I agree completely the study is useless and should be withdrawn. But the author did not prove or even reasonably establish at all that there were 3 and not 8 participants. Not finding 5 out of 8 anonymous participants in a half century old study is to be expected. Even decloaking 3 was remarkable. Look at how hard it is to deanonymize data even now with the computational power, combined with endemic surveillance and monitoring of the public. It's totally unreasonable to conclude the other 5 didn't exist.
Even without the anecdotal information that things were not presented as they were told, the study is still, and always has been, useless. If someone presents at an asylum begging to be admitted, and describing strange symptoms that they say are causing them distress, it is reasonable to admit them for observation.