Americans Resigned to Gun Violence?

Tom H. C. Anderson
October 5th, 2017

New Text Analytics Poll Measures American’s Desire for Change After Vegas Shooting 

In the wake of the awful tragedy in Las Vegas Monday I like many others viewed the news with shock, sadness, disgust and disbelief. And I believe also with a feeling of helplessness. At least that’s what I prefer to think of it as rather than apathy. The number shot and killed was horrific, but not any more horrific than the Sandy Hook shootings which occurred less than an hour away from us here in Connecticut.

What is the tipping point I wondered Monday morning? How bad does it have to get before people demand change, and what does change look like. Certainly for many an outright ban on all guns would be desired I thought, or would it? In the minds of a sizeable group it seems the second Amendment is as important as freedom itself.  But to affect real change we need some consensus if not a majority.

I took the question to representative sample of 1,600 Americans fielded Tuesday – Thursday. Rather than simply asking whether they agreed or disagreed with more gun control, the NRA, second Amendment or some such topic I simply stated “ Gun control is a difficult issue in the US. Reasonably what things do you think could be done about it? [Please as specific as possible]

The kinds of insights possible with open unstructured questions are impossible to get with forced/multiple choice type questions. The comments represented below represent peoples initial thoughts and responses in their own words off the top of their heads without any influence or suggestions.

Most Popular Answer – “Nothing”

Below is a visualization of the free form answers to the question. Frequency of topic answers on the Y axis, and speed of answering on the X axis.

What could be done about gun control poll

The most frequent answer to this question among US adults after the worst shooting in US history is “Nothing”. Those answering nothing gave the question exactly 49 seconds of thought on average before answering. Let that sink in for a minute.

Looking toward the right of the plot, we can see those answers which were given after slightly more thought. For instance, the suggestion of “One Gun Per Person”, while given by extremely few people, was given after slightly longer deliberation. Certainly, the second most popular answer “Background Checks” doesn’t seem to be cutting it.

Relative Frequency of Suggested Solutions

Taking a closer look at some of the more popular suggestions, the chart below tells us a bit about where the population is in understanding the issue.

Gun Violence 2

Summarizing this chart, I would say the average American still doesn’t seem willing to take radical steps to curtail gun access and violence.

Almost none of the suggestions would have stopped the Vegas shooter. Hardly anyone suggested strict gun laws such as those in countries like Japan where guns are no longer a problem. “One Gun Per Person” is an interesting but rare suggestion that would also have helped in the Vegas shooting.

However, many of the most popular suggestions, like “Background Checks”, “Banning Automatic Weapons”, “Mental Health Detection” would not have been helpful. Yet others like “Eliminate Semi-automatics” and “Regulate Caliber of Guns” show a real lack of understanding about guns.

An unexpected suggestion picked up by our software was the banning of “Bump/Slide Fire Stocks”. This was mentioned by almost 2% of those giving suggestions. While this answer does show strong understanding of guns and the ‘hack’ the shooter used to fire as if his guns were fully automatic, it is perhaps one of the scariest answers in the chart, and one that even the NRA seems to be willing to accept.

Banning Bump stocks is not likely to make any serious impact at all on future gun violence and is just a distraction and a scape goat. This is unfortunately likely to become a big talking point on the Hill.

Non- Gun Owners who don’t understand fire arms enough to make distinction between jargon and real improvements to our safety owe it to themselves to get educated enough to debate these issues and rally for real change.

Don’t be apathetic, don’t be helpless, make change happen!


24 thoughts on “Americans Resigned to Gun Violence?”

  1. PS. My colleagues here at OdinText were arguing this morning about whether the post would have been better without my opinion/stance on the topic. I’m curious to hear other researchers thinking on this. Is it ok for us to take a stance on any issues, or does doing so cheapen the research findings and make it look too biased. I do feel there is too much opinion in news, and certainly don’t want that to become a thing in research. Still I felt that since it was a blog post adding it in was ok. Curious to hear your thoughts on that as well?

  2. Your stock photo is a bit disingenuous – it shows a military M-16 variant with a 40mm grenade launcher attached – a “destructive device” strictly prohibited by Federal law.

    ” Hardly anyone suggested strict gun laws such as those in countries like Japan where guns are no longer a problem.”

    Citing government policy in Japan isn’t such a good idea either – Japan has a very high suicide rate (#26 worldwide, US is #48 – and it is believed that the Japanese deliberately under count the rate) and don’t forget that Japan committed some of the worst atrocities in WWII – all part of government policy. And don’t forget that one of the reasons the Japanese didn’t want to invade the US was the belief that there was “a gun behind every tree”. Funny thing about that.

  3. I think it’s very difficult for a researcher to separate himself/herself from the topic being researched – bias comes in a lot of forms. That said, my policy is always to try to make cogent points that are defensible based on what the study found. Analysis is one thing; interpretation is another and is far more subjective. Saying, “Here’s what the data suggests we can/should do” is one thing; saying, “Here’s what I believe we can/should do” is something entirely different, which makes a study look very biased. For instance, pointing out that “Summarizing this chart, I would say the average American still doesn’t seem willing to take radical steps to curtail gun access and violence” is defensible through the data. Saying banning bump stocks is “just a distraction and a scape goat” is not supported through the data you present. It may be supported through other data, but in not presenting that data, it makes it sound like it’s just your own opinion, and may be either an educated or uneducated opinion.

    A bigger concern is the question itself – I wonder how many people said “nothing” because they believe nothing CAN be done given the country’s gun culture and strong pro-gun politicians and lobbyists, and how many because they believe nothing SHOULD be done? Also, I believe some people interpret “gun control” as “take away most guns,” and if so, how many people said “nothing” because they don’t believe in gun control defined that way? The issue itself is not gun control; gun control is a possible solution to the issue of gun violence and mass shootings. I’d love to see a follow-up with a question such as “What should be done to help reduce gun violence and mass shootings in the US?” Or maybe “What types of laws related to guns, if any, would help reduce gun violence and mass shootings in the US?” Might be an interesting exercise in how question wording impacts the feedback we receive.

  4. The prompt question in your text asks respondents what “could” be done about gun violence. The graphic summarizes the data as what “should” be done–two very different concepts. Was also interested to note the distance you placed between “reasonably” and the rest of the sentence. Would not be surprised if this had an effect on interpretations of the question.

    1. I agree that the word reasonably likely created a mind set where some of the tougher decisions such as the Australian solution were discounted as unreasonable. So they are unwilling to take radical steps because the NRA would stop them.

  5. @Kern Thanks, updated visualization to have same terminology as original question. RE question, yes that was purposeful. On topics with high emotional valence such as this I wanted something more measured. Otherwise there is the risk of getting more emotional less useful comments. Based on the data I believe it was the right choice.

  6. Often interested in your weekly poll, Tom. Yes, of course – take a data-driven stance. That’s what we researchers get paid to do – not just “deliver information.” If the data suggests a point of view, it’s our obligation to present that point of view. And if we have other experiences/information that allow us to editorialize a bit more, then that certainly can be brought to bear as well.

    One point about language – your question asked about “gun control.” I wonder if you’d get different responses if you called it “gun violence” or “what can be done about mass shootings.”

  7. Tom, since you you say you are curious to hear our thoughts, i will quote you first. You stated you own thoughts with “What is the tipping point I wondered Monday morning? How bad does it have to get before people demand change, and what does change look like. Certainly for many an outright ban on all guns would be desired I thought, or would it? In the minds of a sizeable group it seems the second Amendment is as important as freedom itself. But to affect real change we need some consensus if not a majority.”

    You did not say explicitly why you chose to use your resources to ask the question in a survey, rather than simply muse or discuss it with those around you? That leads me to ask you the purpose of the blog itself. Is it a way for you to convey information public issues important to you ? Is it a marketing tool for Odin text? If both, might they be in conflict?

    Speaking euphemistically , might it be wise to keep the business of your business business, rather than allow the emotions around a tragedy spill onto the pages of your blog? Or perhaps you could have used the blog to share your thoughts on the tragedy, even the possible causes and solutions without using a survey as a context.

    I just scanned the entire Pew Report addressing all the ways they could differentiate between Democrats and Republicans on social issues and more. They failed to mention that nearly half of Americans of voting age are registered as members of neither of those two parties. Why not? Because divisiveness is more sensational.

    There is enough real and fake sensationalism to go around. Entire industries are built on it. I fear survey research is becoming more susceptible to it. In fact, even during my career, I saw in others, and felt myself resisting, a tendency to blast out some scary fact, or controversial opinion. Human nature I guess. But not a business I care to be part of.

  8. @Walter Thanks. Yes of course, even slight changes in language may slightly change outcome. Operative word being slight, major changes more major outcomes. I thought the question here worked out well.

    TEXT ANALTYICS PRO TIP: Ask question 3 ways, and analyze separately AND also together.

    @Rick I believe only 3-4 of 1600 people referenced other specific countries including Korea and China. As stated above, rather happy with question, but again this is meant for our series of TextPoll/Will it Unstructure Text Analytics Tips blog posts, but would have experimented more with more time/budget.

  9. Tom, the value of your exercise in my view was create this comprehensive list of ad hoc responses that you would now test in a max-diff or other quantitative survey supplemented w/ smart list of questions to more deeply understand the pulse of the country. You have the breadth of responses and can be certain you’ve not left any out by coming up with such a list around the conference table (as a researcher I interpret the “nothing” responses as laziness in answering a complex question vs apathy).
    Agree with others that question-wording about “gun violence” would have been better than calling the issue “gun control” which is already politically *charged* and may have biased those who chose to respond.

  10. @Rachel, very different kind of thing conjoint. Might be interesting here as a follow up sure. My biggest issue with conjoint is how bad it sucks for the respondent, which makes me wonder about data accuracy…

  11. 1-2 minute easy honest unaided top of mind 1 question survey vs 20 minutes of conjoint hell 😉

  12. Tom – I share your frustration with the current state of our gun laws here and the structure of your question reveals the depth of your passion as evidenced by the proposition you set up in the first sentence:
    “Gun control is a difficult issue in the US.”

    “Reasonably”, then, most people think that nothing “could be” done. It’s hard to say what that means, even after 49 seconds (!) of thinking about it. Crazies will be crazies? Congressman will remain beholden to NRA big money and the gun-rights activists they have been so successful in mobilizing? Respondents may have answered in the same way you originally structured the chart; nothing “should be” done. In any event, 49 seconds of thinking that returns “nothing” is troubling, no matter what it means.

    Here’s what I get out of the results:
    Many American are not well informed well on the issue. For example, background checks are in place. The so-called gun show loophole and the private sale loophole on background checks (actually there is only a private sale loophole; “gun sale loophole” is used somewhat incorrectly) is required for firearms purchases through FFL gun dealers.

    Another misconception revealed in the polling – Automatic weapons are banned already.

    Regarding the results, would it be reasonable to combine “Ban Automatic Weapons, “Ban Types/Assault Weapons,” “Eliminate semi automatics” and “Stricter Laws Assault?” That combination adds into the mid 30% range. Depending upon what the actual data looks like, this could this be more instructive rolled up into a single metric with an explanation?

    I’m OK with you taking a position. I don’t view this as the definitive answer. I like @ Rachel Dreyfus’ thinking that this poll is good starting point for further investigation.

  13. Tom, kudos for taking a stand as a citizen and as a researcher. Contrary to popular belief, we do not in our business have to remain objective about everything all the time. In fact I think we do ourselves and our industry a disservice when we proffer mealy-mouthed conclusions and vanilla recommendations in the name of objectivity. So thanks for having the courage to put your beliefs out there.
    That said, as one someone pointed out, the data do seem to be influenced by the way you framed the question, which presupposes that guns are the thing needing control, not the violence committed using guns. It’s an important distinction, and I think it could influence both the results and our interpretation of those results.
    As you pointed out, the most striking thing is the large number who answered “nothing”. There are a lot of reasons a reasonable and thoughtful person might answer “nothing”, and not all of them are about powerlessness in the face of an unresponsive government (thought that is a very valid reason). Someone who believes the person who murdered those people is the problem rather than the tool he used might very well answer “nothing”. We could probably come up with other reasons.
    In the context of this post, your citation of the 49 seconds the average respondent took to answer that question seems to imply that’s not a lot of time (“Let that sink in for a minute.”).
    But everyone’s been thinking about Las Vegas to some degree since the shooting, so respondents are not answering cold. And if someone asked you a question and you paused 49 seconds before answering, I think your interrogator would be likely to think you thought about the answer a LOT. (Time out 49 seconds on your watch.)
    I would love to see you try this again with the question reframed as “Gun violence is a difficult issue in the US. Reasonably what things do you think could be done about it?” and see what you get. Maybe follow up with “Why do you feel these are reasonable solutions to the problem of gun violence?”
    Finally I want to commend you whole-heartedly for your suggestion that non-gun owners should educate themselves about guns. I think one reason (not the only reason) gun owners react as they do to gun control is because of the obvious ignorance some have about guns, how most people use them, and the laws already on the books. Case in point: the second most frequent response was “ban automatic weapons”. Ordinary citizens have not been able to own automatic weapons without a federal license (law enforcement background check, fingerprints, notification of local law enforcement, etc.) since 1934. Those in favor of more restrictive gun control could be a lot more persuasive if they were better educated.

  14. Hi Eric thanks. Many good thoughts. RE combining, yes of course you could, but those are all unique things in my mind. An Assault weapon is not Automatic. A Semi Automatic is not an Assault weapon (necessarily), and of course most semi automatics can be turned into automatic. Still almost all modern guns are semi automatic so the use of the term “semi automatic” to me just shows how many have a viewpoint without knowing anything about guns. That’s sort of like saying bring back muskets.

    1. Tom, I applaud the use of both your research tools, and this forum for discussion. The active component of your question “what CAN be done” is most impactful. Like many, I believe that if after Sandy Hook, we did not make major changes in responsible firearm access, a catastrophe like Las Vegas was all but inevitable. However, in London, where guns are illegal and almost none to be found, the new weapon of choice is sulfuric acid in a toy balloon. Thrown at an individual or a crowd, this has killed at least one, and blinded many. Other types of weapons are improvised more and more. We need to understand what drives individuals with a goal to kill. Only then can we change this tide.

  15. Indeed, “nothing” as a response to what “could be done” means something vastly different than as a response to “what should be done”.

    Daily we hear a little more about the maniac who perpetrated this massacre. Sounds like he used his above average {?} mental abilities to carry out the task of mass murder. While I agree a commercial product devised and marketed to sidestep a law against automatic firearms should be forbidden by law, it is the madness behind the act of using such a weapon on innocent civilians that will seek and find an effective alternative. Stephen Paddock was not an ignorant terrorist who had been recruited by a gang leader. Paddock would have found a means to carryout a mass murder one way or another, his stash of explosive materials is evidence of that, as was his preparation and exploration of alternative venues.

  16. I dont think its a matter of the weapon type… Its a matter of the mental state of the suject holding the weapon. Even if you are retired soldier or former one… They should make you take a phycological test every year, to make sure you are able to have a weapon…

  17. I would have expected you to use “What things do you think should be done about gun control in the U.S.? [Please as specific as possible].” Could implies some knowledge of legislative process, while should gets at the core of their feelings. Your pre-comment about it being difficult ( Gun control is a difficult issue in the US. ) might have been the reason for the pessimism. You should have let them characterize their own perspective on the issue. And your requested “reasonably” might have eliminated ideas that would have been outside the norm.

  18. Super interesting results and following comments. I’m personally comfortable with you taking a stance within the context of a blog post, and it makes the article more interesting (as well as spurs interesting conversation in response). Rather than being sensational, I think the tone you took is more one aimed at provoking people to learn more and be more active, and not just sit back and feel helpless. The one caveat I have would be, as somebody else said above, to be careful with the statement saying that banning bump stocks is just a distraction/scape goat. While I fully agree with you, I think you need to be clearer that you are asserting your opinion here, so that those who disagree are less likely to try to rip apart your entire analysis.

  19. To Rachel’s point on conjoint though, I wonder about the cards and levels, would number of lives lost be a variable… ?

  20. Thanks for sharing and taking a chance with your feelings Tom. Both your opinions and those in the comments show there is a need for this type of conversation in today’s America. As your other blog posts have done this topic again serves to illustrate the power of one question, and given the hot-button topic, also serves to demonstrate in a very real way the importance of question wording. Time well spent.

Comments are closed.