while i understand you reasoning, for the change doesnt your suggestion also use the word accurate in both the question and the answer.
I do get the instinct, but for sure don’t need ‘accurate’ twice. Also ‘the degree to which’ twice. It’s just a mess.
I do appreciate the basic attempt though, and agree that emphasizing that validity is that a thing is measuring what it purports to measure, can be helpful here.
Maybe the wording needs a bit of a tweak?
yea i think it should be c1::the measurement reflects what it is intended to measure (i,e, …
that way the word accuracy doesnt give it away
I went ahead and cleaned it up a little bit. It’s maybe not the most artful phrasing, but it’s better than the previous card and better than the odd suggestion
No problem if you guys disagree. The way I was always taught validity/accuracy vs precision is:
Validity/Accuracy refers to how close a measurement is to the true or accepted value. Precision refers to how close measurements are to each other.
{{c1::Validity}}, also known as {{c1::accuracy}}, refers to {{c2::how close a measurement is to the true or accepted value}}. Seems like a super simple option.
Admittedly my suggestion does change the content of the cloze/card structure. However I’m not sure how disruptive of a change this is given its pretty much just a more concise wording of the original.
Either way I like the current suggestion a lot more than the wording previously.
Neither is incorrect, really. The discrepancy I’ll raise is that not every form of validity is arriving at a ‘value’ per se.
In other words, in social science you might be correlating moods, or affects, or any number of other things that don’t have a precise value. This is often the case in survey design, social science (especially psychology), etc.
In the ‘hard’ sciences you are, of course, more often arriving at a specific value-state. But that’s not an inherent requirement to validity as a whole.
An easy example here would be that you might have a very valid measurement for assessing dynamic balance in older populations. You’re 100% not looking for one specific value, because there isn’t one. Or, maybe more correctly, even any empirical data you’re gathering might be arbitrary in the broad sense except that it points at a valid estimation of fall risk, let’s say.
In this case, the Romberg Test is quite valid but it’s not measuring against a specific value.
That argument makes sense to me. I like your wording more.