Week 5 round up

In Week 5, you had a chance to catch up, but also worked with us to crowdsource some new data. A big thanks to everyone who took part!

By playing the game, you generated data on how scenic photographs taken in different parts of Rio were. Here is a map of the data you created:


You can also view an interactive version of this map here.

You’ll be able to see straight away that some clear patterns have emerged! For example, images around the coast have generally been rated as more scenic. You can also see some patterns which relate to the economics of the city – for example, images taken around the wealthier “Zona Sul” (towards the south east of the map) have attracted higher than average ratings.

Of course, any analysis of this dataset will need to take account of a range of limitations of these ratings. For example, the ratings relate to the photographs, rather than being direct ratings of the locations themselves. At the time of writing, the interactive page reveals that over half of the ten most highly rated photographs were not taken by a Google Street View car but were uploaded independently. This relates to a point many of you raised – pictures from Google Street View do not necessarily provide the most pleasing views!

We’ll be carrying out a very careful analysis of this dataset, and we’ll keep you up to date with our findings. We hope you enjoyed this exercise, and that you now have a greater understanding of how crowdsourcing with online games can help us measure aspects of the world that may not have otherwise been very easy to measure. Thank you all once again for taking part!

Share this article:

This article is from the free online course:

Big Data: Measuring and Predicting Human Behaviour

The University of Warwick

Get a taste of this course

Find out what this course is like by previewing some of the course steps before you join: