« Contact the author | Main | Media misreporting physics »

Definitions

10 April, 2009

I have seen many science teachers argue furiously over accuracy, validity, reliability and precision. Does anyone have some definitions that are universally accepted? I thought this might be a good topic for discussion.

Cheers,
Jeff

Posted by: Jeff Stanger | April 9, 2009 06:57 PM

Comments

Hey Jeff,

When I am explaining this to students (and arguing in the staffroom), I use extreme arguments to highlight the differences.

For example, I say that to measure the temperature of a room, it would not be valid. It may be reliable, but not valid.

Here are my rough definitions for discussion:

Accuracy - how close a result comes to an accepted value (e.g. measuring g).
Validity - relates to the method of an experiment; does it actually test what it claims/aims to?
Reliability - essentially the same as repeatability. Do you get the same results every time? An experiment can be inaccurate but reliable due to a systematic error (e.g. a wrongly calibrated piece of equipment).
Precision - consistent results (e.g. in archery or darts, hitting the same mark every time). It's very hard to distinguish this from reliability - precision refers to a certain degree of consistency i.e. having an answer to 6 decimal places is more precise than 2.

The other comment is that I think that accuracy and precision are best paired/spoken about together, and validity/reliability as another topic.

Other thoughts?

Andrew

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Enter the code shown below before pressing post