Data Resource Roundup, Vol. 9

July 26, 2017

By Bill DeBaun, Director of Data and Evaluation 

Welcome to the ninth edition of the Data Resource Roundup! In this series we periodically share resources, including blogs, courses, white papers and other tools, that cover various aspects of data. Whether it’s better managing and tracking data or getting your organization to become more data-driven, it will all be here in the roundup. Have your own resources that we should feature here? Be sure to let me know about them via email or by putting them in the comments. Previous volumes: 1a1b2345678.

BETTER KNOW A PROBABILITY DISTRIBUTION: Ready to increase your water cooler nerd cred? (I use nerd in the fondest possible connotation of the word in this case.) Look no further than this crib sheet on the various probability distributions commonly used in statistics. This is really fun for everyone. Not sure what a probability distribution is? Crash course. Haven’t thought about distributions since your 8:30 a.m. statistics class? Refresher. Perfect and makes a great birthday present.

DATAVIZ, ALL THE WAY DOWN: There are a number of topics associated with data visualization, such as data cleaning, dashboarding, converting data to new forms, etc. This warehouse of tools has something in all of these categories and more. If you have been tackling a dataviz problem, start here and see if you can’t turn up a solution.

PERCENTAGES ARE 5,330,436% MORE MEANINGFUL AFTER READING THIS POST:  “Using percent here to make the numbers sound bigger and more impressive is backfiring, I believe, because many people don't understand percentages all that well – and in particular percentages way above 100,” Robert Kosara argues in this post about how people have cognitive trouble putting very large percentages into context. Just convert it to “times” instead.

USE YOUR BRAIN: Tableau is here with a thoughtful (pun very much intended) post about how data visualization can use psychology to be more effective. Taking advantage of the way people think about spatial context, shapes and icons, and color can make data visualizations more easily understood and engaging for viewers.

PREDICTING SUCCESS RESPONSIBLY: New America’s Manuela Ekowo asks, “Colleges must ensure that all members of society — including the least privileged of them — have a chance to access a quality education. So how can predictive data be used to increase underrepresented students’ odds of admission rather than reduce them?” Interested in more on predictive analytics in the postsecondary space? New America has you covered.

GETTING BEYOND TRADITIONAL MEASURES: The Association for Institutional Research fields a mailbag question about metrics that colleges and universities can track, beyond retention and graduation. Programs engaged in success work will likely find these insights helpful even if they aren’t prepared to track course-level data themselves.

COPIES OF COPIES OF COPIES: Public Profit has some tips on mission-driven organizations looking to replicate. Given how many NCAN members think about replication on some level, whether it’s to an additional school or city or state, this has some insights that lay out the broad strokes of what to think about at the outset. This is hardly the end-all, be-all of replication, but it is a good place to start.

MAYBE DON’T FOLLOW THESE STEPS: The Nonprofit Technology Network (NTEN) wants to tell you how to ruin your organization’s productivity using a database in five easy steps. If your organization is using a database now or is thinking about migrating to one, go ahead and heed these caveats.

THIS INSIGHT BROUGHT TO YOU BY THE COLOR GREEN, THE LETTER “Q,” AND DATA DISAGGREGATION: Here at NCAN we often extol the virtues of disaggregating data to members. Breaking data down by student characteristics, whether they be demographic or service-driven, can identify hidden income and outcome gaps that programs would want to address. The Annie E. Casey Foundation has a short post on why disaggregation and data presentation both matter when organizations use data to address equity issues.

Back to Blog

Leave a Reply:
Login
 
 
 

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License