A medical company claims it has a test that is 95% effective in determining whether an athlete is using a steroid or not. The company also says that the test gives a false positive in 3% of cases. Suppose this test is given to a group of athletes in which 10% of the athletes are using steroids. What is the probability that a randomly chosen athlete actually uses steroids, given that the athlete's test is positive?