Hey,
can someone please help me interpret pic related. I looked up some tutorials how to understand box-and-whisker-plots but I'm just not getting behind it.
I'd really appreciate it if someone with an understanding of linguistics could help me out here.
I can also provide provide further information if that would be helpful.
It is from Ryuichi Hotta's paper on "s-Pluralisation in Early Middle English and word-frequency".
>>1161816
I'm going to attempt to explain box plots and be of no particular help since I'm just working off my own notes and might be wrong.
The center line is the median value and divides 50% of the values from the other 50% of the values.
The box encompasses 25% of the values on either side of the line and is the interquartile area.
The pinched area is a 95% confidence interval for the median.
Whiskers vary I guess. Generally they're 1.5* the interquartile area and end at the last value in that range.
Circles represent outlying values.
Cross probably represents the average value.
The chart would probably be easier to read if it included a strip chart as well to represent every value and C12b doesn't particularly look like the size of the chart was chosen properly for the range of values. I have no idea what's going on at the bottom of C12b.
Hopefully that helps marginally better than other tutorials since I can't find the paper online so I don't understand the context.
>>1161992
Thanks very much for sharing your notes.
Maybe I'm trying to interpret it based on that and you can tell me if this makes sense according to the mechanics of a box-plot.
So the paper is about how plural markers in Early Middle English developed from things like -um, -es, -as and such to the today known -s. Hotta wanted to research which kind of nouns were effected from that change first, high-frequency or low-frequency nouns.
Earlier in the paper he stated pic related.
So when now looking at the box-plots I assume that the amount of data above the median are the high-frequency nouns and the ones below are the low-frequency nouns. Makes sense, right?
So when I now look at how the data changes over the periods I can see, that at first high-frequency nouns in C12b had a way higher amount of innovative s-types and throughout to C14a both high and low-frequencies became more alike and decreased (which makes sense since mostly all words have by then adapted to the s-plural).
But somehow this doesn't really make sense when comparing it with what he stated in pic related, when he said high-frequency nouns were much slower adapting to s-plural.
So I must be misinterpreting the diagram, right?
May anyone evaluate my interpretation and tell me where I went wrong?
I'm bumping once more today, would really appreaciate it if somebody gave me a hint.
