A group of 225 men with benign prostatic hyperplasia were randomly assigned to take saw palmetto extract or a placebo in a double-blind trial. One year into the experiment 45 of the 112 men in the saw palmetto group (40%) thought they were taking saw palmetto, compared to 52 of the 113 men in the placebo group (46%). Is this difference consistent with chance variation? Conduct Fisher's exact test using a non directional alternative.