I just wanted to say, I did a double take when I opened the supplement to Lunden data file because I thought it was my name on the document LOL (obviously it wasn’t (and she spells her name wrong)).
I agree with most of the sections you removed from your analysis, except for removing vowel reduction. I think it’s important to know if the language values vowel quality because it gives you insight into what kind of language it is (stress-timed, syllable-timed, etc.) which can lead to finding the other motivating factors for stress in that language. Vowel reduction tells you where the stress can’t be, which helps you find where it can. It doesn’t seem important alone but in conjunction with the other correlates it matters. Stress correlates are kind of like puzzle pieces—they don’t really mean anything until you put them together—when you put them together you see the bigger picture.
I liked how thorough you were with explaining what you removed from your analysis and why. I would have liked to see more of a summary of the original paper to be able to compare the differences between the analyses.
Why did you decide to use the binomial distribution?
It’s a little hard to see the different colors of the points in the plot. Perhaps changing stress from color to shape would make it easier to read.
I just wanted to say, I did a double take when I opened the supplement to Lunden data file because I thought it was my name on the document LOL (obviously it wasn’t (and she spells her name wrong)).
I agree with most of the sections you removed from your analysis, except for removing vowel reduction. I think it’s important to know if the language values vowel quality because it gives you insight into what kind of language it is (stress-timed, syllable-timed, etc.) which can lead to finding the other motivating factors for stress in that language. Vowel reduction tells you where the stress can’t be, which helps you find where it can. It doesn’t seem important alone but in conjunction with the other correlates it matters. Stress correlates are kind of like puzzle pieces—they don’t really mean anything until you put them together—when you put them together you see the bigger picture.
I liked how thorough you were with explaining what you removed from your analysis and why. I would have liked to see more of a summary of the original paper to be able to compare the differences between the analyses.
Why did you decide to use the binomial distribution?
It’s a little hard to see the different colors of the points in the plot. Perhaps changing stress from color to shape would make it easier to read.
Either way, I think you did a good job! :)