Terror Alerts vs Elections

Friday, October 8th, 2010

Election Time vs Terrorist Attack
Accusations today over the timings of terror alerts and elections.

The Bush administration used to raise threat levels around campaign time, apparently. Is Obama doing the same with European terror alerts to create a rally-round-the-president effect?

I wondered if there was a correlation between terrorism and elections we could actually see.

See the dynamic graph on Googe Insights

(It tracks the intensity of keywords over time- an interesting barometer of the group mind).

Note: Here I’ve chosen two relevant keywords that have similar intensity. In Insights, if you chose a keyword that’s too intense it dwarves the other and hides the pattern. Example.)

Any correlation?

Books and Store

Our Beautiful Books - Information is Beautiful Information is Beautiful Store

Show Comments ( )

  • http://www.vertissimmo.com/ Abi

    U know what ?

    The French think their governement raise threat levels in order to prevent them from marching and striking. It is quite funny to notice that you believe the US governement is to be blamed for making up terrorist threats in Europe. What’s the point ? It is not like the US were targeted as well !

  • Tim

    Interesting thing. I downloaded the csvs files directly of Google Insights for each keyword (election and terror alert) separately and checked for some correlation. If you lag the terror variable by 13 months you get a correlation of 0.6668 which is also highly significant.

    Here is the graph: http://i52.tinypic.com/16c4843.png

  • Josh

    What about the blue spike around ’08 that is not pre3ceded at all by a red spike?

    • http://stefanco.com Stefan Lasiewski

      > What about the blue spike around ’08 that is not pre3ceded at all by a red spike?

      I don’t really believe in this theory, but if I did I would say that the people who control the terror alerts (The Executive Branch) were not running for re-election in 2008.

      Really, the lack of “terror alert” probably has more to do people paying more attention to the near-collapse of our economy. Between September-November 2008, nearly every financial market suffered a major decline or 20-35%. That’s more then enough stress for anybody, and so things like “terror alerts” probably had less effect.

    • http://stefanco.com Stefan Lasiewski

      Also, check out this Google Insights graph. I added “Stock Market Crash” to the graph, which has a strong correlation with the 2008 elections.

      http://www.google.com/insights/search/#q=election%20time%2C%22terror%20alert%22%2C%22stock%20market%20crash%22&geo=US-CA-807&cmpt=q

      This was quickly done, so someone with more experience might be able to provide better insights.

  • http://robm.me.uk/ Rob Miller

    Since all federal elections in the US are held bi-annually and in November, is there any need for the “Election Time” line at all, given that it also has the potential to confuse matters — the odd little spike in October/November 2005, for example? Just label the elections on the x-axis: November 2004, 2006, 2008, 2010.

    There definitely seems a trend, though, and you can also see, by implication, the extent to which the 2008 election was dominated by the economy, rather than the security-focused elections of 2004 and 2006 (and, I’m sure you’d find if there was data, 2002).

  • Josh Sand

    Definitely a connection, but interestingly enough, there was hardly any at all during the ’08 election. Must’ve been a Bush thing. :P It would be interesting to see this same graph but with other topics included, such as climate change, gay marriage, the words “destruction of america”, etc….

  • http://robm.me.uk/ Rob Miller

    What about the blue spike around ’08 that is not pre3ceded at all by a red spike?

    The 2008 election was fought on the economy, not on security or foreign policy, so there was little discussion of the terrorism threat level.

  • James

    A little strange how the assumption is made that terror alerts are *caused* by the US government.

    Couldn’t a more plausible, Occam’s razor type of explanation be that terrorirsts are timing their plots to impact elections??

    Al Qaeda did this to Spain immediatley before an election and got their conservative govt booted and Spain to withdraw from Iraq.

  • Cass

    Why would there have been a relation in 2008? The terror alert is controlled by the current president… in the 2008 election the current president wasn’t campaigning.

    Any apparent correlation could be exaggerated by a self-fulfilling prophecy. What percentage of “terror alert” searches right now would be fueled by headlines like “Barack Obama accused of exaggerating terror threat for political gain?”

  • MrKern

    The is some research on this topic:

    http://en.wikipedia.org/wiki/Terror_management_theory

    sample:

    The Effects of Mortality Salience and Reminders
    of 9/11 on Support for President George W. Bush
    http://www.psych.ndsu.nodak.edu/hilmert/SPRG/Landau_2004.pdf

  • chuckarama

    It’s possible that we’re just missing the third dataset. We don’t know, because the gov’t owns the monopoly on it. It _is_ possible that terrorists would want to interrupt elections, so the spike before elections could actually be due to real data, but we don’t know, so we’re left to speculate on the root motivations of the “terror threat”. As with all things gov’t, it’s dubious and speculative.

  • tf

    “In a new book, former Secretary of Homeland Security Tom Ridge reveals new details on politicization under President Bush, reports US News & World Report’s Paul Bedard. Among other things, Ridge admits that he was pressured to raise the terror alert to help Bush win re-election in 2004.

    The Bush administration was forced to admit in the days after the 2004 alert that it was based on intelligence three or four years old. Officials then claimed there was a previously unmentioned “separate stream of intelligence” that justified the warning — but offered little tangible information to support their new story.”
    http://www.huffingtonpost.com/2009/08/20/tom-ridge-i-was-pressured_n_264127.html

  • http://espliego.wordpress.com/ cantueso

    Bush never stopped selling fear. I don’t think it was his own idea or initiative. As an idea it is very primitive but efficient, more like a Cheney thing.

    Members of the police here in Spain used to tell me that they kept getting warnings to be careful, but there was never any indication what they had to look out for.

    Besides, worse, online I kept seeing ads saying: IS YOUR FAMILY SAFE FROM TERROR? plus a photo of happy family life and a telephone number to call and find out what 10 steps you can take to make the lives of your beloved ones safer from terror. And of course the cartoonists drawing ladies who looked for terrorists under their bed or under the bath tub.

  • http://www.lonelyshouters.com Sr. Lobo

    The Power of Nightmares by Adam Curtis (BBC) ;)

    • Pugz

      I second that. That documentary is the first thing I thought of when I began reading this. It’s available on youtube. Watch the first 2 and a half minutes to get an idea.

      This is absolutely essential viewing, along with everything else that Adam Curtis has ever made.

      Here’s the link:
      http://www.youtube.com/watch?v=Vt-FyuuWlWQ

      Watch all parts of all the volumes.

  • Matt

    Don’t forget to consider reversing the premise, terrorists may want to cause chaos during the election season.

  • Colin Sachs

    While interesting, this data doesn’t provide any meaningful insight into the potential manipulation of the threat level to affect the outcome of elections.

    Since Google Insights data is derived from “…how many searches have been done for the terms you’ve entered, relative to the total number of searches done on Google over time.” (http://www.google.com/support/insights/bin/answer.py?hl=en&answer=92768) Given this, one can only conclude that there might be a correlation between number of key word searches of “terror alert” and number of key word searches on “election time” by Google users. This only would provide an potential indication of what Google users were focused on finding more information on a given set of topics at a given time.

    This informs us about the interests of the Google users, but nothing whatsoever regarding
    “… the timings of terror alerts and elections.” It is disingenuous to imply otherwise.

  • http://twitter.com/MabandeOnDesign André

    Is Obama doing the same with European terror alerts to create a rally-round-the-president effect?
    Question is: does the terror-talk get people to rally round the prez or do they rally round Republicans?
    Republicans do seem too target the Obama-administration for being soft on terrorism, y’know, so if there’s a terrorism-threat people might just go to the guys who promise the simplest and most “effective” solutions (more guns, more war, hoo-hah!).

  • SebastianH

    Beside the increased alerts and reporting on terror close to elections, there might actually be a substantive relation. While controverse, this was discussed at the 2004 Madrid train bombings which took place three days before the Spanish general elections.

    The argument was that terrorists supported one party over another and chose the timing in order to actively influence the elections.

  • http://ilrespirodellaconsapevolezza.it Luca De Santis

    Well,i was telling the same thing few days ago when i saw the terrorism allarm rising just in the same time of the strike in France…Strange?

  • Mike C

    Interesting analysis. I caution, however, drawing conclusions about political motivations in determining when / what terror alerts to issue because we don’t know how actual terror activity is correlated to political seasons. As an example, look at the Madrid bombings – conducted in a political season to bring about a political reaction (and successful at that). So, maybe terrorists increase their activity around elections and governments respond with more alterts

  • http://www.blackgreengames.com Emily

    As the last comment mentions, the incidence of violent terrorist activity may well coincide with elections. The current US election season is seeing a spike in mail bombs. I’d like to see information on these events side by side with the other. That does bring up the question of how to measure that though.

  • http://pid=1063120&profile_id=6545257&profile_name=trevorruiz512&user_id=6545257&username=trevorruiz512 Carson Kapp

    Hey There. I found your blog the use of msn. That is a really smartly written article. I will make sure to bookmark it and come back to learn more of your helpful information. Thanks for the post. I’ll certainly comeback.