Inconsistent output messages and documentation gaps between correlation methods

[moylettsinead]

When testing different correlation methods and datasets, I noticed inconsistencies in output messages and lack of documentation about method differences.

Test Code:

Test with mtcars dataset using MIC

results_mtcars <- corrp::corrp(mtcars,
cor.nn = 'mic',
cor.nc = 'pps',
cor.cc = 'uncoef',
n.cores = 2,
verbose = FALSE
)

Test with iris dataset using Pearson

results_pearson <- corrp::corrp(iris,
cor.nn = 'pearson',
cor.nc = 'pps',
cor.cc = 'uncoef',
n.cores = 2
verbose = FALSE
)

The output size of the results varies according to the input dataset dimensions. For the mtcars dataset with 11 variables, the output contains 121 observations (11²), while the iris dataset with 5 variables produces 25 observations (5²). The message handling differs between methods: the MIC method provides complete messages in the 'msg' column, whereas the Pearson method shows truncated messages. There is also insufficient documentation explaining the output size calculations, the differences between correlation methods, and their appropriate use cases. The expected output structure is not clearly documented, which could cause confusion for users trying to interpret their results.

To improve the package, the message truncation in the Pearson method should be fixed to display complete information. The documentation should be expanded to explain the output structure and size calculations, along with a comprehensive comparison of different correlation methods. Including examples with various datasets and methods in the README would help users better understand the package's functionality and choose the appropriate method for their analysis.

[moylettsinead]

openjournals/joss-reviews#7319