Treating statistics sensibly

By Ken Perrott 30/09/2010 2

People love to quote statistical studies to support their claims. And often its just a matter of confirmation bias. The statistical studies may not provide the support required – or may suffer from all sorts of flaws.

We see a lot of this in discussions on health, diet and life style. But I have also noticed statistics being liberally thrown around when religion and religious attitudes are discussed. If there is any area ripe for confirmation bias this is certainly it.

Consider this little graphic below which appear at a dating site OKCupid (see The REAL ‘Stuff White People Like’). Just imagine what negative conclusions one could draw about religion from that. To be fair, most references I have seen to it (all atheist – strangely, no religious sites are quoting it) do advise taking it with a grain of salt. (If you are interested have at look at the source. It provides other statistics from the study which will help make sense of this graph).

On the other hand, I have had statistical studies quoted at me which claim to “prove’ the religious people are happier, more honest, more moral, etc. Typical those quoting the studies have never bothered to check out the details and always ignore studies which might have provided different conclusions. In other words the normal confirmation bias.

Religiosity and average income of countries

The Gallup polling organisation, though, has produced some interesting statistics relating to these issues. Recently they reported data showing there was a correlation between religiosity and per capita income (see Religiosity Highest in World’s Poorest Nations). Religiosity was measured by the self declared importance of religion to individuals. Charles Blow of the New York Times provided a helpful graphic of the Gallup data (See Religious Outlier):

When the median responses are plotted for different per-capita ranges we can see a strong relationship between religiosity and per-capita  income:

This is not new. Other researcher have found a similar relationship.

Religiosity and happiness

The Gallup organisation has also reported data indicating the relationship between religiosity and  factors related to the emotional quality of life (see Religion Provides Emotional Boost to World’s Poor).

It appears that the non-religious in countries with a low average annual income were more likely to experience negative emotions than the religious. (These tend to be countries with higher religiosity).

However, for countries with higher income average annual income religiosity did not seem to influence happiness.

Digging deeper

Confirmation bias tempts people to interpret statistics in a way favourable to existing prejudices. For example, the first figure will temp atheists to claim that Christianity is associated with poor writing proficiency. And the religious may claim, on the basis of the Gallup statistics, that religiosity is correlated with individual happiness.

Unfortunately, motivated people will quote such statistics without looking any deeper than the superficial numbers.

Proper interpretation of the first figure requires consideration of sampling bias, both in the general sample (registrants at a dating site) and in the self selection for the different religious classifications.

The Gallup organisation has provided more data allowing a more sensible interpretation of their statistics. Clearly one should not use this data to claim that religion makes people happier generally. A deeper look shows conclusions should be different for higher income countries than lower income ones.

My take on the Gallup data is that religion plays a more functional role in poorer countries. It is more important to the individual in terms of social support and contact and the everyday organisation of life. On the other hand the non-religious person in such an environment may feel less social support, involvement and acceptance. Not conducive to happiness.

In contrast religion has a less important role in more developed, more secular countries. Governments and non-religious organisations are more likely to provide social support (welfare). There are far more avenues open for social involvement in secular societies. Consequently the non-religious are less likely to suffer from social exclusions or opportunities for social involvement.

Similar articles

Enhanced by Zemanta

2 Responses to “Treating statistics sensibly”

  • There was another graph going round the other day, showing that atheists, agnostics, Jews and Mormons had the best general knowledge of religion. That study pointed out that the single biggest predictor of religious knowledge was the level of education achieved.

    So I might wonder if the real message of the first graph is that people with a university degree are more likely to be atheist or agnostic, or Jewish or Buddhist..

  • I think it’s true that in the USA and Europe the education level of atheists/agnostics is generally higher than for Christians.

    But I also think the figures do represent different selection processes. In USA and Europe where Christianity/atheism etc is most common the use of computers is more widespread so the effect of education shows up (there is less selection for education). But in Asia where most Hindus and Buddhists are there would be a selection of the more educated, western, less traditional groups. Both in use of computers and use of dating sites. So I think that pushes the education level, familiarity with English,writing proficiency up for those groups.

Site Meter