Scatterize

Data Exploration Made Awesome

It happens all the time at our lab: someone’s giving a presentation that includes a bivariate scatter plot. And someone raises her hand and says “What if you exclude those two outliers?” And someone else asks “What if you covary for age?” If the presenter is very sharp, those are the next two slides. Much more often, the response is “I’ll try that and get back to you.”

BOOOOOOOO-RING.

Scatterize lets you upload a CSV file, plot the data, and in your browser in real-time, exclude outlier points and include nuisance variables. Every variation of your plot gives you a distinct URL — so you can prepare your chart and send it to a colleague.

Planned features:

• Residual diagnostics
• Spearman’s (rank-order) statistics
• Bulk regression — choose a DV and a set of variables of interest; get a quick t, p, and R^2 for each.

The code is available here at https://github.com/njvack/scatterize, and there’s a demo up at http://someday.example.com/scatterize

Requirements

Requires Python 2.7+, Flask, NumPy, and SciPy.