Computer-mediated communication, compared to face-to-face:

Computer-mediation reduces social PRESENCE

* Less information, different information, less redundancy

* Much harder to exert social pressure--NORMATIVE influence decreases

*Must make things EXPLICIT -- in FTF, much is IMPLICIT

FTF is always Synchronous; CMC can be Synchronous (Chat rooms) or Asynchronous (e-mail, for example, or Motet)

Effects vary depending on the TASK

and also on the FAMILIARITY of members

The task-media fit model predicts that different media will be better depending on the demands of the task for "information richness."

Here's my abbreviated reproduction of the graphic from Hollingshead, McGrath, & O'Connor (1993). Group task performance and communication technology: A longitudinal study of computer-mediated versus face-to-face work groups. Small Group Research, 24 (3), 307-333. Good, marginal, poor refer to the fit between medium and task type. "Rich" means lots of redundant cues, such as facess, sound, words, gestures. "Lean" means there are few cues -- only the text, in the case of e-mail.

Communication medium
Task type	Computer	Audio	Video	Face-to-face
Generate ideas/plans	Good	Marginal too rich	Poor too rich	Poor too rich
Intellective: choose correct answer	Marginal too lean	Good	Good	Poor too rich
Judgment: choose preferred answer	Poor too lean	Good	Good	Marginal too rich
Negotiate conflicts	Poor too lean	Poor too lean	Marginal too lean	Good

Research supports the claim that computer-mediated groups should do better than face-to-face on GENERATE tasks, specifically when brainstorming.

Hollingshead et al. found that FTF groups outperformed CMC groups on intellective tasks, which would not be predicted by this model. However, this finding was for groups that were just forming; in later weeks, both types of groups did about equally well on another intellective task. On one of the negotiation tasks in their study, FTF groups outperformed CMC groups. This is what the model would predict.

A robust finding for decision-making tasks is that synchronous computer-mediated groups take much longer to reach a decision than FTF groups. This may be because normative influence and consensus pressures are weaker (the same explanation for why CMC groups are BETTER at idea generation--they are less inhibited by the presence of others).

Familiarity matters; if you already know the person FTF, then the effects attributed to anonymity and depersonalization should be much weaker.