Starbucks Love Project

I was sent a link to the Starbucks Love Project video on YouTube, with a recommendation to enjoy it. Actually very delightful, and also a potential Global Event for assessment. I learned about it a month after the event, so it was a bit late for entry into the formal series, but worth an exploratory look. The video starts with a description and proceeds with a series of small groups from 156 countries around the world singing the Beatles song, "All You Need is Love". It is very nice to see all these people singing together, quite moving:

On December 7th, 2009 at 1:30pm GMT Starbucks invited musicians from all over the world to sing together at the same time to raise awareness for AIDS in Africa. In that one breathtaking moment, muscians from 156 countries played "All You Need is Love" together. Watch now, as musicians from all around the world come together and share a song.

Join in by lending your own voice to http://StarbucksLoveProject.com Watch streaming video from countries around the world and then join in by singing All You Need is Love yourself. For each video submitted, Starbucks will make a contribution to the Global Fund to help fight against AIDS in Africa. You can also help increase the Starbucks contribution to the Global Fund by submitting a drawing to the Love Gallery.

The global sing-along is part of our continuing efforts to help fight AIDS in Africa. In just one year in partnership with (RED)tm, Starbucks has generated money equivalent to more than 7 million days of medicine to help those living with HIV in Africa.

The analysis I did was of data corresponding to the time of the Youtube video, which is 4 minutes and 4 seconds long. The graph is a rather pretty outcome, with a surprisingly strong trend. It's not appropriate to do statistics on an informal event, and in any case such a short piece or indeed even longer single events are not reliably interpretable, but here is what that 4-minute data segment looks like.

It is important to keep in mind that we have only a tiny statistical effect, so that it is always hard to distinguish signal from noise. This means that every "success" might be largely driven by chance, and every "null" might include a real signal overwhelmed by noise. In the long run, a real effect can be identified only by patiently accumulating replications of similar analyses.