Allen Downey is a professor of computer science working at Olin College of Engineering in Massachusetts. Downey examined data from the General Social Survey (GSS) which shows that the percentage of people with no religious affiliation has increased to 18% in 2010, up from 8% in 1990. This equates to a difference of around 25 million people.
Downey realized that the question of why more and more people are choosing not to align themselves religiously is a complex one, so he examined a handful of factors that he believed contributed to the trend.
The first was the fact that the number of people with a religious upbringing has dropped since 1990. People with religious upbringings are much more likely to identify religiously when they become adults. But while this was a significant factor, Downey claims it only accounted for about 25% of the increase in people choosing not to claim a religion.
Another factor he looked at was education level. 10% more people had a college-level education in the 2000s than in the 1980s, and correlation in the data suggests that this contributed to the trend as well, but Downey concluded it only accounted for about 5% of the increase in people choosing no religious affiliation.
So Downey decided to examine internet usage. Internet use in the 1980s was virtually zero, but by 2010, 53% of Americans spent at least two hours per week online and 25% spent 7 hours or more. This trend correlated very closely with the decline in religious affiliation, as evidenced in the graphs above. Downey’s study concludes that the internet accounts for at least 25% of the decline in affiliation.
Downey offers an explanation as to why he thinks increased internet use has led to less religious affiliation:
“For people living in homogeneous communities, the Internet provides opportunities to find information about people of other religions [or people with no religious affiliation], and to interact with them personally.”
It’s also possible that the causation is the other way around- that the decline in religious affiliation caused the increase in internet usage, but it would be tough to argue that. The third possibility is that some other factor caused both trends. However, Downey believes this is unlikely, saying,
“We have controlled for most of the obvious candidates, including income, education, socioeconomic status, and rural/urban environments.”
If you’re keeping count, Downey’s three factors (upbringing, college education and internet usage) only account for about 50% of the decrease in religious affiliation. So what about the other half?
The only piece of data that correlated with this other portion was birth date. In other words, the later you’re born, the less likely you are to be religious. But obviously, birth date can’t be a cause for the decrease in religious affiliation in and of itself, suggesting there is some other major factor that explains the other 50% of this trend.
What do you think that factor could be? Answer and discuss in the comments section.
Read more from The MIT Technology Review here.