- Dividing the data: Hidden insights are obtained when your data is segmented more. The more you segment your data, the better insights you obtain. In the video, the story of African sub-Saharan region having the lowest GDP vs Child Survival rate is completely different when the data is segmented country wise for that region. We realize that after segmentation, countries like Mauritius has a much higher ratio (than average) as compared to other countries in the same region.
- Treating each data point separately: Each data point can be associated with a different problem and after carefully analyzing the data point, you can provide solutions accordingly. For e.g. the same solution can’t be applied to the poorest of Nigeria v/s the richest of South Africa.
- Usage of Idioms: The idiom you use for visualization should immediately provide you the information that you are trying to convey and what your chart is measuring.
- Checking your legibility: Run it by someone who has never seen your visualization, and ask them to tell you what the chart is supposed to be illustrating. The longer they take, the worse you’ve done.
Reference: https://www.ted.com/talks/hans_rosling_shows_the_best_stats_you_ve_ever_seen
http://online-behavior.com/analytics/effective-data-visualization
Nicely articulated the tips from the Hans Rosling video.
The talk clearly shows how to link two completely different data sets and combine them to present an insight based on one common factor. The main idea is to find the idiom, find the accurate data and support your claim. Even though the visualization seems simple and engaging, the visualization presents a mind boggling correlation between the GDP and Child Survival Rate.