Confirmation Bias

by Don Friedman on July 18, 2011

Confirmation bias is a problem in everyday life and in most professions, so it’s no surprise that it can be a problem during building investigation. People looking at buildings can be biased to prefer modern structure to old (the “it’s old junk” syndrome) or to prefer old structure to new (the “they don’t build ’em like they used to” syndrome). They can prefer steel to concrete or wood to masonry, usually based on their previous experience and training. They can prefer engineers’ reports to architects’, or builders’ reports to both. All of this can be a problem and should be guarded against, but it is normal. This is how we think.

A more serious problem that we have noted occurs when people with training in modern structural systems and materials investigate old buildings with archaic and obsolete structure. Every reasonable person knows they have to research a system that they are not familiar with, but confirmation bias may prevent them from realizing that they are not familiar. To give two obvious examples, concrete-encased steel floor beams look very much like reinforced-concrete floor beams, and cinder concrete looks very much like stone concrete. If an investigator is unfamiliar with the decades of use of cinder concrete in New York, they might see board-forming and assume that the floor is a bar-reinforced stone concrete slab:

It might be. It might be wire-reinforced cinder concrete. It might be wire-reinforced gypsum. The only way to be sure is to check (either by probing or by looking for existing damage that allows probe-like views of the interior of the floor); the only way to know to check is to know that these possibilities exist. Otherwise, a floor slab will be assumed to be concrete and the board-forming marks will provide confirmation of that incorrect assumption.

Previous post:

Next post: