If there is one thing that analysts, historians and journalists can agree on, it’s that the total number of people murdered in Mexico in drug-related violence since 2006 — when the government began its infamous crackdown on drug cartels — isn’t exact. And some would argue that a precise figure may never be known.

Reports rely on different media outlets for their figures, and those outlets in turn may use varying formulas or statistics. Factors include murder trends in their home states and how the Mexican government classifies drug-related homicides — the latter has prompted a fierce debate over what deaths should be included in a data set.

Some media outlets and think tanks rely solely on the Mexican government’s data, which says that about 47,500 people have been killed in drug-related violence since 2006. But that data hasn’t been updated since fall 2011, leaving many to create their own current figures based on varying criteria. For example, the Trans-Border Institute at the University of San Diego, whose figures are cited in this report, relies on numbers from Grupo Reforma, one of Mexico’s major media outlets.

“Depending on what time period and what level of aggregation we are talking about, we estimate that Reforma is usually about 25 percent under what the government reports,” said David Shirk, the director of the Trans-Border Institute. “That’s been consistent over the last few years. So if Reforma says there were 10,000 killings so far this year, we think that it’s pretty safe to assume that there are at least 12,000 or probably 12,500 killings that the government will identify.”

Shirk concludes that Reforma isn’t skirting the issue but instead lacks the resources to cover Mexico’s 31 states and its federal district.

“Think about the limits of a news staff compared to the resources of the entire Mexican federal government for tracking homicides,” he said. “We estimated that the total so far by June was about 6,300 killings this year, putting the total at over 56,000 since the start of the Calderon administration.”

Use the drop-down menu to select a state and compare the government's homicide data with the Trans-Border Institute's statistics.

The TBI says that the Mexican government, which classifies drug-related homicides as “homicides allegedly caused by criminal rivalry,” uses six criteria for classifying a murder as drug-related. To qualify as such, two of the six must be met. They include: victims killed by high-caliber firearms; victims with signs of torture or severe lesions; victims were found at the crime scene or in a vehicle; victims whose bodies were taped, wrapped or gagged; if the murder happened in a prison and involved criminal gangs; or one of several “special circumstances” occurred, including an ambush, a chase or if a written message was left with the body.

“So what the media reports on is a subset of the known number of homicides that are believed to be associated with organized crime,” Shirk said. “So it’s, at best, a rough metric of the overall trend.” Complicating the matter further is that journalists in Mexico are under constant threat.

“There are places like Tamaulipas, [across the Rio Grande from Laredo and the Rio Grande Valley] where coverage is very sparse,” Shirk said. “We don’t know what’s going on because every body is afraid to be there, so I think we have to be very careful when we make assumptions or draw conclusions from the data that we are seeing.”

Molly Molloy, a researcher and librarian at New Mexico State University in Las Cruces, N.M., has chronicled the homicides in Ciudad Juárez, across the border from El Paso, since violence skyrocketed there in 2008. Her totals are based on what is reported in El Diario de Juárez, the Mexican city’s major newspaper. Like Shirk, however, she said the data is a rough estimate at best.

“They [government officials] don’t really tell us how they devise [a figure] other than to say they count certain things as being organized crime-related, and they don’t count certain things,” she said. “But so many of the crimes are not investigated, 95 percent of them by the government’s own figure.” I think all the homicides should be counted because the violence has such a character, such an influence on society.”

To get to the current estimate, Molloy said she has taken the monthly average as noted by Mexico’s Secretaría de Seguridad Publica (the federal public security agency), about 1,700, and added that to the Mexican government’s figures as of last September, for a total of about 62,800. Molloy said she has seen other media outlets calculate that 60 percent of Mexico’s homicides are attributable to the drug wars.

According to the country’s statistical agency, Instituto Nacional de Estadistíca y Geografía (INEGI) and the Secretaría de Seguridad Pública, there have been about 99,660 homicides since 2007, 60 percent of which is about 59,800, close to the 62,800 figure Molloy says is a decent assumption. That still represents a 3,000-person discrepancy, however, illustrating how difficult determining an exact number is.