Wednesday, July 3, 2013

Are TSA screeners really slower than amateurs?

By Prof. Alan (Avi) Kirschenbaum*

A recent report published on NBC under the title, “TSA screeners slower than amateurs, but more accurate” raises some serious questions. 

The paper, which appeared in an obscure journal that focuses mainly on physical aspects of vision cognition, is certainly an eye catcher, until you look more carefully at the researchers’ results and try to figure out if the innuendos about the accuracy and speed of TSA trained screeners inability to match “amateurs” are what they actually found. 

Several issues are involved here. The first is that the “amateurs” were in fact college students, the proverbial “cannon fodder” of academic researchers in their quest to conduct experiments. The second is the relevance of these results for airport security.

The experiment revolves around identification of the letter “T” (sometimes) placed on a screen with various other “Ts” but not exactly the same or in the correct position. According to the results, accuracy and time in identifying the “T” differed. 

The differences, as far as I could determine, were not statistically significant, meaning they could have all been achieved by sheer chance. Moreover, this experiment was done without the usual external noises and pressures that are part of the daily life of security screeners.

Even more dubious is the substance of the comparison: screeners are trained to identify objects while college students focus most of their efforts on dealing with and comprehending words. No great wonder that students saw the letter more quickly, but when it came to accuracy the screeners beat them out. The rest is all interpretation, with the article’s authors bringing us the earth-shattering news that consistency is the key to better performance.

So what does this have to do with airport security? Does this mean that hiring amateurs is better than investing in training screeners? As airport managers want to increase throughput of passengers, no doubt that 1–2 second difference in spotting that “T” may make a difference. But what about the consequences?

It all comes down to accuracy versus speed. The screeners were more accurate – meaning fewer false alarms and greater chances of spotting the prohibited items. The college students were quicker but made more mistakes. 

As a security manager, which one would you opt for? Part of that answer was revealed in the BEMOSA Project where up to 40 percent of the security employees stated they bent, broke, ignored and even went against the rules – despite knowing full well what was on the screen and who was in front of them.

This headline enticer requires a lot of caution when one realizes that we are dealing with a sanitized experiment in a controlled environment where the results may have very little to do with the real social world outside the experiment’s booth. 

This is particularly true in the case of airports which are complex social organizations with rich and vibrant social networks that the BEMOSA project has amply revealed. Therefore, it might be wise to keep both eyes open the next time you see this kind of article. 

---

 *Prof Alan (Avi) Kirschenbaum, founder and CEO of Kirschenbaum Consulting.