People contribute their services for free in many different ways. In the U.S., for instance, about one-quarter of adults offer voluntary service to community groups. Internationally, many people contribute to online communities such as Wikipedia and YouTube. What do they get out of it?

This is a topic of some discussion among researchers. Many argue that contributors would reduce their contribution in large groups because of free-riding. Theory suggests that the more people in the potential pool to contribute, the less each one of them would contribute. However, this theory cannot explain why people would still contribute even when the pool has millions of people.

Michael Zhang of HKUST and Feng Zhu of USC examine how contributions may change with group size in the context of Chinese Wikipedia. Their methodology cleverly exploits the blocks by censors in Mainland China. They found that as the number of users dwindled, so did the number of contributors. This finding cannot be easily reconciled by the classical "ree-riding" theories.

"We attribute the cause to social effects: contributors receive social benefits from their contributions and the shrinking group size reduces these benefits. We find the more contributors value social benefits, the greater the reduction in their contributions after the block," they say.

Chinese Wikipedia was founded in 2002 for Chinese speakers in Mainland China, Taiwan, Hong Kong, Singapore and other regions. The blocks in Mainland China started in 2004 and although they were at first revoked on appeal, the appeals were ignored in October 2005. From then until the summer of 2008, when the Olympic Games were staged, Chinese Wikipedia was mostly off-bounds in China.

These specific, externally-determined time limits made Chinese Wikipedia a real-life laboratory for exploring the effects on contributors to the public good. Moreover, the nature of Wikipedia made it possible for the researchers to trace thought_leaderships to contributors, identify those who still used the site after blocking, and determine how their contributions altered after the blocking.

The authors focus on a window of four weeks shortly before the October 2005 blocking, and four weeks after, and find contributions drop even for those who are not blocked.

"The block significantly reduces contribution levels and in addition, the number of [Chinese] characters added drops more than the number deleted. A back-of-envelope calculation suggests individual contributions drop by 42.8 per cent on average as a result of the block," they say.

The effects are particularly felt among those who might be more motivated by social effects to contribute to Chinese Wikipedia. The website allows contributors to set up user pages and user-talk pages, and the authors suggest those who do so are more likely to derive greater social benefits from contributing.

"People who care to write on these pages are more active in socializing with other contributors, and those who contribute more to user pages and user-talk pages also contribute more to Wikipedia thought_leaderships. Our results show they are affected more by the block and thus suggest that social effects (e.g. social interactions) could provide an important motivation for contributing," they say.

Overall, the study shows that group size can affect the level of contributions and those who care more about the social benefits of contributing react more strongly to changes in the group size than those who value them less.