Creative Commons license icon

Reply to comment

I think your position on the direction of causation is sound, but representing a -0.56 Pearson correlation coefficient as "there may or may not be" is a bit misleading. It's more about the variance of that correlation - i.e. is X always correlated with Y by a consistent amount - regardless of that amount. (or: "Do the points fit a straight line, regardless of its angle?")

What I think you're getting at is refuted by the grey area, showing uncertainty. We don't know the level of significance; 5% is common, though it could also be 10%. But at that level, even in the most optimistic case, going from 0 to 100% religiosity averages out to more than a 0.5 drop in happiness; and with the same data, it could be over 2. The best-fit line shows a ~1.5 drop, from ~6.5 to ~5. This is pretty big considering that difference between the most to least happy countries is only 3.5.

My reading of that graph is that religiosity isn't necessary for happiness (which shoots a hole in those "hole in the heart" arguments); and that, for whatever reason, more religious countries tend to be significantly less happy, to a relatively large extent. Of the 14% of countries which achieved a 7 or higher happiness rating, none were above 60% religiosity, even though this group had almost half of the countries sampled.

The graph shows clear exceptions - Brazil is relatively religious, but happy; Bulgaria is irreligious, but unhappy. But they are exceptions, and merely indicate that happiness isn't solely correlated with religiosity. Which makes perfect sense - Brazil has a lot of things going for it, despite its problems; Eastern Europe has many issues.

Reply

  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <img> <b> <i> <s> <blockquote> <ul> <ol> <li> <table> <tr> <td> <th> <sub> <sup> <object> <embed> <h1> <h2> <h3> <h4> <h5> <h6> <dl> <dt> <dd> <param> <center> <strong> <q> <cite> <code> <em>
  • Lines and paragraphs break automatically.

More information about formatting options

CAPTCHA
This test is to prevent automated spam submissions.
Leave empty.