Older people often like to think back to the “good ol’ days” when everything was better. In many cases this is merely the feeling of nostalgia and seeing the past in a better light than it actually was. But of course, as with everything, sometimes this golden past did exist.
In this post I look at data to see whether there were more mass shootings in the past than today. The main question I’m after is are mass shootings becoming more common nowadays than they used to be in the past? The implications are huge: if they are becoming more common, then we must look at what changed over time. We will be closer to identifying the reasons for these terrible acts.
I look at the number of mass shootings from 1965 to 2013. It appears that before 1965, mass shootings were very rare. The reason for this could possibly be the unavailability of suitable weapons because of the absence of technology or their high price relative to the general populace’s income.
The data is the following:
What we can directly see is that the first half of the eighties and the second half of the 2000’s as well as arguably the 2010’s so far were quite bad periods compared to the rest. The last column (M/P) refers to mass shootings relative to population size. More precisely, it is the number of mass shootings per 100 million residents. Basically, that statistic takes it into account that in a larger society there are bound to be more mass shootings, simply because there are more potential shooters. When looking at that statistic it really is the early eighties and the late 2000’s that stand out and maybe the late seventies.
Let’s graph the above table to see whether these half-decades of higher mass shooting activity seem significantly worse than the rest:
You can clearly see two things: the total number of shootings (while quite volatile) seems to increase with time (blue curve). Despite this, the number of shootings per 100 million residents is rather characterized by stagnation (red curve). What one can deduce from this is that while there may be more shootings nowadays (and hence you’ll hear more about shootings on the news), but the number of mass shootings seems to be growing merely because of the increase in population.
Last but not least, let us look at the data and see whether the increase in the mere number of mass shootings over time is statistically significant. When regressing the number of mass shootings on time, the slope coefficient is not statistically different from 0; meaning that statistically speaking even the increase in the mere number of mass shootings is questionable (i.e. what we see may be nothing more than just some random fluctuation).
If we extrapolate naively and assume that from 2010 until the end of 2014 we will have (10 / 3.5) * 5 = 14 mass shootings, then the slope coefficient of the regression does become significant albeit only at the 10% level. What this means basically is that depending on future trends we will be able to tell whether there is an increase in the absolute number of mass shootings over time. Hopefully, the answer will be no.
If we look at the mass shootings per 100 million inhabitants data, then just as the graph above indicates, there is no trend whatsoever. The slope coefficient is effectively 0, even after doing an extrapolation similar to the one above (which by the way would yield an M/P of 4.53 in the last 5-year period). The bottom line: there is clearly no upward trend in mass shootings per capita as of now, which is good news.
What can we deduce from all this? Well, first of all that mass shootings seem to be quite random in some sense. There are terrible periods like the late 2000’s, but there are relatively peaceful periods like all of the nineties. What we also see is that, the number of mass shootings relative to the population size is definitely not increasing. Furthermore, even the absolute number of mass shootings has an ambiguous trend. One thing is for sure: we are in (or hopefully we’ve just been through) a rough decade of mass shootings. There is a high probability therefore that a more peaceful period is ahead of us, because what we’ve been experiencing lately is not normal, it is above average.
Of course, the current higher-than-average rate of shootings could be the beginning of a new era of more mass shootings. Alternatively, it could also be followed by a relatively peaceful decade (or hopefully decades and cenutries) just like the so-far worst 5 year-period (the early eighties) was followed by the most peaceful 15-year period in the sample.
So was the past a golden age? Certainly not in a black-and-white sense. But only future trends will tell whether our society has become more dysfunctioning over time. As of now, it appears that the past was no better than the present in terms of mass shootings.
What else could be done with this data? It could be interesting to put these numbers together with some soft facts like what kind of gun regulations and things like this were in effect in each period. Maybe one could find a correlation between the number of mass shootings and some other variable.