First off, for the matter in question, you will never get honest answers as to the proportion of escorts who have been subjected to some form of sexual abuse. In fact, the definition is a bit in doubt because some feminist organizations have defined ALL sexual activity with an escort to constitute violence against women.
So I acknowledge up-front that this information will never be known. Therefore, no statistics are valid. EVEN IF you had to pass a test in order to be an escort that you HAD to have been sexually abused in order to be qualified; a certain proportion of women would lie in order to meet the qualification. That's life.
HOWEVER -- while statistics cannot tell me anything about a given individual; they can tell me enough to help guide wise decision making.
Take registered sex offenders generally. WHY are they registered? Because the rate of recidivism is sky high.
Any given registered sex offender MIGHT have been wrongly convicted; and may not re-offend because he never offended in the first place.
BUT if, as a group, a woman excludes registered sex offenders from her dating pool she will decrease the likelihood that she will be harmed.
So even though we can't tell anything about a given individual from a group statistic; we CAN find enough information to help guide decision-making in improving our odds of successful outcomes.
As yet another example -- look at smoking. I believe it is 1/3rd of people who smoke until the day they die develop a serious smoking-related illness. 2/3rds do not. So, clearly, any individual smoker may not experience harm. However, you never know if you are going to be 1 in 3, or 2 in 3 -- so avoidance of smoking can be a wise choice.
The fact you can't tell something about an individual doesn't negate the value of statistics in decision making because a great deal of life is not about absolutes -- it is about odds.
Same reason that we use condoms. It decreases the odds of disease transmission.