| Type: | Package |
| Title: | Strongest Neighbor Coherence |
| Version: | 0.1.0 |
| Maintainer: | Kevin E. Wells <kevin.e.wells@usm.edu> |
| Description: | Computes Strongest Neighbor Coherence (SNC), a structural diagnostic that replaces Cronbach's alpha using top-k correlation structure. For methodology, see Wells (2025) https://github.com/TheotherDrWells/snc. |
| License: | MIT + file LICENSE |
| Encoding: | UTF-8 |
| RoxygenNote: | 7.3.2 |
| Imports: | stats |
| Suggests: | knitr, rmarkdown |
| VignetteBuilder: | knitr |
| NeedsCompilation: | no |
| Packaged: | 2025-07-09 20:34:33 UTC; w10105397 |
| Author: | Kevin E. Wells [aut, cre] |
| Repository: | CRAN |
| Date/Publication: | 2025-07-14 17:30:02 UTC |
Print Method for SNC Objects
Description
Prints summary output for an object of class "snc".
Usage
## S3 method for class 'snc'
print(x, ...)
Arguments
x |
An object of class |
... |
Ignored. |
Value
No return value. Called for side effects (prints formatted summary).
Strongest Neighbor Coherence (SNC)
Description
Computes Strongest Neighbor Coherence (SNC), a rotation-free structural diagnostic that evaluates how well each item aligns with its top-k most strongly correlated neighbors.
Usage
snc(R, k = 2, factors = NULL, digits = 3)
Arguments
R |
A square item correlation matrix (symmetric, 1s on the diagonal). |
k |
Integer. Number of strongest neighbors to use for each item (default = 2). |
factors |
Optional. A vector of factor assignments for items, used to compute group-level means. |
digits |
Number of decimal places to round to (default = 3). |
Value
An object of class "snc" with:
- overall
Mean SNC value across all items
- items
A data frame of item-level SNC values
- factors
(Optional) A data frame of factor-level mean SNC values
Examples
R <- matrix(c(1, .6, .3, .6, 1, .5, .3, .5, 1), 3, 3)
rownames(R) <- colnames(R) <- c("Item1", "Item2", "Item3")
snc(R)