Reliability and Validity

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Reliability and Validity

JanEmilLarsen
In 'Back-to-the-moment", in the page "It is about reliability" utterings  containing "I  think" ctr. "I am absolutely certain" of the dating 1489 is given as examples of different reliability.

However, is this rather a statement of (level of) validity, while a statement of reliability  could be eg. dating 1489 (high reliability),  or dating "sometime late in the 15. Century" (lover reliability)?

A and B agrees on the same value (same reliability?), but B is (more) certain that the value is valid (not same validity)....
Same accuracy <> Same validity ...

Reliability and Validity are standard terms in scientific/statistic assesments of measurements.

Which is in fact the concept used here?
Reply | Threaded
Open this post in threaded view
|

Re: Reliability and Validity

JanEmilLarsen
To partly answer my self:

It is probably a reliability meausure .
But this has nothing to do with that A and B agrees on the same value.
Even if they disagreed about the year, the degree of reliability would be the same.
Perhaps the example obscures this.
Reply | Threaded
Open this post in threaded view
|

Re: Reliability and Validity

roenbaeck
Administrator
To define our usage of the word reliability, it is a truth value, albeit extended from {0, 1} to the interval [-1, 1], where -1 means that the opposite of a statement is true, 0 that the statement is completely uncertain, and 1 that it is absolutely true. There may be more appropriate wordings that could have been used, but since this sort or "leaked" out, we decided to stick with the word reliability.