I would like to see the raw data, i.e. the list of (supposed) HPE's (divided by industry) AND know how (the heck) the recruiters managed to get their hands on i

I would like to see the raw data, i.e. the list of (supposed) HPE's (divided by industry) AND know how (the heck) the recruiters managed to get their hands on it. I mean, provided that the survey is actually a valid representation of HPE's, 68% is too high to be "casual", it means that the recruiters have some way to identify those HPE's (separating the wheat from the chaff), but if they (the recruiters) can do that, then the percentage should be very near 100%. It is more likely that a percentage very near to 100% of employees in a given industry (let's say for the sake of the example software or IT) are contacted monthly (some weekly) and that a certain amount of employees in another industry (let's say still for the sake of the example manufacturing or architecture) are never or seldom contacted. Or - seen in another way - how often (still divided by industry) are APE's (Average Performance Employee) and LPE's (Low Performance Employee) contacted by recruiters (if contacted at all)?