It seems a common sense that online opinions are good for the development of democracy. One important reason is that opinions expressed online are more free of social control and influences. Empirically, we do find more opinions expressed online than offline. But it does not mean opinions are more diverse.
Traditional approach to social influence on public opinions paid much attention on the attitude changes (e.g, conformity, consensus and so on) . Researchers on online opinions tried their best to compare the attitude changes due to social influences between Internet and offline context. Finally, they concluded that there are differences of social influences. Online opinions are less susceptible to the external factor. It’s good news for democracy.
However, these researchers have ignored one important aspect of social influence on public opinion: the expression besides the change. In the anonymous and immediate situation online, there may be little change for user’s opinions, however, the social factor may show great impact on their expression online. Such as the antagonism mechanism addressed, online users are more inclined to express when they are against others opinions than when they are for others’ opinions. It’s rather different from what attitude change perspective claimed.
The expression behavior is much important than the change of attitudes. When researchers collected opinions online, they just collected the expressed attitudes. If users did not have any expression because of any social influence, we can't measure their opinions. All the opinions online is determined (at least partially determined) by the expression rather than change of attitudes. And these behaviors of expression are influenced by social factors online, thus social influence still show their impact on public opinions online.
Internet is probably good for development of democracy, but at least is not a ideal one.
