Is it possible to express
has a higher variance
in the sentence
Dataset A has a higher variance than dataset B
as an adjective? Would
Dataset A is more variable than dataset B
be the right choice?
-
Is more variative.– Rusty CoreCommented Feb 5, 2019 at 20:17
-
A quick NGrams search shows that 'more variable' >> 'higher variance' > 'more variant' (and checking the examples, in formal/academic writing).– MitchCommented Feb 5, 2019 at 20:26
Add a comment
|
3 Answers
The common way of saying this is to use the word variant:
1 : manifesting variety, deviation, or disagreement
2 : varying usually slightly from the standard form
// variant readings
// variant spellings
So, in your case, you would say:
Dataset A is more variant than dataset B.
Generally, a comparison needs to be made to something else. So, in this case, dataset A is more variant than dataset B when it comes to some statistical norm.
For a more explicit statement, you could say something like this:
In comparison to the norm, dataset A is more variant than dataset B.
If you're talking about the elements within a dataset, then you are comparing them to each other. In that case, you could use variance if you rephrased your sentence slightly:
The elements of dataset A show more variance than those of dataset B.
answered Feb 5, 2019 at 16:35
-
1To me, "more variant" does not mean "having a greater variance" in the statistical sense. Maybe it's just me, but that part of the answer seems wrong. "...show more variance" is good, though. Commented Feb 5, 2019 at 20:21
-
This sounds like it should be write but NGrams shows that it is hardly ever used in actual practice.– MitchCommented Feb 5, 2019 at 20:26
The dataset cannot be more variable from my opinion. It just is. However variance is a property of the dataset, and it sounds OK in the very short example provided.
Edit to improve after comment: From my understanding, the dataset is a set of (let's say) numbers, acquired after some measurements (or any other activity). Therefore, the dataset cannot change (unless new measurements are done, leading to a new dataset). So the dataset cannot be variable. Let alone being more variable.
-
1Thanks - the point is: From a statsitical point of view, "Variance" is an exactly quantifiable property. Therefore, a dataset can have a higher variance than another. I.e. if "variable" would be the correct choice of adjective, "more variable" should also make sense. Commented Feb 5, 2019 at 11:15
-
@user1934212: Thanks for pointing out incomplete answer :)– virolinoCommented Feb 5, 2019 at 11:23
-
Thanks for pointing out an incomplete question :-D See update. Commented Feb 5, 2019 at 11:39
-
You can force a "simplification of language" and ignore the comments from some guys like me. Only then
variablecan be used to describe something that does not change. :D But I may be wrong, you did not clarify if the dataset is fixed or volatile. The point is: if you need to usevariance, you are not in the business to create romantic poems. Therefore, you should remain to the strict definition of words, without simplifications. My 2 cents ;)– virolinoCommented Feb 5, 2019 at 11:53 -
On the other hand, whatever you use is your responsibility. So yes, go ahead and use anything :)– virolinoCommented Feb 5, 2019 at 11:55
Dataset A is more variable than dataset B.
Yes, this is perfectly good English. The only issue is whether variable is precise enough.
Variance is a very specific mathematical concept. If you want to say that the variance of A is greater than the variance of B, mathematically, I don't think there's an adjective that will do the job. Just use the noun:
Dataset A has a higher variance than dataset B.
...or...
You can use high-variance as an adjective, as in “a high-variance dataset”. After seeing that a few times, readers will have no problem with sentences like:
Dataset A is higher variance than dataset B.
With higher-variance datasets, it is important to check that the results are significant.
This is a bit jargony, though. I think “has a higher variance” is better.
