Friday, July 15, 2011


I just came across this graphic over at Chart Porn. What story would you tell with this data?


  1. It would be an even better story as a Gapminder viz, so we can see the movement over time.

  2. It would seem that certainly as GDP increases there is a similar increase in the belief of evolution, but that is only part of the story. It would be interesting if this was an interactive viz of some kind where other factors could be included to aid interpretation.

    The fact that the USA is an obvious outlier gets me scratching my head, and as such the line of best fit hasn't taken the USA into account as the result would differ considerably. It seems reasonably obvious that wealthier nations would be more educated and would therefore (possibly) see things from a more scientific viewpoint, but again this is only part of a larger story. I would like to see how education standards fit into this process. Do wealthier, more educated (as a whole) nations have more rigid beliefs about things like this?

    Religion itself is also something that would play a part. I would be interested in charting the percentage of believers in evolution per nation against the percentage of people who have some religious beleif. It should in theory produce a straight line, but would it? Some people consider themselves spiritual and yet still believe in evolution rather than the concept of a world created by a deity.

    Certainly is food for thought. Good though provoking post Cole.

  3. Forget the story, what's the sample size? And, for the record, there was a third possible response ("Not Sure") which would flatten the curve.

    Outside of that, I like that the data points are color coded by geography. I just wish the background were a crisp white instead of something that looks like it was cut out of the Financial Times.

  4. tells more or less the same story! I don't have data on it, but I imagine that belief in evolution vs. "religiosity" is pretty linear, which explains why they're the same graph flipped over. (pg. 77 of Gelman et al.)

  5. I found more information here about the study including sample size. Haven't the time to look through it yet, but it might answer some more questions.