We then use the highly significant p-value as proof that we should consider a quasi-Poisson model instead to account for the difference in the true variance from the variance assumed under the model. \\textbf{Original comment:} \\textit{iii) The authors should be more descriptive about the tests realized.
← all excerpts
The impact of differences in text segmentation on the automated quantitative evaluation of song-lyrics.
1
—
—