nature of the dependent variables calls for the use of maximum likelihood estimation (Long, 1997). I analyze the nominal, general gun control variable with multinomial logit, and I analyze the dichotomous automatic weapon ban opposition, background check opposition, and electoral influence questions with logit.³ All multivariate regressions are weighted to be representative of the U.S. population.

Results

Descriptive Results

Table 2 presents the distribution of respondents’ answers to the dependent variable questions. A majority of respondents believe that gun control laws should be made stricter than they are currently, and only about 10% of respondents want gun control laws to be made less strict. This findings are similar to the September 2013 Gallup poll in which 49% of respondents said that laws covering the sale of firearms should be made more strict and 13% said that they should be made less strict. A majority of respondents in these data favor a ban on semiautomatic guns, and an overwhelming majority of respondents favor a federal background check law for gun purchases. The Gallup poll also asked respondents for their opinions about a ban on the manufacture, sale, and possession of semiautomatic guns in December, 2012; they found that 44% of respondents supported such a ban whereas 51% opposed it. Note, though, that the ABC News and Gallup questions are worded in distinctly different ways, which likely accounts for the different sample statistics.

Table 2.

Distribution of Respondents’ Answers to Gun Control and Election Questions.

In general, do you think gun control laws should be made more strict, less strict, or kept as they are now?
More strict	525 (55.9%)
Kept as they are	322 (34.3%)
Less strict	93 (9.9%)
Do you favor or oppose a nationwide ban on semiautomatic weapons including some rifles, pistols, and shotguns that have detachable magazines, allowing them to rapidly fire a high number of rounds?
Favor	508 (54.6%)
Oppose	423 (45.4%)
Do you favor or oppose a federal law requiring background checks on all potential gun buyers?
Favor	844 (88.7%)
Oppose	108 (11.3%)
Is it possible you would ever vote for a candidate who does not share your views on gun policy, or is this issue so important that you could not vote for a candidate who disagrees with you?
Yes, would vote	414 (45.0%)
Depends	65 (7.0%)
No, would not vote	442 (48.0%)

Overall, these findings support the claims made by Democratic politicians during the gun control debates that followed in the wake of the Sandy Hook shooting. However, the answers to the electoral influence question provide a clue to explain how the Democrats’ efforts to pass new gun control legislation during the 113th Congress failed even though it seems that a majority of citizens support the proposals contained within the Democrats’ bills. The public is closely divided on the question of whether or not they would vote for a candidate who does not share their opinions about gun policy, and only a minority of respondents to this survey indicated that they are confident they would vote for such a candidate. This finding seems to support politicians’ belief that votes on new gun control legislation could endanger them during the next election (Wozniak, 2014).

Multivariate Analysis

Table 3 presents the results of the multinomial logit regression analysis of respondents’ opinions about gun control policies, in general, on the demographic, cultural value, and political variables. The results indicate that possessing a gun in one’s home, concern about the state of society’s morals, belief that the country is on the wrong track, and a positive opinion of the NRA are positively and significantly related to the likelihood that people will say that gun control laws should be “less strict,” as opposed to “more strict.” The results also indicate that possessing a gun in one’s home, belief that the country is on the wrong track, a positive opinion of the NRA, being registered to vote, being Republican or Independent, and holding a conservative or moderate political ideology are positively and significantly related to the likelihood that people will say that gun control laws should be “kept as they are,” as opposed to made “more strict.”

Table 3.

Multinomial Logit Regression Results of Opinion About General Gun Control on Predictor Variables (Relative Risk Ratios With Standard Errors in Parentheses).

Variable	Relative risk ratio
“Less strict” vs. “More strict” (Omitted category)
Gun possession	5.17 (2.42)***
Concern about crime	0.34 (0.48)
Libertarianism	1.41 (0.80)
Concern about societal morals	6.72 (5.65)*
Country on wrong track	3.23 (1.86)*
NRA opinion	17.14 (7.25)***
NRA affiliation	0.49 (0.26)
Voter registration	0.62 (0.36)
Republican	1.84 (1.31)
Independent	2.22 (1.47)
Conservative	1.02 (0.52)
Moderate	0.63 (0.36)
White	0.74 (0.51)
Hispanic	0.42 (0.36)
Evangelical	0.64 (0.25)
North Central	0.34 (0.23)
South	0.44 (0.25)
West	0.35 (0.24)
Constant	0.03 (0.03)***
“Kept as they are” vs. “More strict” (Omitted category)
Gun possession	1.70 (0.46)*
Concern about crime	0.40 (0.36)
Libertarianism	1.07 (0.55)
Concern about societal morals	0.69 (0.45)
Country on wrong track	1.98 (0.59)*
NRA opinion	6.74 (1.80)***
NRA affiliation	0.84 (0.33)
Voter registration	2.00 (0.78)†
Republican	2.22 (0.82)*
Independent	2.23 (0.69)**
Conservative	3.37 (1.35)**
Moderate	1.90 (0.71)†
White	1.09 (0.38)
Hispanic	0.61 (0.26)
Evangelical	0.70 (0.21)
North Central	1.06 (0.39)
South	0.68 (0.24)
West	0.95 (0.35)
Constant	0.03 (0.02)***
Log pseudolikelihood	−494.53
N	734
McFadden’s adjusted R²	0.21

†

p ≤ .10. *p ≤ .05. **p ≤ .01. ***p ≤ .001.

Table 4 presents the results of the logit regression analyses of the remaining dependent variables. Consistent with other studies that examined multiple different types of gun control, these results reveal that factors that predict opposition to a ban on semiautomatic weapons are not identical to the factors that predict opposition to background checks. Possessing a gun in one’s home, belief that the country is on the wrong track, positive opinion of the NRA, and holding a conservative or moderate political ideology are positively and significantly related to the likelihood that people will oppose a ban on semiautomatic weapons. Holding a libertarian concern about big government, holding a positive opinion of the NRA, and having a conservative or moderate political ideology are positively and significantly related to the likelihood that people will oppose a federal law requiring background checks on gun purchases. It is also important to note that the “concern about crime” variable was dropped from the background check regression because it was a perfect predictor; all but one of the respondents who identified crime and violence as the country’s most important problem also opposed background checks.

Table 4.

Logit Regression Results of Opposition to a Semiautomatic Weapon Ban, Opposition to Background Checks, and Electoral Influence on Predictor Variables (Odds Ratios With Standard Errors in Parentheses).

Variable	Dependent variable
Variable	Opposition to semiautomatic weapon ban	Opposition to background checks	Electoral influence
Gun possession	1.51 (0.36)^†	1.55 (0.55)	0.97 (0.20)
Concern about crime	1.58 (1.41)	^a	1.06 (0.78)
Libertarianism	1.88 (0.98)	0.38 (0.22)^†	1.11 (0.53)
Concern about societal morals	2.13 (1.40)	3.42 (3.03)	0.58 (0.37)
Country on wrong track	2.31 (0.60)***	1.09 (0.58)	1.02 (0.24)
NRA opinion	6.64 (1.73)***	5.92 (2.16)***	1.12 (0.26)
NRA affiliation	0.89 (0.33)	0.83 (0.32)	2.00 (0.63)*
Voter registration	1.20 (0.46)	1.13 (0.74)	0.65 (0.21)
Republican	0.68 (0.22)	1.58 (1.20)	1.24 (0.36)
Independent	1.22 (0.34)	2.65 (1.80)	0.93 (0.23)
Conservative	2.38 (0.76)**	7.61 (5.29)**	1.40 (0.39)
Moderate	2.00 (0.57)*	3.38 (2.41)^†	1.02 (0.26)
White	0.82 (0.27)	0.92 (0.40)	0.42 (0.11)***
Hispanic	0.80 (0.34)	1.26 (0.95)	0.81 (0.28)
Evangelical	0.96 (0.24)	1.19 (0.47)	1.18 (0.26)
North Central	0.67 (0.22)	0.69 (0.37)	0.95 (0.28)
South	0.79 (0.23)	0.41 (0.23)	0.94 (0.25)
West	1.17 (0.40)	0.56 (0.32)	1.19 (0.36)
Constant	0.15 (0.07)***	0.01 (0.01)***	2.00 (0.88)
Log pseudolikelihood	−405.06	−192.43	−475.69
N	732	732	722
McFadden’s adjusted R²	0.16	0.13	0.01

The “concern about crime” variable was dropped from the regression because it was a perfect predictor of opposition to background checks.

†

p ≤ .10. *p ≤ .05. **p ≤ .01. ***p ≤ .001.

In contrast to the effects of different cultural values on peoples’ attitudes toward gun control policies, people’s willingness to vote for a candidate who does not share their opinions about guns was only significantly affected by two factors. Holding a positive opinion of the NRA is positively and significantly related to the likelihood that people will say that they would not vote for a candidate whose gun opinions differ from their own. On the other hand, Whites were significantly more likely than racial minorities to say that they would or might vote for such a candidate depending on circumstances. I do note, though, that these results should be viewed with caution because the miniscule adjusted R² indicates that this model does a poor job of predicting people’s responses to this question (though the overall regression is statistically significant).

Discussion and Conclusion

This article compliments both the national gun control debate and the scholarly literature on public attitudes toward gun control by providing additional evidence of the overall stability of mass opinion over time, as well as contributing an analysis of political factors that helps explain the partisan nature of this issue debate. Like many other polls conducted over the past half century, these data from the CBS News/New York Times poll indicate that a (slim) majority of Americans supports the prospect of making gun control laws stricter and, furthermore, supports the proposal to ban semiautomatic weapons. The data also indicate that a much larger majority of Americans supports background checks for gun purchases. Unfortunately, this data set does not measure people’s opinions about proposals to ban or restrict the possession of handguns or hunting rifles, which are usually the gun control proposals that are most strongly opposed by Americans.

The fact that the distribution of gun control responses in this poll is so similar to responses from previous gun control polls, as well as the overall stability in the Gallup gun control items reported in Figure 1, suggests that the Sandy Hook shooting did not dramatically alter the distribution of support or opposition to gun control among the American people—just as previous mass shootings also failed to alter the overall stability of public opinion on this issue (Kohut et al., 2012; Smith, 2002). In that sense, President Obama’s appeal that “We can’t tolerate this anymore. . . . we must change” fell on deaf ears; it appears he failed to convince the roughly 45% of Americans who oppose new gun control proposals or weapons bans to change their minds. On the other hand, the central components of the bills proposed by Democrats in the U.S. House and Senate in the wake of the Sandy Hook shooting were a renewed semiautomatic weapon ban and a proposal to tighten the nation’s background check system (Allen, 2012; Wozniak, 2014). These data indicate that a majority of Americans do support these specific proposals. How, then, can we explain the fact that the Democrats’ bills failed to pass either chamber of Congress?

I argue that the pattern of relationships between the theoretical predictors and attitudes toward gun control provide one possible answer to that question. It is important to note that the majority of Americans who support a semiautomatic weapon ban and the proposal to make gun control laws stricter is a slim majority; the American public still appears to be relatively closely divided over gun control proposals that would affect firearm ownership in one form or another. Scholars have argued that this division in gun control opinions falls along the lines of cultural and political cleavages in American society (Celinska, 2007; Harding, 1998; Kahan & Braman, 2003; Kleck et al., 2009; Semet & Ansolabehere, 2011; Spitzer, 2011; Tyler & Lavrakas, 1983; Vizzard, 2000; Wolpert & Gimpel, 1998). This study supports that conclusion.

I find some limited support for the argument that the gun control debate is a proxy for deeper debates about cultural values in America (Kahan & Braman, 2003; Kohn, 2004), though I find that not all cultural concerns emerge as significant predictors when included together in multivariate analysis. The relationship between a libertarian concern about big government and opposition to gun control is weak in these findings. Concern about big government decreased opposition to background checks, but this effect was only marginally significant. Combined with the fact that the sign of this effect is the opposite of theoretical predictions and it affected no other dependent variable, I am not confident in the reliability of this finding.

On the other hand, the measures of anxiety about moral decay in American society generated more consistent findings. Respondents who were concerned about the state of society’s morals were significantly more likely to say that gun control laws should be less strict, and respondents who feel that the country is on the wrong track also favored making gun control laws less strict and were more likely to oppose a semiautomatic weapon ban. These findings are consistent with the theoretical perspective that some Americans view the right to own guns as being primarily about “ . . . a man’s right to defend his wife, his family, his freedom, and his country” (Pat Buchanan quoted in Harding, 1998, p. 202). In other words, these results suggest that feelings of uneasiness about mainstream culture (or at least a specific perception of mainstream culture) are part of “gun culture” (Kohn, 2004).

The instrumental perspective also finds some support in these data. The presence of a gun in one’s home made respondents significantly less likely to say that gun control laws should be made stricter and significantly more opposed to a semiautomatic weapon ban (though this latter finding is only marginally significant). On the other hand, concern about crime did not technically generate any statistically significant effects, though the near complete opposition to background checks among people who are concerned about crime suggests that scholars should continue to test the effect of instrumental crime variables on people’s attitudes toward guns and gun control.

However, viewed in total, these results indicate that the strongest, most consistent predictors of people’s opinions about gun control are political. Political conservatives and moderates were significantly more likely to say that gun control laws should be kept as they are rather than made stricter, more likely to oppose a semiautomatic weapon ban, and more likely to oppose background checks than were liberals. Similarly, Republicans and Independents were significantly more likely than Democrats to say that gun control laws should be kept as they are rather than made stricter. Even controlling for cultural beliefs, political ideology, partisanship, and demographic characteristics, though, the single strongest predictor in most models was a respondent’s opinion about the NRA (judging by the magnitude of this variable’s relative risk and odds ratios). Respondents who hold a favorable opinion of the NRA were significantly less likely to say that gun control laws should be made stricter and significantly more likely to oppose semiautomatic weapon bans and background checks.

This pattern of results reaffirms the conclusion that the American gun control debate is politicized and polarized and provides additional evidence that the NRA is a major agenda-setter on this issue (Vizzard, 2000; Wolpert & Gimpel, 1998). No matter what President Obama or the Democrats in Congress said about the need for new gun control in the wake of the Sandy Hook shooting, these findings suggest that slightly less than half of the American public listened instead to the counter-arguments of the NRA. Furthermore, the electoral influence model shows that people who are connected to the NRA (personally or vicariously) are significantly more likely to say that they would not vote for a candidate who holds different opinions about guns and gun control. This evidence suggests that gun control votes may be a “litmus test” for NRA-affiliated voters, which likely explains the failure of the new gun control bills in the 113th Congress. Representatives of conservative and rural districts where the NRA is influential (including some Democrats) perceived that they had a strong electoral incentive to oppose the gun control proposals (Allen, 2012; Martin & Sherman, 2013; VandeHei & Allen, 2013). Indeed, a former spokesman for the NRA related stories from his career of times when the NRA was able to flip politicians’ votes in their favor by mobilizing NRA members in the politicians’ districts. He states, “The NRA’s membership, if it had one trait, one political trait, they vote. . . . That’s the deal. You are a politician. You want to get elected. You want votes. NRA has votes” (Breslow, 2015).

Does this mean that gun control will be perpetually gridlocked in American policymaking? Perhaps not. The final, interesting finding of this study is that White respondents were significantly more likely to say that they would consider voting for a candidate who does not share their gun control opinions. Even though race was not a statistically significant predictor of people’s gun policy opinions in these multivariate analyses, other studies find that Whites are more likely to own guns and less likely to support gun control than racial minorities (Celinska, 2007; Kahan & Braman, 2003; Kauder, 1993; Smith, 2002; for contrary evidence, see Semet & Ansolabehere, 2011; Wolpert & Gimpel, 1998). Perhaps, then, politicians are incorrect to infer that NRA affiliates who oppose gun control would definitely vote against them if they support gun control.

Here we reach the limit of our knowledge. Measures of people’s opinions about the NRA have been largely absent in scholarly tests of the predictors of attitudes toward gun control, and I am aware of no other study that directly measures how people’s vote choices might be affected by candidates’ gun control positions. The findings of this analysis suggest that the intersection between attitudes toward the NRA, attitudes toward gun control policies, and the influence of gun politics on citizens’ vote choice would be an important avenue for future study. In light of the contradictory effects of the NRA membership and White race variables (which are certainly overlapping spheres) on the vote choice question, scholars should test whether or not divisions exist within the community of NRA members in regard to opinions about gun control policy and the degree to which a politician’s position on gun control would be the single most important determinant of an NRA member’s vote choice. To pursue this line of inquiry, scholars likely need to gather new data that includes an adequate battery of items to measure numerous gun control policy opinions, opinions about and membership in the NRA, and personal gun ownership. Few existing data sets contain that full combination of items.

Limitations

This study possesses several limitations. First, I had to construct many of the theoretical predictor measures using imperfect proxies. Had this data set contained questions specifically written to measure people’s beliefs about the (a) instrumental effectiveness of gun control laws to prevent crime, (b) libertarianism, and (c) concern about moral decay in modern America, those variables might have emerged as more consistently significant in the regression analyses. Still, it is noteworthy that the present cultural values variables did generate some significant effects that were consistent with theory even with imperfect operational definitions.

Second, these data are cross sectional, so the findings are necessarily correlational, not causal. In addition, the absence of data over time makes this study vulnerable to a possible omitted variable. This survey included no direct measures of respondents’ exposure to or knowledge of politicians’ statements about gun violence and gun control in the wake of the Sandy Hook shooting, so I cannot directly test whether political framing effects altered respondents’ policy opinions. In the wake of tragic events such as mass shootings or terrorist attacks, politicians often discuss the need for safety and protection, which can raise people’s fears and “mortality salience.” Recent research suggests that people who are primed to think about their own mortality sometimes express more conservative policy opinions and become more likely to support politicians who discuss themes like national security and crime control (Burke, Kosloff, & Landau, 2013).⁴ In a related vein, previous studies have found a significant relationship between news consumption or TV crime drama viewing (Dowler, 2002; Holbert et al., 2004) and attitudes toward gun control. Thus, the absence of news consumption items in these data means that I may be failing to measure a significant influence of political framing.

However, there are also reasons to believe that unmeasured framing effects would reinforce or interact with the current findings, not overturn them. I have concluded that people’s political beliefs and affiliations are the strongest predictors of their attitudes about gun control, and numerous framing scholars find that framing effects are moderated by people’s preexisting political beliefs and ideologies. That is, liberal voters are receptive to liberal messages but resistant to conservative messages, and vice versa for conservative voters. People are not blank slates, and politicians cannot push them to hold policy positions that are incommensurate with their prior beliefs, particularly when they are exposed to competing political messages from both sides of a policy debate (Chong & Druckman, 2010; Druckman, 2001; Druckman & Nelson, 2003; Slothuus, 2010; Sniderman & Theriault, 2004). Furthermore, politicians’ ability to change voters’ opinions is at its weakest when they are attempting to alter highly salient, strongly held opinions—which likely describes opinions about guns and gun control (Lecheler, de Vreese, & Slothuus, 2009).

Indeed, Haider-Markel and Joslyn (2001) found just that. They conducted two experiments in which they exposed participants to different frames about the Columbine High School mass shooting created using actual rhetoric from the media coverage of the shooting; these frames emphasized individual rights versus public safety and placing blame for the tragedy on gun laws versus violence in the media. In their first experiment, they found that exposure to the individual rights frame made Republican respondents even more supportive of concealed carry laws, but neither frame had any significant effect on Democrats’ opinions. In their second experiment, they found that only Democrats were significantly affected by the “blame gun laws” frame, while only Republicans and Independents were significantly affected by the “blame violence in the media frame.” They conclude,

Our results . . . suggest that frames are unlikely to influence the attributions of citizens who are not predisposed to the message. Thus, if elites make use of partisan frames, they are unlikely to “convert” anyone. Rather, frames are most likely to reinforce or activate partisan attributions. (Haider-Markel & Joslyn, 2001, p. 536)

To my knowledge, this is the only study that directly measures framing effects on public opinion about gun control.

In light of the findings of Haider-Markel and Joslyn (2001) and the present study, I hypothesize that effects of exposure to politicians’ or journalists’ arguments about guns and gun violence on public opinion about gun control will be moderated by people’s political ideology, political partisanship, NRA membership, and attitude toward the NRA. Future studies should test these framing interaction effect hypotheses, ideally with experimental or panel data to measure opinion change over time after exposure to alternative frames.

Conclusion

Analyzing data from a national public opinion poll conducted 4 months after the Sandy Hook Elementary School mass shooting, I find that a slim majority of Americans favor a semiautomatic weapon ban and proposals to make gun control laws stricter, and a large majority support a federal background check law. In contrast, the public is more closely divided on the question of whether or not they would vote for a candidate who does not share their opinions about guns. Consistent with previous research, I also find that both instrumental concerns and cultural beliefs are significantly related to people’s opinions about gun control, but the strongest, most consistent predictors of people’s gun control preferences are their political beliefs and affiliations. Despite President Obama’s appeal that “this time must be different,” these findings indicate that the Sandy Hook shooting did not significantly alter the polarized and politicized nature of the gun control debate in America.

Footnotes

Acknowledgements

I wish to thank Andrew Fasullo, Brittne Lunniss, and Adam Pittman for their research assistance on this project. I also wish to thank the Roper Center Public Opinion Archives for making the data analyzed in this study publicly available.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Kevin H. Wozniak is an assistant professor of sociology at the University of Massachusetts Boston. He studies the politics of punishment and criminal justice. He received his PhD in justice, law, & socity and American politics from American University. He was also a 2012-2013 Congressional Fellow of the American Political Science Association. During his fellowship, he worked as a legislative aid for Representative Bobby Scott (VA-03), a member of the House Democratic Gun Violence Prevention Task Force.

References

Allen

(2012, December 16). After Connecticut school shooting, Washington quiet on gun control. Politico. Retrieved from http://www.politico.com/story/2012/12/washington-silent-on-gun-control-85130.html

Allison

P. D.

(2002). Missing data. Thousand Oaks, CA: SAGE.

Barron

(2012, December 15). Gunman massacres 20 children at school in Connecticut; 28 dead, including killer. New York Times, p. A1.

Barry

C. L.

McGinty

E. E.

Vernick

J. S.

Webster

D. W.

(2013). After Newtown: Public opinion on gun policy and mental illness. The New England Journal of Medicine, 368, 1077-1081.

Brennan

P. G.

Lizotte

A. J.

McDowall

(1993). Guns, southernness, and gun control. Journal of Quantitative Criminology, 9, 289-307.

Breslow

J. M.

(2015, January 6). NRA Insiders on the politics of guns. NPR. Retrieved from http://www.pbs.org/wgbh/pages/frontline/government-elections-politics/gunned-down/nra-insiders-on-the-politics-of-guns/

Burke

B. L.

Kosloff

Landau

M. J.

(2013). Death goes to the polls: A meta-analysis of mortality salience effects on political attitudes. Political Psychology, 34, 183-200.

Burns

Crawford

(1999). School shootings, the media, and public fear: Ingredients for a moral panic. Crime, Law and Social Change, 32, 147-168.

Cao

Cullen

F. T.

Link

(1997). The social determinants of gun ownership: Self-protection in an urban environment. Criminology, 35, 629-658.

10.

CBS News/New York Times Poll. (2013). Poll # USCBSNYT2013-04B. Storrs: The Roper Center, University of Connecticut.

11.

Celinska

(2007). Individualism and collectivism in America: The case of gun ownership and attitudes toward gun control. Sociological Perspectives, 50, 229-247.

12.

Chong

Druckman

J. N.

(2010). Dynamic public opinion: Communication effects over time. American Political Science Review, 104, 663-680.

13.

Dowler

(2002). Media influence on attitudes toward guns and gun control. American Journal of Criminal Justice, 26, 235-247.

14.

Druckman

J. N.

(2001). The implications of framing effects for citizen competence. Political Behavior, 23, 225-256.

15.

Druckman

J. N.

Nelson

K. R.

(2003). Framing and deliberation: How citizens’ conversations limit elite influence. American Journal of Political Science, 47, 729-745.

16.

Epstein

R. J.

(2012, December 16). Connecticut school shooting: Obama: “These tragedies must end.” Politico. Retrieved from http://www.politico.com/story/2012/12/obama-to-speak-at-newtown-vigil-85155.html

17.

Epstein

R. J.

Epstein

(2013, January 16). Obama guns plan meets fast resistance. Politico. Retrieved from http://www.politico.com/story/2013/01/obama-gun-control-plan-is-announced-86280.html

18.

Erskine

(1972). The polls: Gun control. Public Opinion Quarterly, 36, 455-469.

19.

Fisher

J. E.

Mason

R. L.

(1981). The analysis of multicollinear data in criminology. In Fox

J. A.

(Ed.), Methods in quantitative criminology (pp. 99-125). New York, NY: Academic Press.

20.

Fox

(1991). Regression diagnostics. Newbury Park, CA: SAGE.

21.

Gibson

(2013, January 16). House conservatives hit President Obama on guns. Politico. Retrieved from http://www.politico.com/story/2013/01/overreach-on-guns-say-house-conservatives-86273.html

22.

Goss

K. A.

(2004). Policy, politics, and paradox: The institutional origins of the Great American Gun War. Fordham Law Review, 73, 681-714.

23.

Haider-Markel

D. P.

Joslyn

M. R.

(2001). Gun policy, opinion, tragedy, and blame attribution: The conditional influence of issue frames. Journal of Politics, 63, 520-543.

24.

Harding

D. R.

(1998). Public opinion and gun control: Appearance and transparence in support and opposition. In Bruce

J. M.

Wilcox

(Eds.), The changing politics of gun control (pp. 196-224). Lanham, MD: Rowman & Littlefield.

25.

Heath

Weeks

Murphy

M. M.

(1997). Gun attitudes and fear of crime. Journal of Offender Rehabilitation, 25, 147-157.

26.

Holbert

R. L.

Shah

D. V.

Kwak

(2004). Fear, authority, and justice: Crime-related TV viewing and endorsements of capital punishment and gun ownership. Journalism & Mass Communication Quarterly, 81, 343-363.

27.

Kahan

D. M.

Braman

(2003). Less persuasion: A cultural theory of gun-risk perceptions. University of Pennsylvania Law Review, 151, 1291-1327.

28.

Kauder

(1993). One-gun-a-month: Measuring public opinion concerning a gun control initiative. Behavioral Sciences & the Law, 11, 353-360.

29.

Kleck

(1996). Crime, culture conflict and the sources of support for gun control. The American Behavioral Scientist, 39, 387-404.

30.

Kleck

Gertz

Bratton

(2009). Why do people support gun control? Alternative explanations of support for handgun bans. Journal of Criminal Justice, 37, 496-504.

31.

Kohn

(2004). Shooters: Myths and realities of America’s gun cultures. New York, NY: Oxford University Press.

32.

Kohut

Doherty

Dimock

Keeter

(2012). Views on gun laws unchanged after Aurora shooting. Washington, DC: Pew Research Center.

33.

Lecheler

de Vreese

Slothuus

(2009). Issue importance as a moderator of framing effects. Communication Research, 36, 400-425.

34.

Lewis-Beck

M. S.

Jacoby

W. G.

Norpoth

Weisberg

H. F.

(2008). The American voter revisited. Ann Arbor: University of Michigan Press.

35.

Lindaman

Haider-Markel

D. P.

(2002). Issue evolution, political parties, and the culture wars. Political Research Quarterly, 55, 91-110.

36.

Long

J. S.

(1997). Regression models for categorical and limited dependent variables. Thousand Oaks, CA: SAGE.

37.

Martin

Sherman

(2013, January 16). Why President Obama’s gun plan may be doomed. Politico. Retrieved from http://www.politico.com/story/2013/01/house-shows-little-appetite-for-gun-reform-86251.html

38.

Mathews

(2008, November 4). It was guns, not race, that affected Bradley. Politico. Retrieved from http://www.politico.com/news/stories/1108/15220_Page2.html

39.

Nocera

(2012, December 19). Republicans resist assault weapons ban. Politico. Retrieved from http://www.politico.com/story/2012/12/republicans-are-not-keen-on-assault-weapons-ban-85285_Page2.html#ixzz3Qc5qKRRZ

40.

Parti

(2013, January 22). NRA spent $2.5M lobbying in 2012. Politico. Retrieved from http://www.politico.com/story/2013/01/nra-spent-25m-lobbying-in-2012-86557.html

41.

Patterson

S. C.

Eakins

K. R.

(1998). Congress and gun control. In Bruce

J. M.

Wilcox

(Eds.), The changing politics of gun control (pp. 45-73). Lanham, MD: Rowman & Littlefield.

42.

Robbers

M. L. P.

(2005). The media and public perception of criminal justice policy issues: An analysis of Bowling for Columbine and gun control. Journal of Criminal Justice and Popular Culture, 12, 77-95.

43.

Roberts

J. V.

Stalans

L. J.

(1997). Public opinion, crime, and criminal justice. Boulder, CO: Westview Press.

44.

Schildkraut

Hernandez

T. C.

(2014). Laws that bit the bullet: A review of legislative responses to school shootings. American Journal of Criminal Justice, 39, 358-374.

45.

Semet

Ansolabehere

(2011, November). Profiling and predicting opinion on gun control: A comparative perspective on the factors underlying opinion on different gun control measures. Paper presented at the Conference of Empirical Legal Studies, Chicago, IL. Retrieved from http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1884661

46.

Slothuus

(2010). When can political parties lead public opinion? Evidence from a natural experiment. Political Communication, 27, 158-177.

47.

Smith

T. W.

(2002). Public opinion about gun policies. The Future of Children, 12, 154-163.

48.

Sniderman

P. M.

Theriault

S. M.

(2004). The structure of political argument and the logic of issue framing. In Saris

W. E.

Sniderman

P. M.

(Eds.), Studies in public opinion: Attitudes, nonattitudes, measurement error, and change (pp. 133-165). Princeton, NJ: Princeton University Press.

49.

Spitzer

R. J.

(2011). The politics of gun control (5th ed.). Boulder, CO: Paradigm Publishers.

50.

Tau

(2013, January 19). Bill Clinton to democrats: Don’t trivialize gun culture. Politico. Retrieved from http://www.politico.com/story/2013/01/bill-clinton-to-democrats-dont-trivialize-gun-culture-86443.html

51.

Tonso

W. R.

(1982). Gun and society: The social and existential roots of the American attachment to firearms. Washington, DC: University Press of America.

52.

Tyler

T. R.

Lavrakas

P. J.

(1983). Support for gun control: The influence of personal, sociotropic, and ideological concerns. Journal of Applied Social Psychology, 15, 392-405.

53.

VandeHei

Allen

(2013, January 8). The “hell no” caucus. Politico. Retrieved from http://www.politico.com/story/2013/01/behind-the-curtain-the-hell-no-caucus-85872.html

54.

Vizzard

W. J.

(2000). Shots in the dark: The policy, politics, and symbolism of gun control. Lanham, MD: Rowman & Littlefield.

55.

Weinger

(2013, January 17). Ted Cruz: Gun push will have 2014 political price. Politico. Retrieved from http://www.politico.com/story/2013/01/cruz-gun-push-has-2014-political-price-86351.html

56.

Wilson

H. L.

(2007). Guns, gun control, and elections. Lanham, MD: Rowman & Littlefield.

57.

Wolpert

R. M.

Gimpel

J. G.

(1998). Self-Interest, symbolic politics, and public attitudes toward gun control. Political Behavior, 20, 241-262.

58.

Wozniak

K. H.

(2014). A political science assessment of the House Democratic Gun Violence Prevention Task Force. PS: Political Science and Politics, 47(2), 550-552.

59.

Wright

(1981). Public opinion and gun control: A comparison of results from two recent national surveys. Annals of the American Academy of Political and Social Science, 455, 24-39.

Public Opinion About Gun Control Post–Sandy Hook

Abstract

Keywords

Trends in Poll Data

Theoretical Predictors of People’s Attitudes About Guns and Gun Control

A Dearth of Political Factors

Method

Data

Dependent Variables—Gun Control and Electoral Politics

Independent Variables—Instrumental Concerns

Independent Variables—Cultural Values

Independent Variables—Political Characteristics

Control Variables—Demographic Characteristics

Plan of Analysis

Results

Descriptive Results

Multivariate Analysis

Discussion and Conclusion

Limitations

Conclusion

Footnotes

Acknowledgements

Declaration of Conflicting Interests

Funding

Notes

Author Biographies

References