Championing data analysis at the second A2DataDive

One year ago, Ann Arbor welcomed its first datadive, a weekend-long event open to the campus and local community that brought together data geeks, people who were curious about data analysis, and people from nonprofit organizations looking for help answering some research questions about their data sets. As one of the clients last year, we learned immensely from the A2DataDive debut. This year, School of Information masters students Nikki Roda and Claire Barco resumed their leadership roles for its second manifestation.

The A2DataDive is always a jam-packed, super productive weekend. I’ll try to do my best to give you a very condensed overview of the event. After I give you the low-down on round 2 of the datadive in 2013, I also want to highlight some outcomes from round 1 last year. Stay tuned for a follow-up blog post with highlights from interviews with this year’s participants.

Photo CC BY Regents of the University of Michigan. Taken by Dave Malicke.

Overview of A2DataDive 2013

There were four local nonprofit organizations and over 100 registered participants for this year’s event, which was coordinated by the School of Information, with Open.Michigan included among the sponsors. This year, the organizers arranged several short data jams lead by a resident data librarian at the university to introduce people to some of the nonprofit datasets and basic principles of data analysis several months before the weekend-long event. Also new this second round, the organizers created the role of “data ambassadors,” who are student volunteers that liaise with the nonprofits months ahead of time to prepare the data and research questions for the big weekend.

Here are the final presentations:

Outcomes from A2DataDive 2012

For A2DataDive round 1, the Open.Michigan initiative was in the unique position of both co-organizer and client. The two clients were Focus Hope and the African Health OER Network. The African Health OER Network is one of our flagship projects here at Open.Michigan. It’s not technically a nonprofit organization but rather a joint project of University of Michigan and several other nonprofit educational institutions. The datadivers invited us to be a client anyway.

Our African Health OER Network datadive project had two paths – one to analyze our YouTube analytics data and another to analyze our social network from CiviCRM. We left the event with a few new tools, useful visualizations, and deeper insights into our audience and collaborators for the African Health OER Network. Some concrete examples of how Open.Michigan has integrated our new knowledge over the past year:

  • One of the groups did a word frequency cloud of the comments on our YouTube videos, which showed the words “thanks” and “thank you” were among the most frequent terms from users. We included that tidbit in multiple reports and presentations about the project. We even ran updated analyses several times since then as we received new user comments. We were pleased to confirm that the theme of gratitude was still strong.
  • The month following the 2012 datadive, we had a funding opportunity to send one colleague from an African partner institution to present at an international conference. In order to identify which individual to send, we consulted one of the network visualizations, which showed the most- connected, most-active individuals based on their participation in past events.
  • In October last year, I facilitated the Health OER Tech Africa Regional Workshop in Ghana, which brought together technologists, multimedia specialists, and instructors from health sciences institutions across Africa. The 3-day workshop included 22 staff members from 12 institutions across six countries. On the second day, we had a 90-minute data jam (a mini datadive) where participants analyzed YouTube analytics data from participating institutions.
  • Last year, in prep for the datadive, some of us from the Open.Michigan team wrote Python scripts to pull data from the YouTube and Google Analytics APIs, which generated the spreadsheets that we used the A2DataDive in February 2012 and which I adapted for the mini-datadive in October 2012. We’re now working on a project to share those analytics in real-time on the associated course and resource pages on our website. We’ll keep you posted!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s