angler is currently still in the developmental stage, and thus has limited usage (which we illustrate below)
Sexual dimorphism (i.e the differences between males and females) is largely prevalent in the natural world, and has been widely observed to be exhibited in a range of patterns across various clades. A critical issue that hampers the study of sexual dimorphism is the lack of standardized measures in the literature to calculate sexual shape and size dimorphism. angler computes a set of commonly used measures of sexual size dimorphism (SSD) and shape dimorphism (SShD) from landmark based morphometric data for future studies to conveniently calculate standardized effect sizes of sexual dimorphism. The package additionally computes a bootstrapped-based standard error for the various measures included in the package.
Installation
This package is not available on CRAN as of this date (although it is intended to be submitted soon). You can install the development version of angler from GitHub by running the following line of code:
devtools::install_github("ntmv/angler")Quick user guide
We demonstrate the use of the current version of the package using the pupfish dataset which is exported with the package. The first step before performing an analysis using angler is to use our sex conversion function, in order to convert the sex vector of the input dataset into a compatible format.
# Load package
library(angler)
# Load pupfish example data
data(pupfish)
# Convert sex
pupfish$Sex <- with(pupfish, ConvertSexFormat(Sex)); levels(pupfish$Sex)
#> [1] "f" "m"Calculating sexual size dimorphism
angler offers two measures of sexual size dimorphism: 1) The SDI Measure (Lovich and Gibbons 1992) and 2) The log ratio of sexual dimorphism, log(F/M) (adapted with alterations from Smith 1999). We demonstrate the use of both:
# SSDI
with(pupfish, SSDI(CS, Sex))
#> [1] -0.1624368
# Log ratio
with(pupfish, SSDLog(CS, Sex))
#> [1] -0.1505185Smith 1999 has the log ratio with male biased being positive and female biased being negative (suited to mammalian research). We can have the statistic operate in that manner by setting positiveM = TRUE.
Pending Task List
- Clean up documentation of functions
- Add parallel support for strata functions which take long time to run + vectorize
- More error checks for sample sizes for within-strata bootstrapping
- General function check
- Testing
- Update README with examples
- Add vignette and pkgdown website for code documentation and examples
- General formatting