Earlier in the month CNN published an article proclaiming "Juarez Shedding Violent Image, Statistics Show." With the inclusion of "statistics" in the title I was immediately drawn to investigate and see exactly what the article was claiming, and what sort of data they were using to reinforce those claims. The article begins by stating that "statistics from the Chihuahua State Attorney General's Office show that Juarez ended 2012 with the fewest killings the city has seen since drug-related violence encapsulated it five years ago." Here is some of the data they cite concerning death tolls in 2012 as opposed to other years:
-750 homicides in 2012 versus 2,086 people killed last year.
-Between 2007 and 2011 more than 9,000 people were killed, with the peak coming in 2010, when Juarez saw a record 3,116 homicides, or about 8 murders per day, according to figures released by the attorney general's office.
-In 2007, there was an average of about 0.82 murders per day. In 2012, there were two murders per day, according to the attorney general's office.
After reading this article I decided to consult other news sources to see what they had to say about the violence in 2012.
From July 14, 2012 the El Paso Times reports in their article "Juárez slayings decreased 59.8% first half 2012":
-The number of slayings in this city of 1.3 million during the first six months of this year dropped 59.8 percent compared with slayings in the same period of 2011, according to figures from the Chihuahua state prosecutor's office.
-Statistics on the state's prosecutor's website indicate there were 1,623 killings between January and June 2010, and 1,322 in the first half of 2011. This year, there had been 653 slayings as of June 27, the website said.
-Zarate said there were 952 killings from January through June this year, compared with 1,642 during the same period last year.
On January 13, 2012 the NY Daily Times reports in"Mexico releases drug war death toll estimate: One killing every half hour":
-The toll linked to the anti-cartel campaign launched by Calderon in 2006 continued to climb; there were 12,903 killings between January and September of 2011. Total deaths stand at over 47,515, according to the government tally.
-Ciudad Juarez, across the border from El Paso, Texas, had the highest number of violent deaths at 1,206, and from September 2010 to September 2011, the number of drug war-related deaths rose 11% compared to increases of 70% and 63% in previous years.
My thoughts on these reports:
First and foremost it was difficult to find many articles or reports published in 2013 as opposed to 2012: we are comparing partial data from 2012 with one reported set of data published in 2013. Also, the availability of data concerning this topic is of great concern, which the NY Daily News reports in saying "the Attorney General's Office posted the figures after months of prodding from human rights groups, the press and a disgruntled state agency." Thus the accuracy of the data could thus be called in to question simply by the fact that it was difficult to obtain. If the Mexican government was truly so hesitant to release the statistics on deaths in Juarez, as the NY Daily News claims, you could speculate that the numbers might be even greater than what is officially reported.
Also, there seems to be some discrepancy in the data concerning 2012 as CNN claims that there were "750 homicides in 2012" while the El Paso Times quotes Army Gen. Emilio Zarate in stating that there were "952 killings from January through June this year." Another issue that concerns me is that the articles do not do a very good job of categorizing and defining "murders" in Juarez. Likely they are implying that all murders in discussion are drug-related, but this is never stated specifically, and furthermore the category of "drug-related deaths" is nebulous and could encompass a multitude of definitions. Couldn't "drug-related deaths" refer to those killed while selling drugs just as easily as people killed using drugs? And what about the relatives or friends who are so often caught in the cross fire-- aren't those deaths "drug-related" as well? Thus while CNN applauds the drop in homicides in Juarez in 2012 it also credits this drop to things such as "the withdrawal of federal agents -- often accused of corruption," and a "team effort and coordination between local police and civil society organizations." This in turn implies that all homicides in Juarez are directly connected to the drug war. While the decrease in homicides in 2012 is no doubt a good thing, to me it seems a mistake to directly correlate it to a decrease in drug-related violence and conflict between cartels.
Sources (all accessed 1/28/13):
- "Juarez shedding violent image, statistics show," CNN.com, January 05, 2013
- "Juárez slayings decreased 59.8% first half 2012," ElPasoTimes.com, July 14, 2012
- "Mexico releases drug war death toll estimate: One killing every half hour," NYDailyNews.com, Friday, January 13, 2012.
Thanks for writing, Julia. I like your analysis of the situation. One question I would like to raise is if their numbers (assuming they are correct, which is something you are questioning) are statistically dissimilar from one another. What is the standard deviation of homicides in Juarez over the past 15 years? In other words, are these fluctuations significant?
ReplyDeleteUsing the data provided by http://justiceinmexico.org, I calculated the standard deviation (I assumed a perfect bell curve) of Mexico's national homicide statistics from 1997-2012 to be approximately 3665. This means that fluctuations of plus or minus 3665 from year to year are not significant, as they lie within 1.0 sigma (47.7%) on the bell curve. For the year 2011, the total number of homicides was 37,409 and for 2012 was 30,488. The difference between these two years is 6921, or about 1.9 sigma. Normally, figures are only considered to be statistically significant if above 2.0 sigma (95.4%+).
Mexico might be on the right path to fixing their country, but only time will tell if this will be a lasting trend.
Thanks,
-Lee
Sources:
http://justiceinmexico.org/data-portal/2480-2/
http://en.wikipedia.org/wiki/Standard_deviation
Julia,
ReplyDeleteI think you brought up a really good point in stating that political agendas might be to blame for this discrepancy in numbers. I think political agendas can be huge motivators for people to either consciously or unconsciously bias the statistics that they are presenting. While hesitancy to release the number of homicides may be the motive influencing the discrepancy in this data, there are a lot of other political motives that can likewise influence statistics. The example that immediately comes to my mind deals with the recent presidential election in which we were bombarded with statistically backed claims from both political parties. Differing political agendas in that instance oftentimes resulted in the same situations being described using completely opposing statistics. Before we can trust the validity of any statistical or scientific claim, I think it is important to recognize these types of biases, political or otherwise, that may have influenced the collection and interpretation of data.
Julia, I really enjoyed your anaylysis along with the topic. Personally, as a mexican i get to hear more and see more than what a lot of american newspapers portray. A big part of the number discrepancy can be seen through the amount of dead-bodies found. If a body isnt found, the murder is not accounted for. Or for example, if a lot of bodies are found a certain day, even if the bodies have been dead for months, those bodies would be accounted on the month the body was found. Great topic. See you in class. By the way, we could argue somehow that assuming political biases firsthand is a bias, dont you think
ReplyDelete-Xavier Maqueo (7)