Studies Show Guns ADD Risk of Negative Outcomes – The Standard Model Part 3 of 5 (Light Over Heat #43)

This video continues my series systematizing the dominant academic approach to understanding Gun Culture 2.0, what I call “The Standard Model of Explaining the Irrationality of Defensive Gun Ownership.”

The model has 6 points, and in this 3rd video, I discuss point 3: how guns are seen to add risk of negative outcomes.

Links to videos 1 (Light Over Heat #41) and 2 (Light Over Heat #42) are below.

ACADEMIC TRIGGER WARNING: I got carried away discussing the methods employed in public health research on guns as a risk factor (I am a professor, after all), so this video is longer and more tedious than average. AND I also took the last third of the video in which I critique the public health research and put it in a separate video that will run next week.

In this video and the next I focus in particular on an article by David Studdert and his colleagues from the Annals of Internal Medicine in June 2022, “Homicide Deaths Among Adult Cohabitants of Handgun Owners in California, 2004 to 2016: A Cohort Study.”

It is an excellent model of how scholars attempt to isolate guns as a risk factor when the gold standard of causality — the randomized controlled experiment — is not possible.

Please surf over to my “Light Over Heat” YouTube channel and SUBSCRIBE to follow, RING THE BELL to receive notifications, and SHARE so others can learn about this work.

If you are able to support this work financially, please consider making a contribution:

Donate with PayPal button

Buy Me A Coffee


  1. 3 videos for the price of one! I feel like I am back at the University sitting for a full seminar! Cool!
    (Trigger reset warning: of course like David, I am a member of the academic community).


      • 1.
        The authors’ admission that “we could not determine whether the nonowners in our sample were killed by the actual handgun owners or handguns that defined them as exposed….” is sufficient to disregard this study. They also dissemble: the supplement records that only 46.2% of perpetrators were a domestic partner or family member of the victim.

        That 53% of the homicides occurred away from the home “was an unexpected result, not clearly related to the handguns we used to define the exposure in our study….” So they simply excluded those cases to bring their results more in line with their hypothesis. They also chose to count all 173 homicides where the location could not be determined (7.5% of the total) as occurring in the home.

        “Firearms were used in 65.2% of all homicides.” The other 34.8% should have been discarded, but were not.

        “[C]ohabitants of handgun owners had substantially higher rates of homicide at home…. Rates of homicide occurring away from home were also higher.”

        This clearly indicates an elevated environmental risk of homicide unrelated to the presence or absence of a firearm in the home — a huge potential confounder not once addressed by the authors.

        “[F]or every 100 000 nonowners whose cohabitant acquired a handgun, 4.03 more died by firearm homicide in the ensuing 5 years than died among nonowners whose household remained gun-free.”

        That’s .0008% per year. This is described as a “substantially elevated risk.”

        The authors also deliberately left out 3.2 million non-owners who lived in a home containing a gun at the start of the study. No rationale was offered for this exclusion. Were the risk of falling victim to homicide to prove lower in this group, it would undermine the premise that exposure to a gun, and not environmental factors (which might motivate first-time acquisition of a firearm), is the cause of homicide victimization by a cohabitant.

        “[A] putative confounder … would need to both increase the risk for homicide by firearm by a factor of 8.35 and be 8.35 times more common among cohabitants of owners than cohabitants of nonowners.”

        That factor is entirely plausible, if a small subset of the owner-cohabitant cohort both has elevated homicide rates and is exposed to the confounder. If any attempt was made to break out data by demographics, it was not reported. In any case, sex and ethnicity were imputed by an arcane calculation and assigned to individuals as fractions, e.g., “20% white, 60% black, 15% Hispanic, 1% Asian, and 4% other.”

        “[W]e only partially accounted for long gun ownership, although this omission is unlikely to have a large effect on our estimates for 2 reasons: Fewer than 20% of firearm owners in California own only long guns and handguns are used in approximately 90% of homicides by firearm.”

        The inclusion of 1/5th more subjects with 1/10th the homicide rate would have a rather large effect. Comparing long-gun-only, handgun-only, and mixed-type households might also be informative, and lead to an examination of the demographic differences between owners of long guns vs. handguns as a likely confounder of the handgun-only data the authors chose to focus on.

        Further, the authors would be at pains to explain how, in households with both long- and handguns, the presence of the handgun increases risk of homicide, but the presence of the long gun does not.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.