Sloan JH, Kellermann AL, Reay DT, et al. "Handgun Regulations, Crime, Assaults, and Homicide: A Tale of Two Cities." N Engl J Med 1988; 319: 1256-62.
methodological and conceptual errors:
- attempted a simplistic single-cause interpretation of differences observed in demographically dissimilar cities and cultures
- purported to evaluate the efficacy of Canadian gun control without evaluating the situation before the law
- the Vancouver homicide rate increased 25% after the institution of the 1977 Canadian law
- failed to acknowledge that, except for Blacks and Hispanics, homicide rates were lower in the US than in Canada
Sloan, Kellermann, and their co-authors attempted to prove that Canada's gun laws caused low rates of violence.  In their study of Vancouver, the authors failed to compare homicide rates before and after the law. As Blackman noted,  they had ignored or overlooked that Vancouver had 26% more homicides after the Canadian gun ban, an observation that should warrant scientific exploration and generate a healthy skepticism of the authors' foregone conclusions. Blackman's critique and analogy were so "on target" as to be amusing:
"... The Vancouver-Seattle 'study' is the equivalent of testing an experimental drug to control hypertension by finding two ordinary-looking, middle class white men, one 25 years old and the other 40, and without first taking their vital signs, administering the experimental drug to the 25-year-old while giving the 40-year-old a placebo, then taking their blood pressure and, on finding the younger man to have a lower blood pressure, announcing in a 'special article' a new medical breakthrough. It would be nice to think that such a study would neither be funded by the taxpayers nor published in the [New England Journal of Medicine]." 
Since its publication this article on gun control is among those most frequently cited, though this small scale (two cities) study has been thoroughly debunked by three large scale (national and multi-national) studies.    Kellermann and Sloan's biased interpretation of their data, asserting that guns are to blame for crime, assaults, and homicide, is even refuted by their own statistics.
Kellermann and Sloan glossed over the disparate ethnic compositions of Seattle (12.1% Black and Hispanic; 7.4% Asian) and Vancouver (0.8% Black and Hispanic; 22.1% Asian). The importance? Despite typically higher prevalence of legal gun ownership amongst non-Hispanic-Caucasians in the US,  the homicide rate was lower for non-Hispanic-Caucasian Seattle residents (6.2 per 100,000) than for those in adjacent Vancouver, Canada (6.4). Only because the Seattle Black (36.6) and Hispanic (26.9) homicide rates were astronomic could the authors make their claim. [See Graph 14: "Ethnic and Racial Groups -- Seattle and Vancouver" & Graph 15: "Homicide Rates by Ethnic and Racial Group -- Seattle and Vancouver"]
Could guns have some special evil influence over Blacks and Hispanics, but not others? Hardly! The authors failed to identify the inescapable truth. The roots of inner-city violence lie in the disruption of the family, the breakdown of society, desperate and demoralized poverty, promotion of violence by the media,   the profit of the drug trade, the pathology of substance abuse, child abuse, disrespect for authority, and racism -- not in gun ownership.
For an even-handed and scholarly cross-cultural comparison of guns, violence, and gun control, the reader is referred to Kopel's compendium.  If one reviews homicide and suicide data, despite high levels of gun ownership and high levels of gun control, the US fares well in comparison with many countries, even those supposedly "non-violent" nations whose gun controls the US is invited to emulate, such as Japan. How do US homicide, suicide, and intentional fatality (combined homicide and suicide) rates compare with other nations? [See Graph 10: "International Suicide Rates Comparisons"; Graph 16: "International Homicide Rates Comparisons"; and Graph 17: "International Intentional Fatality (Homicide+Suicide) Rates Comparisons"] Certainly the determinants of the levels of violence in a society are many and complex.