mode helix
NOW LIVEEmpower your end users with Explorations in Mode.Try it now

Analytics Dispatch 8/23/2021

Visualizing ordinal variables

ONLY CONSTANT IS CHANGE
Visualizing Ordinal Variables
“My sense is that we often don’t know what to do with ordinal variables, partly because they don’t fit neatly into traditional analysis routines. Ordinal variables are not numeric, they’re not categorical, they’re strange... [O]rdinal data is also somewhat complicated to visualize whenever there’s any sort of change involved.” - Octavio Medina

DOUBLE, DOUBLE, TOIL AND TROUBLE
How Data Science Pinpointed the Creepiest Word in “Macbeth”
It’s not the word you’d expect—and it appears in this very sentence. - OneZero

RESEARCH SAYS...
Mitigating Dataset Harms Requires Stewardship: Lessons From 1000 Papers
Efforts to implement higher ethical standards and transparency in the machine learning dataset creation process can be more effective if we understand of how datasets are used in practice in the research community. - arXiv.org

SOURCE LOCAL
Keep Your R Scripts Locally Sourced
A really bad debugging session completely broke this person’s mental model for how one bit of R code should work. - Higher Order Functions

STUDY UP
Introduction to Causal Inference from a Machine Learning Perspective
Here’s Sean Taylor on why causal inference is important for working data scientists: “Typically in causal inference, researchers try to estimate some quantity of interest one time for a publication. In industry, we must build systems to reliably estimate quantities, at scale, over time, for a variety of contexts.” - Brady Neal

decorative particle

Get our weekly data newsletter

Work-related distractions for every data enthusiast.