Brian Hollar reflects on happiness and Iceland:
While it is true that people in wealthier countries are happier than people in poorer countries, there is only weak correlation inside those countries between happiness and wealth. (In other words, if you live in a wealthy nation, earning a lot of income probably won’t buy you a whole lot of happiness relative to a modest income.) My intuition says that a similar relationship may hold true for marriage and religiosity — cross-country comparisons and intra-country comparisons may lead to two entirely different results. It is entirely conceivable that countries with higher divorce rates and lower levels of religiosity are “happier” than countries with lower divorce rates and higher levels of religiosity, while at the same time people in each country who have higher religiosity and lower divorce rates are happier than their fellow countrymen.
P.S. — People generally are happier the freer they are. When freedom and wealth both increase, people have greater opportunity and ability to express true preferences about both religion and marriage. This could lead to less preference falsification within society — potentially leading to higher divorce rates and lower rates of religious participation. (Timur Kuran has some excellent work on preference falsification called Private Truth, Public Lies.) I would expect societies with lower levels of preference falsification to have stronger correlations between happiness, marriage, and religiosity.