For instance, USADA explains:
Education is an extremely important pillar of an effective anti-doping program and is the first line of defense in protecting the rights of clean athletes. USADA provides extensive anti-doping education to thousands of athletes each year, helping athletes and support personnel understand their rights and responsibilities in regards to the drug testing process.WADA says something similar:
WADA believes that a long-term solution to preventing doping is through effective values-based education programs that can foster anti-doping behaviors and create a strong anti-doping culture.We'd all like to believe that education leads to changed behavior. I've come across this issue frequently in areas of science studies, where the assumption that more education leads to changed behavior has been pretty well debunked for a long time. There is actually an enormous literature on this topic. Education doesn't necessarily change behaviors in predictable or desired directions.
Thus, given this broad understanding, it is a bit surprising that the anti-doping community relies so heavily on a claim that education is an effective anti-doping strategy. It is almost an article of faith in some circles.
But the evidence that is available doesn't support such claims or faith. In fact, 2007 WADA published the result of a sponsored study which reviewed the literature on the effectiveness of education in anti-doping (drawing upon work related to bullying, illegal drug use, etc.). That study concluded (PDF):
Educating athletes about the effects of drugs, and specifically the adverse effects, may be a means of deterring drug use 53. However, this contention is not supported in the present review. Studies that employed interventions designed solely to provide subjects with information regarding drug use (anabolic steroids) in either a balanced format 40 or with an emphasis on adverse consequences 39, were not effective to altering subjects’ attitudes or intentions. Whilst providing information on drug-related issues and improving knowledge is necessary, an effective programme must also address the myriad of other variables that impact upon the decision to use performance enhancing drugs, such as alternatives to drug use, peer and media resistance training and decision-making skills.More broadly, there simply isn't much evidence of effectiveness:
There are several limitations to the existing research. Firstly, whilst a number of studies successfully modified attitudes or intentions related to drug use, they were less effective in reducing drug use behaviours. Too few studies have conducted long-term follow-up to determine whether intervention effects persist over time and whether they translate into reduced drug usage. Secondly, all studies to-date have employed self-report questionnaires to gather data, without corroboration by other methods. Because the issues addressed in this research relate to illegal or taboo practices, or indeed social desirability favouring exaggeration, the validity of findings may be compromised due to under and over-reporting. Research in this field has primarily targeted college or university athletes, with programmes co-ordinated primarily within the athletic department. College and university athletes differ from other athletes in that they are subject not only to the pressures of sporting performance but also expectations regarding their academic achievement, which may create a unique set of circumstances surrounding the decision to engage in drug use. Accordingly, it is not appropriate to generalise the methods and findings of these studies to other settings and populations. Research has mostly been confined to prevention of anabolic steroid use, with less attention paid to other performance enhancing and recreational drugs, such as growth hormones, amphetamines or cocaine.The more fundamental problem of course is that there is very little evidence about the prevalence of doping in sport, much less on the sorts of approaches that are more likely to influence that prevalence. Anti-doping efforts have a long way to go to become more evidence based.