How can we compare the F2 values of the vowels produced by ten speakers? There are only two vowels to be investigated. Therefore, 20 vowels are produced in total. I believe paired t-test is the way to go but I am not sure. That's why I need your help. Thanks in advance.
5
-
Is there only those parameters you want to investigate? So you don't take into account gender.– amegnunsenDec 4, 2019 at 11:03
-
Or surrounding consonantal / vocalic context? Or age, geography?– user6726Dec 4, 2019 at 15:33
-
All the vowels were produced by males, aged 20-26. They are all native speakers and use the standardized language.– Geralt of RiviaDec 4, 2019 at 16:58
-
I am fairly sure that paired t-test is not correct. This is really not a linguistics question, it's a statistical question. If you have 10 people saying "sit" vs. "seat" once, you don't have enough data to be drawing an inference.– user6726Dec 5, 2019 at 1:06
-
This was a hypothetical question and all I wanted to know was which statistical analysis one should use when comparing F2 values of, say, enough data. If you think paired t-test is wrong choice, could you help me understand which one should be used?– Geralt of RiviaDec 5, 2019 at 6:06
Add a comment
|