Wandering and meeting Sarah, Edouard and Ludovico

This week, I found a link to David Ranzolin’s The Data Analyst as Wanderer: Pre-Exploratory Data Analysis with R. In it David considers “answering questions about the data at two junctures: before you know anything about the data and when you know only very little about the data”.

He used metaphor of wandering to discuss data analysis. He observed “data analysts may wander but are not lost. This post is for data analysts ready to wander over their data (with R)”.

David’s questions are:

  1. What is this?
  2. What’s in this?
  3. What can I do with this?

These sprang to mind when I found Sarah Milstein’s How to Fail When You’re Used to Winning (A guide for managing morale while pushing for innovation). In her post, Sarah observed:

Innovation is a buzzword for our era. It evokes the promise of profiting tomorrow from today’s changes in technology. The word innovation implies a clean, crisp path. That’s a lie. In fact, innovation requires enormous amounts of failure — which then presents leadership challenges.

She adds:

But any team that must experiment constantly will fail a lot, and repeated failure almost always depresses people. (Original emphasis)

This part of her post struck me forcefully:

a certain amount of failure is inevitable. Accountability lies not just in individuals taking responsibility, but in teams having a consistent way to learn from those episodes. (Original emphasis)

Sarah concludes her post with this exhortation:

Your path to succeeding at failure and maintaining morale will not be linear. You’ll stumble along the way and find yourself wanting to pretend you didn’t just trip. But stick with it. Teams that can maintain good spirits during hard times tend to win, and nothing feeds morale like success.

Waldemar Januzczak was the guide in my next phase of wandering. Here in Australia they re-showed his 2009 documentary on Edouard Manet. In part of the documentary he discusses Manet’s Old Musician painting with Juliet Wilson-Bareau. Juliet pointed out Manet’s techniques in the picture and his ability to create texture in his composition.

One example was the shoes of the two young boys:

Juliet’s observation was that by rubbing the existing paint of the shoes rather than adding white, the picture takes on a different perspective.

Juliet’s knowledge of Manet made this granular insight particularly powerful and sent me off thinking about how each of use sees nuances in performance and the data of those performances. She returned me to David’s three questions for wanderers: What is this?; What’s in this?; What can I do with this?.

Fortunately, Ludovico Einaudi was there to help me with these contemplations. One commentator noted of him “For Einaudi, composition can happen in different ways. He improvises at the piano, invents melodies in his mind, but also hears them during his sleep”.

Daniel Keane (2017) said of Ludovico’s music “Throughout its course, one is never sure whether one is listening to something very old or very new.”

I think this sentiment resonates with our experiences of data wandering and why David Ranzolin’s prompts are so helpful.

Photo Credit

Le Vieux Musicien (The Yorck Project, public domain)

180309 Finds

During this week I came across some excellent resources. This post is an aide memoir for me but I hope it might be of interest.

Open Badge Template

This design your own open badge template appeared in the Wapisasa sandbox.

Data Culture

Rahul Bhargava, Catherine D’Ignazio and the Stanford Center on Philanthropy and Civil Society, have worked together with 25 organizations to create the Data Culture Project.

Data Science Text

Pablo Casas (2018) has published a data science live book. The introduction to Why this book?

The book will facilitate the understanding of common issues when data analysis and machine learning are done.

Building a predictive model is as difficult as one line of R code:

my_fancy_model=randomForest(target ~ var_1 + var_2, my_complicated_data)

That’s it.

But, data has its dirtiness in practice. We need to sculpt it, just like an artist does, to expose its information in order to find answers (and new questions).

Michael Clark’s Documents

Michael notes of his online resource shared on GitHub:

Here you’ll find documents of varying technical degree covering things of interest to me or which I think will be interesting to those I engage with. Most are demonstration of statistical concepts or programming, and may be geared towards beginners or more advanced. I group them based on whether they are more focused on statistical concepts, programming or tools, or miscellaneous.

Michael is the Statistician Lead for the Advanced Research Computing consulting group, CSCAR, at the University of Michigan. He provides analytical, visualization, code, concept, and other support to the larger research community.

March Madness

A tweet from Mara Averick led me to Sam Firke’s discussion of March Madness college basketball.

Sam shares his guide for March Madness predictions (an estimate how likely it is that Team A beats Team B, for each of the 2,278 possible matchups in the tournament).

Sam shares a link to Gregory Matthews and Michael Lopez’s (2014) paper Building an NCAA mens basketball predictive model and quantifying its success. Gregory and Michael were winners of the 2014 Kaggle competition to predict the outcome of the tournament.

Photo Credits

…so… you’re saying the round one would please her more? (Pim Geerts, CC BY-NC-ND 2.0)

sweet16wide (Andy Thrasher, public domain)

Thinking about analytics

A line in Colin Beer’s Learning analytics and magic beans (1 March 2018) “Learning analytics requires a learning approach …” sent me off thinking about how discussions about learning analytics in education might facilitate conversations about analytics in sport.

I produced this slide deck to think out loud:

Photo Credit
Making Sense Of The Data: For You And Your Coach (Jacquie Tran, 2014)