The Curious Link Between Online Gaming and Chinese Regional Culture

There had never been an easy way to map the rich diversity of Chinese culture until demographers started analysing the playing patterns behind online games 

With more than 1.3 billion people, China is the most populous nation on Earth. It is also one of the oldest civilisations that is today filled with a rich diversity of peoples and cultures speaking some 292 different languages.

But classifying and mapping these cultures is a tricky business, not least because of the enormity of the task. Gathering and analysing data on this huge scale simply hasn’t been possible.

Until now. Today, Xianwen Wang at the Dalian University of Technology in China and a couple of pals, have created a detailed map of Chinese regional cultures by studying the popularity of online games in different parts of the country. The result is a fascinating insight into an aspect of Chinese culture that is otherwise largely impenetrable to westerners.

First some background. Historically, demographers have thought of Chinese cultures as divided in various simple ways. The simplest of these is to divide the country roughly into two parts: north and south along the line of the Qinling Mountains and Huaihe River. Another is to divide the country into three parts: the eastern, central and western regions. There are also six economic zones, 34 administrative divisions and so on.

How well these divisions capture the cultural differences between regions in China is open to question. “These classifications are rather rough,” say Xianwen and co. There is certainly little published evidence to go on.

So these guys have taken a different approach. They point out that card games and board games are strongly embedded in Chinese culture. Consequently, there are many hundreds of different games played all over the country.

What’s interesting about these games is that they have evolved differently in different regions. “For example, there are more than 30 kinds of mahjong in China,” say Xianwen and co.

People in Shanghai play Shanghai mahjong whereas people play both Ningbo mahjong and Hangzhou mahjong in the neighbouring province of Zhejiang. A similar variation exists in card games.

Xianwen and co’s idea is that the popularity of these games in different parts of the country is a proxy for the cultural differences in these places. So two regions are culturally similar if the same games are equally popular.

That’s the theory—but where to get the data to take explore the idea? It turns out that exactly this kind of data is available from gaming websites, which offer online versions of all these varieties of games for anybody to log on and play.

The biggest of these websites is called QQ, an instant messaging-type service run by the Chinese internet company Tencent (incidentally, the third largest internet company on the planet after Google and Amazon).

QQ offers almost 200 different types of board game or card game. And of its 800 million users, about 8 million are playing games at any instant.

To get an idea of who was playing which game in what part of the country, Xianwen and co took a snapshot on 9 August 2013 at peak usage time, just before 10pm. They found some 7.55 million players online and gathered data about which game they played and what part of the country they were in.

The first impression from this map is that there is a clear distinction between playing patterns in north and eastern China versus western China. “Most nodes are clustered together in the east coastal regions [and] north China plain (including Beijing, Tianjin and Shandong province),” they say.

Western China is almost completely dark by comparison, a reflection of the economic contrast across the country. Younger people are particularly attracted to the northern and eastern economic zones who in turn are more likely to play online games.

But the data allows much more detailed analysis. For example, Xianwen and co map the popularity of 13 variations of mahjong around the country. “Each mahjong game has its own sphere of influence, there may be some overlap between two games, but generally the boundary of games is distinct,” they say.

As a kind of control, they also map the geographical popularity of more modern games such as Tractor and Happy Bullfight that do not have region variations and are not so closely linked to traditional Chinese culture. These games turn out to have little geographical diversity —their popularity is evenly distributed around the country.

Finally, they use the variations in popularity of 47 different games to create a detailed map of cultural variation across China. This clearly maps out 17 well-defined regional clusters across the country.

In general, the boundaries of these clusters coincide well with the geographical boundaries of provinces, although there are a few exceptions.

That’s a fascinating study. Most westerners have a poor understanding of the cultural variations across China—indeed this kind of insight must be hard even for Chinese authorities to pick up.

That’s the power of data mining. This kind of analysis of huge datasets is straightforward and produces unprecedented detail and insight. A powerful and valuable tool that is clearly destined to reveal more secrets.

Ref:arxiv.org/abs/1310.0621: Games and Culture: Using Online-gaming Data to Cluster Chinese Regional Cultures