This month Nature Methods gathers together 35 “Points of View” columns on data visualization. This is a great resource for scientists and others presenting research data – from using pencil and paper for data exploration to typography and color. Check it out at: http://blogs.nature.com/methagora/2013/07/data-visualization-points-of-view.html.
Choose a subject, any subject. Andy Kirk has published a great directory of sites and services for accessing (mostly) structured data sets at his site, Visualizing Data. It’s nicely organized into these categories:
I can not wait to start exploring!
Nathan Yau provides a great summary of resources for improving your data visualization skills in this post on Flowing Data.
- Read the classics of Edward Tufte, like the Visual Display of Quantitative Information and others (including Yau’s excellent books)
- Gather data
- Do visualizations
- Share visualizations and get feedback
- Go back to your readings and start again
Yau’s post is titled Getting started with visualization after getting started with visualization – and points to iterative aspects of learning by reading, doing, sharing, doing again.
The New York Times‘ Bits Blog profiles Alastair Croll’s book Lean Anaytics at
Interesting ideas about experimentation, failure and leadership.
Experimentation, of course, involves a lot of failure, as failure is where most learning takes place. Data around the failures of others are collected and studied as part of the overall process now. Data on failure is cheaper to create, and cheaper to come by. That is another way of saying that people are more likely to make new and interesting mistakes, instead of the same old ones, which is probably a good thing.
One big result of this failure-driven world, Mr. Croll says, is that organizational leadership is changing toward a more structured learning environment. “In the past, a leader was someone who could get you to do stuff in the absence of information,” he says. “Now it’s the person who can ask the best question about what’s going on, and find an answer.”
Favorite quote from Nick Kolegraff’s Do You Need a Data Scientist? on the O’Reilly Media site:
Focus on getting accessible quality data and solid reporting. Then worry about data science. You’ll save money and efficiency.
Good data is not always big data, so let’s not over-engineer our solutions.
Beautiful visualizations as well as inspiration for learning R
Geography of Tweets
Good ideas in the Harvard Business Review blogpost, “A Data Scientist’s Real Job: Storytelling”. It’s an account from the DoSomething.org staff who review gobs of data in order to determine the best story to tell to move their readership towards action.
Using Big Data successfully requires human translation and context whether it’s for your staff or the people your organization is trying to reach. Without a human frame, like photos or words that make emotion salient, data will only confuse, and certainly won’t lead to smart organizational behavior.
Data gives you the what, but humans know the why.
So…how to avoid the data deluge and compile data that motivates towards change?
- Look only for data that affect your organization’s key metrics.
- Present data so that everyone can grasp the insights.
- Return to the data with new questions.
The important conclusion is that the data scientist’s job is qualitative – “asking questions, creating directives from our data, and telling its story.”
Read the full post at:
NPR aired a great story today about squishy data that is perceived to be more solid that it is. The happiness index is an example that we’d expect to be subjective, but then so is the measurement of unemployment.
Marshall McLuhan is evoked in this commentary on content as influenced by the medium.
From the “Greatest Hits” of the Pew Internet Survey –
One of the “mega
takeaways” – “People love their libraries even more for what they
say about their communities than for how libraries meet their