Updating the 1315 Chart

Drug Spending vs Past Month Use Graph

BLOG:  When my animated chart recently went somewhat viral, there were some significant, and on-point, questions about how the Addiction Rate vs Federal Spending represented what the research for my film project has shown: spending on drug control has no correlation to drug usage in this country. So I decided in the interest of transparency to recreate the chart using only metrics with online resources available to verify the data.

This required a few changes.

  • first, I’ve switched from Addiction Rate to Past Month Prevalence (ie. how many people have used drugs in the past month) as that data is more readily available online even though, as I believe, Addiction Rate is a better indicator of the severity of the drug problem;
  • second, I have adjusted all financial totals for Federal Drug Control Spending to 2012 dollars;
  • third, I declined to input Federal Budgets from 2004-2009 because of reporting inconsistencies in those budgets (explanation here);
  • fourth, I made some estimates based on marijuana data ( approximately 90% of all drug use in the 1970s) in order to come up with an estimate of total drug use for 1973 and 1976;
  • lastly, I have charted everything as it relates to 100 citizens.  I made the last choice because I did not think that a straight per capita relation would make sense.  Would people understand that 0.087 people per capita used drugs in the Past Month in 2011?  Or is it easier to understand that 8.7 people out of 100 used drugs in the past month?  I think the latter.

I would not be surprised if, in the eyes of statisticians and those more statistically inclined than myself, I’ve somehow still confused the proper way to do this.  If that’s the case, please let me know in the comments. Also, if anyone has some better resources to fill in any data gaps, I’d love to hear about it.

Additionally, I would love to find the costs, shown over time, that express the financial cost of drug prohibition on our criminal justice system and on our prison system. If anyone has a heads up on an online resource for that, please message me.  I feel confident, as Jack Cole expresses in my trailer, that the costs associated with this failed system of prohibition is at least $1.5 trillion over the course of the drug war.  One last note – the $545 billion figure is a tad low as it does not account properly for expenditures in the years 2004-2009 as noted above.

Here is a Google Spreadsheet with all the data sourced (opens in new window).

And below is the resulting graphic. (Click to enlarge and see the animated version).

Drug Spending vs Past Month Use Graph

6 Responses

  1. Pingback: Questions on The 1315 Project Chart | Matt Groff

  2. Pingback: Spending Money on the Drug War Really Isn't Lowering Drug Use | News 47News 47

  3. Andrew Gilbert says:


    As I told Serena Dai in an email, I respect what you are doing and the effort. I also have a lot of respect for your ability to make any film, something I would be completely incapable of.

    I am also somewhat sympathetic to some of the issues around this topic. I might be the first person to rail against mandatory sentencing guidelines that punish poor people disproportionally to rich, to big government programs running out of control, and to excessive investment in state control at the expense of personal liberties. The social and political chaos that is an untended consequence in many countries is disturbing.

    That said, I also have a big issue with pseudo – social science as a means to promote pre-conclusive agendas. And I fear that is more what you are focuing on and accomplishing than truly adding value to the discussion. This might sound harsh. But would ask you consider that it isn’t “closed” minded to hold someone to a standard of high integrity in their data and analysis. Quite the contrary, it is about as “open” minded as you can get. And not holding yourself to that standard puts you on par with just about every other big wind blowing progandist floating around these days.

    Some Lingering Criticisms:

    The area under the curve is a summation. Making up a number to put into the shaded area that isn’t backed by the data is mathematically incorrect. It is also somewhat unethical, if you realize you know what you are doing. Your numbers still don’t add up to $545B. The spreadsheet gives roughly $400B. If you adjust for the Bush year weirdo accounting, which is fine, than you are still only at $460B or so. I would ask, perhaps plea, in the name of objectivity and accuracy you please stop misrepresenting the actual data by making up a number to make things “look better”. I am no expert, but happy to point out some places where you can brush up on math and statistics fundamentals.

    A more minor point on data. I applaud your posting the actual sheets you are using to construct the data. But you are also clearly “interpolating” some values to actually produce the charts. I don’t have a problem with the interpolation, as long as you note and justify it. You appear to do this, which is great. But you don’t include the “interpolated” version of the data in the sheets. Might be more clear if you did so. I would still retain the raw data as well so people can see what you did and how you get where you got. By not having the interpolated sheet, one has to jump from raw data to a graphic, and the numbers don’t match. This raises alarm bells. It also makes it harder for others to play with things using the same basis as you did. See below about correlation.

    More on judgement. I would argue you are still somewhat grossly misrepresenting the $$ outlays. Per the ONDCP numbers / PDF, which is here http://www.whitehouse.gov/ondcp/the-national-drug-control-budget-fy-2013-funding-highlights, roughly 35% of the spending for 2011 is on treatment. If you are trying to make a statement about correlation and causation, I would question how you justify including these $$. They are just as much a reflection of the social “costs” of drug abuse as they are a “preventive” measure. This debate is a pretty emotionally loaded one, and one “meme” which is popular is that the “war” is “causing” drug use. I have a big issue with this leap to judgement. Fine if people want to make it, but there is more burden of proof than just screaming so. And by including treatment $$ in your graph you are both misrepresenting the effort to correlate (or not) preventive $$ outlays and use.

    Third, Serena Dai, in the Atlantic Wire posting, makes the statement that there is “clearly no correlation”. Actually, Serena, there is. Matt, your data shows this. Eh, at least there potentially was. If you look at 1976-1986 it is pretty significant. I tried to roughly estimate the values from your graph (as could not use your sheets – see above) and do a quick check. I got roughly -0.83. This is significant. The question then becomes, did the first incremental dollars make a big difference, and everything since then has been “wasted”. This would be an interesting question to pursue. Another possibility is the makeup of the dollars has changed in some way. Also potentially interesting.

    In the end, I would come back to the fundamental criticism. You are at significant risk if you goal is mainly to “prove a point” which you are already convinced is true. We all carry our biases and agendas around, which is fine. I have to make the assumpition your goals are noble. But the more you “believe” in something, to me anyway, the more care you need to exercise in being thorough, objective and as accurate as you can.

    Andrew Gilbert

  4. Bill says:

    Hi Matt,

    I used your graphic on our site with your explanation as to the numbers, thanks again for that and your comment about the update!

    I’m saying the following from the point of view that I’m really looking forward to seeing what you produce, but if this is going to really have the best chance at success, the numbers are going to have to be bullet proof to statistical errors/bias etc. This is simply because with wider distribution and hopefully MSM interest, ‘the other side’ is going to look to discredit it – and if there’s low hanging fruit like Andrew’s comment it won’t take long.

    Is it worth roping in someone with a statistical background – beg, borrow, steal their time?

    All the best and cheers,

  5. Pingback: Obama: Fighting State-Led Marijuana Legalization a 'Low Priority'

  6. Pingback: Drug War Effectiveness | infoGraphi

Leave a Reply