Happiness, Wealth and Charity – Correlation, NOT causation

15 09 2010

You may have heard of the study that came out recently that showed a relationship between charitable giving and happiness.  The results, in a nutshell, showed a stronger correlation between happiness and charitable giving than between wealth and charitable giving.

The reason I bring this up is because I want to make sure we all have our critical thinking caps on when we’re reading about studies like this.  When the media report on such studies as this one, media consumers commonly assume that the results illustrate a causal relationship when it is really only correlational in nature.  If that is Greek to you, there’s an explanation just ahead.

I first heard about this study on the radio and later read about it online.  Here is the article (for the record, I don’t usually read the Christian Science Monitor, I just followed a link to the article).  The first correlation/causation issue arises in the title of the article, “Degree of charity depends on happiness more than wealth,” which implies that happiness causes charitable giving (ie. charity depends on happiness, ergo happiness happens then charity follows).  Then the first line of the article, “So it’s true. Money doesn’t buy happiness. Giving does.” implies the opposite, that charitable giving causes happiness.  It may seem like a trivial thing to bring up and maybe in this case it is.  It probably doesn’t matter much in which direction causation flows when it comes to giving and happiness.  But the fact is that confusing correlation and causation can cause people to arrive at incorrect and possibly harmful conclusions.

In any research methods course they will pound into you that “correlation does not equal causation!”  The example that I remember most vividly from my college days is that there is a correlation between ice cream sales and crime.  When ice cream sales increase, crime also increases.  This is true.  What is not true is that which is tempting (and sensational) to believe, that one causes the other.

It would be very interesting if people responded to crime by drowning their sorrows in more and more ice cream.  Or if the elevated blood sugar resulting from increased consumption of ice cream caused some people to go a little nuts and commit crimes.  That would make for great news.  But all we can accurately say is that when one increases, we see an increase in the other.  One possible explanation of the ice cream/crime phenomenon is that ice cream sales increase during warm weather and warm weather is also a more likely time for criminals to be engaged in illegal activity.  But that is also just a guess.  That’s the nature of correlation, we can see a relationship but we can’t draw conclusions about which causes which.

The whole point is that this study on charitable giving, like so many other studies that show up in the media, measures correlation only.  It is interesting to see that these things are related, but that’s all we can know from a correlational study.  So I exhort my readers to be good critical thinkers and be wary of drawing erroneous conclusions from correlational studies based on the assumption of causation.




Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: