The Bias of Bias research

[Boyd Baumgartner]

https://www.bloomberg.com/news/articles ... r-it-looks

Make no mistake, fingerprint identification can be cast in the light of behavioral economics as it is the assessment of value and action based on that value. This jumped out at me given the Brendan Max video and his critique of the performance studies and appeal to the Human Factors group.

From the article:

They either rely on small sample sizes, misinterpret individual errors for systematic biases or underestimate how people absorb information based on how a fact or question is framed.

Small sample sizes *cough* Dror

And this is the rhetorical device Mr. Max uses, it's the misinterpretation of individual errors for systematic biases. It was interesting how he was able to 'see beyond the data' to put out the real potential error rates 'that they don't want you to know about' (Sound on for maximum benefit) (And weren't even based on any actual data...you know...the standard he claimed fingerprint identification failed to meet). Although, to be fair, that's something that Ralph Haber trotted out way back in 2009 and included in his book Challenges to Fingerprints.

The secondary issue of how people absorb information based on the study design (standard for conclusions?) and how information is absorbed based on how a fact is framed plays right into our discussion of the proficiency testing as well as proposed OSAC scales and why specifically they need to be validated before they're implemented, but those are topics for a different thread.

[Dr. Borracho]

Make no mistake, fingerprint identification can be cast in the light of behavioral economics as it is the assessment of value and action based on that value.

Isn’t it funny how a concept linking such seemingly unrelated disciplines as economics and fingerprints can crystalize in the minds of diverse individuals simultaneously?

In the December issue of The Atlantic, an article titled “The Road From Serfdom” by Danielle Allen explains the shift in government decision making that has exacerbated the gap between rich and poor. Ms Allen postulates that government policy has shifted from caring about individuals to considering only the effect a policy has on an overall population, even though the majority of citizens may be reduced to the equivalent of serfdom in the Middle Ages.

One paragraph in particular jolted me to the same realization that Boyd makes in the above quoted passage. Ms Allen wrote,

In the late 20th century, economics established itself firmly as the queen of the policy-making sciences. Up until then, before the emergence of digital computing power and the spread of numbers-based social science, people who were trained as lawyers, not as economists, had dominated policy making. The shift is documented in recent research by the sociologist Elizabeth Popp Berman. The difference in outlook between economists and lawyers is immense. Whereas economists seek out rules that are in theory universal – mathematical principles that apply everywhere, and are blind to context – legal thinking is fundamentally about the institutions of specific societies and about how institutions actually work in specific situations. This is not to say that we can always count on lawyers to see real people or that lawyers went away. The point is that a different way of thinking – emerging first in economics – has ascended across a wide range of professions.

In his book The Economists’ Hour, published last September, Binyamin Appelbaum goes into great detail explaining how the post-World War II reliance on economists to direct government policy has resulted in drastic shifts in not only the economy, but in all phases of American life. Unless you are some kind of economics wonk, you will find the shift in governmental philosophy and policy outlined in this book alarming. It explains a broad range of trends in American life in the last half of the 20th Century, not all of them good, and some of them very bad for much of the population.

Ms Allen’s ideas are so insightful that the last sentence of the above paragraph merits repeating:

The point is that a different way of thinking – emerging first in economics – has ascended across a wide range of professions.

Alas, we are among the victims of this new way of thinking.

[Boyd Baumgartner]

Isn’t it funny how a concept linking such seemingly unrelated disciplines as economics and fingerprints can crystalize in the minds of diverse individuals simultaneously?

For me behavioral economics addresses the fact that there are biological, psychological and socio-cultural influences to decision making that you can't necessarily isolate or model via statistical means alone. While it's fun to say that it's subjective as a perjorative, it's equally subjective to give the illusion of objectivity by slapping some statistical models on it that are just reflections of the subjective decisions.

You need to ask different questions from the different angles that these schools of thought provide. Taking a one dimensional approach to a complex task like fingerprint examination is just plain silly. Staying siloed in forensic science research causes tunnel vision.

Is it important to know? Yes. But is it also important to know how to talk beyond it to the tasks that we're also engaged in for which there's no corresponding JFI article? Absolutely.


EconTalk is one of my favorites. (filter by date/month is on the lower right hand)

Both Binyamin Appelbaum and Gerd Gigerenzer were recent guests.

[ER]

Small sample sizes *cough* Dror

I still insist that Dror's biggest sin in all his research was this article title:

"Contextual information renders experts vulnerable to making erroneous identifications"

Yes, there was a small sample size, but even worse... There were ZERO erroneous identifications.