DAHLIA LITHWICK argues in favor of the TIPS program. But the example she uses of a successful “snitch” strategy is the mandatory-reporting requirement for suspected child abuse. Lithwick says this has been a success, and that it has produced few false accusations. This surprises me: my wife, a forensic psychologist who deals with violent juveniles, holds a very different opinion.
Reader James Daniels shares these doubts, and writes:
It strikes me that comparing the repeated observation of children’s behavior and injuries by trained teachers and doctors isn’t similar at all to the untrained opinion developed by the plumber fixing your sink. Even worse, the mandatory reporting acts generate 3 million tips per year, of which about 2 million turn out to be false (according to [this report]). This is a 66% false positive rate by trained professionals. Given that terrorism has a much lower rate of occurance in the population than child abuse, can you imagine how quickly the FBI would begin to ignore the millions of tips generated by TIPS? Even with 3000 terrorists at large, 3 million annual calls would give a false positive rate of 99.9%. Further, if the FBI were to actually follow up on these helpful clues (on the order of thousands per day), the several million disgruntled voters smeared by the program would shut it down before the next election.
Given the low base rate of terrorism, any unselective test is going to generate disproportionate quantities of false positives. Worse, the real terrorists might start using TIPS against us, by filling it with chaff. Alternatively, they might hide their wares behind a false wall when the cable guy comes over. Basically, even if the problem were a lack of information at the top (rather than gross analytical incompetence), TIPS would prove to be a wasted effort.
I think this is right. As cases like the Al Qaeda webhacking incident illustrate, the system can’t deal with the information it gets now. Who’s going to analyze those tips, nearly all of which will be useless, to extract the good ones?