Add effect size functions

[mccarthy-m-g]

Hi Paul,

Nice package.

As a fellow expert in nice numbers, I was wondering what your thoughts are on adding effect size functions to the package, based on a measure of effect size I have tentatively named the Nice Index.

Formulas

The Nice Index provides a measure of just how nice the numbers in a vector, matrix, or data frame are, and can be measured in absolute or relative terms.*

The Absolute Nice Index is given by:

$$ \text{Nice}_a = \text{count(nice numbers)}. $$

The Relative Nice Index is given by:

$$ \text{Nice}_r = \frac{\text{count(nice numbers)}}{\text{count(not nice numbers)}}. $$

*Apologies for the not so nice formatting, GitHub’s new math rendering apparently has some kinks, and not the good kind.

API

In terms of an API for these effect sizes, I think something like this would be… noice.

how_nice(x, type = c(“absolute”, “relative”)

Keep it nice,
Michael

[Paulj1989]

Haha. This is excellent Michael. If you fancy putting a PR together for this, it would be very welcome (I imagine the global {nice} community are gagging for additional functionality). Failing that, I'll take a look when I get a moment!

Stay nice!