The unit bias

Nominal variables matter, even when we are deciding how much to eat on our plates:

To test [the unit bias], the researchers left a bowl of M&M sweets in the
hallway of an apartment building with a sign that read “Eat Your Fill:
please use the spoon to serve yourself”. Some days they left a
tablespoon-sized scoop, other days they left a quartercup scoop that
was four times as big. Passers-by could obviously help themselves to as
little or as much as they wanted regardless of which spoon was
provided, but on average, 1.67 times more M&M’s were taken on the
days the big scoop was left compared with the tablespoon-sized scoop.

another experiment, the researchers found that, measured by weight,
significantly more pretzels were taken by passers-by when a
complimentary bowl of 60 whole pretzels was left in an apartment
building, compared with when a bowl of 120 half-pretzels was left. And
it was a similar story when either a bowl of 80 small Tootsie rolls (an
American snack bar) or a bowl of 20 large Tootsie rolls was left in an
office building.

In other words, throughout the study, people
took more food when the unit on offer was larger. “Consumption norms
promote both the tendency to complete eating a unit and the idea that a
single unit is the proper portion”, the researchers said.

Here is further information.  There is a lesson for macroeconomics in here, somewhere.


Were people supposed to eat 1/4 of a large Tootsie Roll and leave the other 3/4 of the Tootsie Roll back in the bowl?

Doesn't this study show more about personal etiquette than personal consumption? In essence, the choice one makes when they are confronted with a set of free items which is being offered to the public. People will take as much as they desire without being (or appearing) selfish and taking more than their 'fair share'.

The more interesting statistic to track might be what percentage of the entire bowl of pretzels or Tootsie Rolls was taken by each free-sampler. Does the portion size decrease as the total amount in the bowl decreases? Or does it stay relatively constant?

Yes. Stuff put out in public is one thing....people don't want to seem greedy. But I have a feeling, if anything, that broken pretzels and mini tootsie rolls disappear faster than their bigger compadres if they are consumed in an home environment. Too easy to say, just another tiny one...

I think this, or something similar, is well-known in the grocery business. When goods are marked at say, 2 for $3.00, customers have a strong tendency to buy pairs, even though if price of one unit is $1.50.

Isn't there also an effect from assuming that other people have some idea what they are doing? If everyone else is eating a portion of size X, then that is some (strong) evidence that size X is more-or-less appropriate.

A small fixed cost per scoop regardless of scoop size generates the result without recourse to bias, no?

Outside the realm of food:
A newborn (human) child is called "large" if she's over eight pounds.
Or she's called "large" if she's over 4000 grams.

The thing is, one is 10% larger than the other. This isn't just nomenclature; it affects the observational and treatment protocol.


I just wanted to say WOW!!! goose bumps and e-motions, the design of your web page really got me!!! Check my sites ;)

Comments for this post are closed