Distortion:
"Research shows that kids who use marijuana weekly are nearly four times more likely than nonusers to report they engage in
violent behavior."
— Office of National Drug Control Policy, Marijuana Myths & Facts, from the web at
http://www.whitehousedrugpolicy.gov/publications/marijuana_myths_facts/myth4.pdf" target="_blank">http://www.whitehousedrugpolicy.gov/publications/marijuana_myths_facts/myth4.pdf, last accessed April 13, 2006.
This is a misinterpretation of research. ONDCP claims an association between drug use and anti-social behavior. They imply yet don't specifically claim causality and quote parts of some research while omitting findings that don't support their assertion.
In this myth, the ONDCP claimed to be supported by two studies:
Greenblatt, J.
Adolescent selfreported behaviors and their association with
marijuana use. Substance Abuse and Mental Health Services Administration (SAMHSA). Based on data from the National Household Survey on Drug Abuse, 19941996,
1998.
Friedman, AS; Glassman, K; Terras, A.
Violent behavior as related to use of marijuana and other drugs. Journal of Addictive Diseases. 20:4970,
2001.
First, the Greenblatt piece published by the
Substance Abuse & Mental Health Services Administration. She observes near the beginning of the article that
"In completing the YSR [Youth Self Report], youths were asked to read the list of 118 statements and indicate if the statement was not true, somewhat or sometimes true, or very or often true for them. Although causal conclusions about the relationship between substance use and problems cannot be drawn from the NHSDA data alone, these data provide a useful complement to other studies. While the reported behaviors are not necessarily caused by the use of marijuana or, conversely, the cause of marijuana use, there appears to be a strong positive correlation between the reporting of certain behaviors and reported frequency of marijuana use. The more frequent the use, the more likely the 12-17 year olds were to report problem behaviors."
Her report actually concludes:
"This report shows that among those age 12-17, past year marijuana users were more likely than nonusers to report problem behaviors in the past 6 months. Further, for the majority of items measured, the more frequent the use, the more likely the youths were to report problem behaviors.
"The more frequent users were more likely to be the older youths (6 out of 10 were age 16-17), white, male, to live in a metropolitan area and the West. They were more likely than less frequent users to have moved in the past year and are less likely to live in a 2-parent family. Frequent marijuana users were more likely than less frequent users to report delinquent behaviors such as running away from home, stealing, and cutting classes or skipping school. They were also more likely than less frequent users to report aggressive behaviors such as destroying things that belong to others and physically attacking people. Monthly or more often users were more likely than less frequent users to have driven under the influence of alcohol or drugs or sold illegal drugs in the past year. From a psychological view, youths who used marijuana in the past year reported many behaviors symptomatic of anxiety and depression. Users were 2 to 4 times more likely than nonusers to report they think about killing themselves or that they deliberately try to hurt or kill themselves. They weremore likely than nonusers to say they were unhappy, sad or depressed and that they feel "no one loves me". The users were more likely than nonusers to report that "others are out to get me" and "I am suspicious".
"Regardless of whether the problem behaviors preceded marijuana use or marijuana use preceded the behaviors (which we are not able to ascertain from the NHSDA), it is apparent from these data that the marijuana users are exhibiting many signs of anxiety and depression and exhibiting delinquent and aggressive behaviors far in excess of the nonusers. Further, there appears to be a high correlation between the presence of many of these reported behaviors and the frequency of marijuana use."
The Friedman research which ONDCP references is even more limited and less generally applicable to society as a whole. As noted in the article itself,
"Thus, the sample for this study was not strictly representative of an
inner-city, low SES, African-American community. It is, rather,
roughly representative of that part of such a community that had been
involved in the use of drugs." (pp. 56-7)
The article notes further that:
"Drug use by persons with antisocial personality could reasonably be
expected to increase any tendency to act violently. On the other hand,
such drug use has also been postulated to be an attempt to assuage
violent tendencies through self-medication. To the extent that this
might occur, it indicates that having violent tendencies can predispose
to substance use/abuse, which is the converse of the relationship proposed
for study in this paper. Accordingly, it is a potential confounding
factor that should to be controlled for in the analyses performed for
our study." (p. 52)
And finally, the authors note specifically that:
"A disclaimer, or a note of caution, is indicated against over-generalizing
the findings of a linkage between marijuana use with drug selling
in the inner-city and with involvement in serious types of criminal and
violent behavior. These significant marijuana-violence linkages that
have been found for this study sample may not apply to a representative
sample of the general population. The findings presented here
may be specific for the sample of this study: an inner-city, relatively
low SES, African/American sample. As postulated in the introductory
section of this paper, marijuana use during adolescence is fairly widespread
in this study sample, especially within specific peer groups.
The regular users of marijuana maintain contact with the sellers of
drugs, and thus become more familiar with the criminal life style,
which may lead to a tendency to engage in drug selling themselves,
and thus to a greater likelihood of committing violent illegal offenses.
The drug sellers from whom they originally obtained the cocaine and
other drugs during their adolescence, most likely were adolescent
peers who grew up in similar circumstances to their own. The majority
in the sample need the money. Some are helping their families financially
with some of the money they earn from selling drugs. Thus, a
peer bonding and friendship develops between the buyer/user and his
drug provider. The buyer/user becomes a new seller, and eventually
finds himself in circumstances in which engaging in violent illegal
behavior is routine and is considered to be acceptable.
"These findings on the degree of relationship of substance use to
violent behavior may be somewhat inflated since we do not have
available for control purposes, data on all the possible factors, in
addition to substance use, that may be involved in violent behavior,
(i.e., all of the relevant characteristics, behavior and life circumstances
of the subjects, that predispose to violent behavior). The fact that there
were available as many as 51 such relevant characteristics for use as
control variables in the analyses, may be considered to be a relative
strength of the study. On the other hand, it is a weakness, or a limitation
of this study, that data on some of the factors or influences that are
known to predispose to violent behavior were not available for the
analyses. An outstanding example of such an influence is the amount
of time spent during childhood and adolescence in watching TV programs
and films that present violent behavior in an interesting and
exciting manner. Such entertainment programs sometimes present, as
heroic figures, characters who use drugs and engage in violence. In
any case, the lack of more complete control data should not be a
significantly greater problem for determining the effect of the use of
marijuana on violent behavior, than this lack would be for the effect of
the use of any other type of drug. Thus, it would not explain why the
degree of marijuana use was found to have a greater degree of relationship
to certain types of violent behavior, when compared to the
degree of cocaine/crack use." (pp. 68-9)
Other research examining the question of marijuana and violence was published by the Journal of Contemporary Criminal Justice in 2004. In
"Teasing Apart the Developmental Associations Between Alcohol and Marijuana Use and Violence," Wei et al. examined data from the Pittsburgh Youth Study on 503 young males. They concluded:
"With regard to the associations between early frequent marijuana use and
later violence, our conclusions are similar to those of White et al. (1999), in
that what we are seeing is a selection effect. In other words, marijuana use is
more atypical during early adolescence and becomes more normative with
age, and the subset of males who begin marijuana use at younger ages are at
elevated risk for several serious outcomes, including poly drug use, violence,
and property offending. It is likely that this subgroup of males is inherently
more deviant, engaging in multiple problem behaviors at earlier ages, choosing
deviant peers, and being more likely to manifest their individual propensity
for aggression and antisocial behavior later on. Our findings reinforce
the benefits of primary prevention efforts that address multiple risk factors
early on, as well as early intervention with high risk or aggressive males.
"Because the proportion of violent individuals who used marijuana frequently
was larger than the proportion of frequent marijuana users engaging
in violence, and because the prediction of violence from earlier frequent marijuana
use was mediated by common risk factors, our results do not indicate
that early frequent marijuana use causes later violence. Rather, we conclude
that frequent marijuana use and violence co-occur because they share common
risk factors (e.g., race/ethnicity, hard drug use). It is important to keep in
mind that marijuana has been used for centuries and is the most widely used
illicit drug today and that the majority of marijuana users do not engage in
violence (Boles & Miotto, 2003). Our findings indicate that intervention
with young violent offenders to prevent or treat substance use problems
may be more practical than targeting marijuana users for violence prevention."
Back to top
