Technical term related to measuring targets
December 31, 2010 5:45 AM   Subscribe

What is the technical term for people working to meet a target, not the thing the target is trying to measure?

When a target is brought in to measure something, there is a "law" that says within a certain amount of time people will - consciously or subconsciously - modify their behaviour to meet the target not that which the target was intended to measure.

I know there is a technical term for it (I've seen it before) but my google-fu is failing - anyone out there got any ideas what it is?
posted by garius to Writing & Language (16 answers total) 4 users marked this as a favorite
 
I've only ever heard it described as a "target-driven culture" -- and usually in reference to education or health. Like when a school shifts focus from a broad curriculum to just what's on an exam, because of league tables.
posted by londonmark at 6:02 AM on December 31, 2010


In education there's "teaching to the test".
posted by XMLicious at 6:10 AM on December 31, 2010 [4 favorites]


Goodhart's Law?
posted by TheophileEscargot at 6:13 AM on December 31, 2010


Gaming the system.
posted by drdanger at 6:43 AM on December 31, 2010


From the land of dog training, I can point to a post that refers to some sources that might have the answer, although the books and talks mentioned are all about the problems with setting targets and the teaching to the test phenom.

I don't recall Goodhart's law coming up in Pink's book.
posted by Lesser Shrew at 6:47 AM on December 31, 2010


Sometimes this is called measurement dysfunction. The Joel on Software website had two articles that went into this (spurred by Joel reading Robert Austin's book, Measuring and Managing Performance in Organizations): Measurement, and The Econ 101 Management Method. There was a more recent article on the same website, but I can't find it right now.
posted by Houstonian at 6:53 AM on December 31, 2010 [1 favorite]


Shifted burden. When people come to believe that their job at the widget factory is to hit targets rather than to produce widgets.
posted by txmon at 6:58 AM on December 31, 2010


Rubric.
posted by brent at 7:04 AM on December 31, 2010


I thought I had heard Observer Effect in my business classes
posted by ansate at 7:14 AM on December 31, 2010


From a wikipedia article on Demand Characteristics:

In research, and particularly psychology, demand characteristics refers to an experimental artifact where participants form an interpretation of the experiment's purpose and unconsciously change their behavior accordingly.
posted by ThatCanadianGirl at 7:26 AM on December 31, 2010 [1 favorite]


In the organizations literature, I've learned this rule as: "the folly of rewarding A, while hoping for B" (1975 paper about the effect).
posted by eisenkr at 7:28 AM on December 31, 2010


It might be of value to look up static and dynamic efficiency which can have implications as regards meeting societal targets in the short and long term.
posted by biffa at 7:29 AM on December 31, 2010


Sounds similar to "the map is not the terrain".
posted by cosmicbandito at 7:44 AM on December 31, 2010


What you measure is what you get.
posted by ewiar at 10:06 AM on December 31, 2010


I think you're looking for the Observer Effect. I'm assuming you're referring to a situation in which a researcher observes people/animals to form some generalised behavioural hypotheses, but the presence of the researcher and the situation of being observed result in the people/animals changing their behaviour, thus skewing the results.
posted by Xianny at 11:07 AM on December 31, 2010


I don't think this is the same as the "observer effect"; it can occur even without any observer (although in the obvious business examples there is an actual or implied observer).

Other terms in the same neighborhood:

A "perverse incentive", where you try to encourage some result by adding an incentive but there's an easier way to achieve the incentive that you don't like. For example, maybe you want to incentivize doctors to cure patients as quickly as possible so you give them a bonus based on what fraction of their long-term patients are healthy; perversely this encourages them to immediately kill any patient who might end up with a lingering illness.

The "law of unintended consequences" is the more general notion that trying to do something (perhaps by measuring an effect) may have a different outcome than you wanted.

"Working to rule" (or "malicious compliance") refers to intentionally working towards the stated goals instead of the commonly-understood goals, say as a form of protest.
posted by hattifattener at 12:53 PM on December 31, 2010


« Older Stripping non-alphanumerics in a UTF8 friendly way   |   Am I being irrational? Newer »
This thread is closed to new comments.