How Non-Facts Become Facts
August 11, 2005
I am as frustrated with society as a pyromaniac in a petrified forest.
- A. Whitney Brown
Not long ago the editors of a prominent journal decided not to accept a manuscript submitted by this author because it lacked “empirical support” for its premise. In fact others majority who reviewed the article agreed that the data presented did not provide enough “empirical support” for some of the claims by this author.
While it is difficult to know how or if the article may not have been “professional” enough for the journal, there remain questions about just how “empirical” some of the “empirical support” in domestic violence research is.
Is scientific research always “factual?” Or is it possible, as Benjamin Disraeli once observed, “There are three kinds of lies: lies, damned lies, and statistics?” Is it possible that a researcher’s social bias and political agenda can be more important than objectivity and that some researchers set objectivity aside in favor of their pursuit of gender equality (Macionis, 1997)?
The “fact” is that scientific citations do not always provide “empirical support.” It is important that professionals, researchers and domestic violence advocates understand that some times scientific evidence presented in prestigious journals may not always be synonymous with the truth.
Donald Dutton, his work document he is an unbiased domestic violence researcher, believes there is what he labels an academic domestic violence research paradigm:
I mention this because it appears to me that a scholarly paradigm has developed where the same group of authors mutually cites each others work and generate one model of family violence; the father is the batter, the mother is the victim, the child is victimized by observation of the father’s violence. This is the essence of an academic paradigm. A social reality is created that directs belief and focus of future research and disregards
conflicting data (Dutton, 2005)
Central to the search for the truth must be scientific objectivity and a lack of bias. To reach the truth, most researchers agree, the researcher must be or at least attempt to be, objective and value free. However, feminist researchers, because of what they are, must believe that most scientific research has ignored women and because of that past ignorance their contemporary research must always focus on the condition of women’s rights being surprised by men.
Hence, feminist research can not objective because it demands the feminist researchers be biased. Feminist researchers, because of their social and political beliefs and their personal and professional objective, provide us with research that is primarily in pursuit of gender equality and this hinders not helps their search for the truth (Macionis, 1997).
Central to their domestic violence research and concerning the issue of domestic violence it seems, is their need to marginalize and minimize heterosexual male victimization at the hands of heterosexual females.
The search for the truth for some feminist researchers and all gender feminist researchers begins and ends with the belief that women are the victims of domestic violence and men their victimizers.
The gender feminist researchers can acknowledge homosexual domestic violence because in homosexual domestic violence the perpetrator is a male and in lesbian incidents the female perpetrator is acting like a male. In child, sibling and elder abuse the female perpetrators are acting out their violence against others because of their past victimization at the hands of males or their witnessing victimization of other females at the hands of males.
Rather than observe, recognize or become enlightened by data to the contrary gender feminist researchers simple minimize, marginalize or ignore facts that are contrary to their political and social agenda (Dutton, 2005). The gender feminist researcher will recognize only data that agrees with their beliefs.
If data does not agree with their firmly established position, men are the perpetrators and women their victims, that data will be ignored or when it can not be ignored, the gender feminist will claim that specific data is flawed.
A Non-Fact
The Journal of the American Medical Association lists on its website the key and critical objectives of JAMA http://jama.ama-assn.org/about_current.dtl. It lists that its number 1 objective is: “To maintain the highest standards of editorial integrity independent of any special interests.”
In their article, “Risk Factors For Injury To Women From Domestic Violence,” somehow the journal has missed its mark by a very wide margin. This author is not sure if JAMA dropped its high standards or has given into a special interest group. Regardless, this particular article does little to lend credibility to JAMA.
The following sentence appears twice on the first page of above article presented in the December 16, 1999 issue of The New England Journal of Medicine (JAMA): “Domestic violence is the most common cause of nonfatal injury to women in the United States.” The article is about injuries to women, and as do most domestic violence articles, it excludes injuries to men http://content.nejm.org/cgi/content/short/341/25/1892.
This domestic violence injury non-fact has been written many times and so often presented as fact by the electronic and print media, it has actually become accepted as a fact by the general public and as this JAMA article documents, this non-fact is also accepted as fact by many professionals, researchers and domestic violence advocates. It will also be cited by many other professionals and hence it begins the cycle of non-truth.
What should be particularly troubling to everyone is that this domestic violence injury non-fact is presented as fact by one of the more prestigious medical journals in the world. What is even more troubling is that data that escaped these researchers that refutes their claim is only a mouse click away. Research that once could take days or months is now, because of internet search engines, is only minutes or hours away.
It is co-authored by 9 medical doctors and two staff members. Once JAMA reports that domestic violence is the most common cause of nonfatal injury to women in the United States, the JAMA article itself will be used to cite that non-fact as fact.
If you read the article or visit the abstract on the JAMA website you will see that this article has been cited by 14 other articles. Now researchers can provide 15 instances of “empirical support” to document this non-fact as fact. This non-fact to fact to “empirical support” process is not very complicated.
The JAMA article has three citations for its “most common injury” claim. The first citation was: “Family and other intimate assaults – Atlanta, 1984. MMWR Morb Mortal Wkly Rep 1990;39:525-9.” Results are at http://www.cdc.gov/mmwr/preview/mmwrhtml/00001707.htm.
What is a fact is that the above study provides little to no data to support the JAMA researchers claim. It is a very small study and the majority of the participants are African American women living in an urban setting. The report contains only a total of 150 reported non-fatal incidents. That information is gleaned from police reports and presents no comparison with other non-fatal injury reports.
The only information in the MMWR study that might be seen as supporting the JAMA researchers claim is a paragraph that notes another study that claims domestic violence is responsible for more injuries than motor vehicle accidents, rape, and mugging combined. However, the MMWR study includes a warning about that study the JAMA researchers chose to ignore.
In fact, the MMWR study warns that this other small study also takes place in an inner city emergency room where the population is almost exclusively African America women living at the lower end of the socioeconomic strata. These women from an inner city minority community do not accurately represent a cross section of America women. In fact the rate for injuries in the MMWR study notes that African America women were injured three times more often than white women.
The second citation in the JAMA article is: “Grisso JA, Wishner AR, Schwarz DF, Weene BA, Holmes JH, Sutton RI., A population-based study of injuries in inner-city women. Am J Epidemiol 1991:134:59-68.” An abstract of the results from that study can be found on line at
http://aje.oupjournals.org/cgi/content/abstract/134/1/59.
These JAMA article researchers appear to ignore the fact that the study above they cite actually demonstrates that the major cause of injury to women were falls not domestic violence. Further, it reports “that very little is know about nonfatal injuries to women.” It concludes that, “More work is needed to understand the nature of injuries occurring to young women in urban communities.
This study also does not provide a cross section of American women. The majority of the participants are from a poor, urban, African American community. For any study to be relevant nationally that study must collect data from a sample population that represents the entire population. These two studies clearly do not meet that very basic random sampling standard needed to produce accurate results.
How can these JAMA article researchers conclude that the information in either of these first two studies is empirical scientific data that provides “support” for their domestic violence injury claim? The almost total lack of “empirical support” for their premise in these first two citations does not bode will for the objectivity of these researchers.
In fact, the third citation the JAMA article cites is worse, as difficult as that seems, than the first two: “Stark E, Flitcraft A., Spouse abuse. In: Surgeon General’s workshop on violence and public health: source book, 1985: Centers for Disease Control, 1986:AS1-SA43.” What is a fact is that the works of Stark and Flitcraft document they are gender feminists and “in fact” this fact was never presented as “fact” by the Attorney General.
The Centers for Disease Control (CDC) is on record that they will not recognize this “fact” as being presented by them or the Attorney General as a “fact” at the conference http://www.mincava.umn.edu/documents/factoid/factoid.html. The JAMA article researchers either ignore or are unaware of the CDC disclaimer.
What is more surprising than the non-fact citations presented in the JAMA article, is the fact that these 9 doctors and two staff members failed to notice that two out of their three citations lead them to the CDC.
The Facts
How is it possible that these domestic violence injury claims that are so wrong, could be accepted by JAMA? How is it possible that out of nine medical doctors or two staff researchers involved in this JAMA article, that not a single one of them thought to question data that is so easily refuted? In fact, they make claims that their citations do not.
All that researchers and others who are concerned about this nonfatal injury claim have to do is to visit the CDC website to find out that their claim that domestic violence is the most common cause of nonfatal injury to women in the United States is without “empirical support” http://www.cdc.gov/ncipc/wisqars/.
The CDC website http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5017a4.htm documents the estimated number of nonfatal injuries treated in hospital emergency departments for the year 2000. What caused these doctors and staffers and what causes so many other people to be so wrong about or blind to this real and relevant data?
The CDC website documents unequivocally that domestic violence is not the most common cause of nonfatal injury to women in the United States. Just a cursory view of CDC data documents that falls cause the most injuries to women. In fact there are a number of unintentional injuries listed before any type of intentional injury, intimate partner or not, is listed.
There is not a single scientific study anywhere in the United States, or in fact elsewhere, that document the JAMA domestic violence claim to be true. The truth does not seem to stop domestic violence advocates from repeating the JAMA domestic violence claim over and over again and it can be found on many domestic violence websites.
The findings from the National Violence Against Women Survey, the Full Report of the Prevalence, Incidence, and Consequences of Violence Against Women, document that men and women are nearly twice as likely to be injured on the job than during a rape or physical assault (Tjaden & Thoennes, 2000).
The Bureau of Justice Statistics Special Report, “Violence-Related Injuries Treated in Hospital Emergency Departments” documents that all violent assaults account for only about 3% of women’s injury related visits to emergency rooms. Of those approximately one third of violence-related injuries reported by females were committed by their intimate partner.
Fallacious Facts
This JAMA article establishes another less than “reputable citation” for gender feminists and others who want to continue with this hoax. The truth is that these researchers for this JAMA article, did not present a single citation that can actually document their domestic violence injury claim has “empirical support.”
How is it possible that these doctors, researchers and a prestigious medical journal remain so unconcerned or uninformed about such an obvious truth? Is it possible that the gender feminist ideology has become more important to some domestic violence researchers than the truth?
______________________________________________________________________________
Richard L. Davis (Lt.ret) is the author of Domestic Violence: Fact and Fallacies, an adjunct instructor of criminal justice courses for Quincy College at Plymouth and the VP of www.Familynonviolence.org. He may be reached at rldavis@post.harvard.edu
