Listing of classic R datasets used in various packages
I found all the talk about statistics to be really really annoying. I mean, dude, it's a sample size of -2-. Experiments? Sure. But statistics aren't going to help you in a situation where you are compare two groups of size 1.
"Cubism.js is a D3 plugin for visualizing time series. Use Cubism to construct better realtime dashboards, pulling data from Graphite, Cube and other sources. Cubism is available under the Apache License on GitHub."
"Cube is a system for collecting timestamped events and deriving metrics. By collecting events rather than metrics, Cube lets you compute aggregate statistics post hoc. It also enables richer analysis, such as quantiles and histograms of arbitrary event sets. Cube is built on MongoDB and available under the Apache License on GitHub."
Markdown package for the R statistical software environment. Handy for publishing R code (e.g. to Rpubs) and for integrating latex, comments and other text more seamlessly into code.
"Project Bamboo is currently piloting a directory of tools, services, and collections that can facilitate digital research. This evolution of Lisa Spiro's DiRT wiki includes new ways of browsing and commenting on the entries. Please send us feedback on how to improve the site!"
Includes a pretty interesting discussion of dealing with overdispersion in mixed-effects models (e.g., count models) by using individual-level fixed effects. It also includes a whole series of citations.
Incredible answer to this question that really makes it clear. If I teach stats, I'm going to use whuber's example.
An example of R code that tests for a mediation relationship.
Resources on statistical mediation curated by David MacKinnon. Very thorough.
David Kenny's homepage. Especially useful for resources related to mediation and moderation, but also full of stuff on other sorts of statistical modelling techniques
A somewhat hard-to-find-but-useful set of ggplot2 examples buried on Hadley Wickham's site.
*Sigh*. Another opt-in survey. This one from the Ada Initiative.
Take Blackduck's results with a giant fist full of salt, but do look at these trends.
It turns out, Google has data on enough search terms that something is always quite highly correlated with your curve! This is a crackpot conspiracy theorists dream!
A bunch of R code to fit your power law distributions.
I've heard great things about this podcast, but haven't yet got a chance to listen to it.
