Religion and corruption

There is no corruption-free country in the world—every country experiences some level of corruption. But the huge differences in the level of corruption between countries is striking. This much is well known.

What is less well known is how corruption is related to religiosity—the most religious countries tend to be the most corrupt countries. But how do I know this and what conclusions can we draw from it?

The chart shows public sector corruption vs religiosity for 140 countries (this is every country for which both corruption and religiosity data are available). See "Methodological notes" below for a fuller explanation of the datasets used.

The chart clearly shows that the most corrupt countries (Somalia, Sudan and Afghanistan) are very high on the religiosity scale whilst the least corrupt countries are very low on the religiosity scale (Denmark, New Zealand and Finland). It can be seen that this general relationship holds true for most countries.

It is interesting to look at the outliers—countries that are more corrupt (or less corrupt) than expected for their religiosity. I've drawn diagonal dotted lines to show the outliers. The countries that are more corrupt than their religiosity would suggest are all communist or ex-communist countries. I have marked Russia, Vietnam and Czech Republic but all the countries below the lower dotted line fall into this category.

The countries that are less corrupt than expected are a mixed bunch. Qatar and UAE are both Muslim countries but Malaysia and Singapore have many religions—Muslims being the biggest single group in Malaysia and Buddhists holding that position in Singapore.

It would be interesting to map GDP/capita (income) vs corruption. That may correlate even more closely than religiosity vs corruption. Nevertheless, the chart does illustrate the point that highly religious countries tend to be much more corrupt than irreligious countries.

Finally, correlation is not causation. So I am not saying that being religious MAKES people corrupt. It may, but this data cannot establish that. However, there is one conclusion that we can draw from this data; being religious does not make people honest and not being religious does not make people dishonest.

...........

Methodological notes.
Corruption is abuse of power for private gain. Transparency International is an organisation set up to monitor and expose corruption around the world. It has published the Corruption Perceptions Index (CPI) annually for 20 years.

The CPI rates the level of public sector corruption in a country on a scale of 0-100 where 0 is the highest possible level of corruption and 100 means totally free of corruption. Of course, corruption is hidden so it cannot be measured directly but the CPI uses multiple indicators and cross checks and is respected by governments and international organisations.

I have used the 2014 CPI data for my analysis which is the latest available.

A link to Transparency International's website is given below. There you can find a full description of their methodology and detailed data for 177 countries.
http://www.transparency.org/cpi2014/results

I use Gallup Poll research to measure religiosity. Gallup runs a poll from time-to-time which asks respondents "Is religion important in your daily life?" The least religious country measured in this dataset was Estonia with only 16% answering "Yes". There were three countries in which 100% of respondents answered "Yes"; Bangladesh, Niger and Somalia.

This Gallup data was from 2009 which is the latest available. Since both religiosity and corruption change slowly, it is unlikely the difference in collection dates will distort the correlation to a material extent.

The full dataset is available here:
http://en.wikipedia.org/w…/Importance_of_religion_by_country