Important decisions are often made by groups rather than single individuals. Scientists are quite pessimistic on the functioning of committees whose members have to collect information. The reason is that information has properties like a public good. The public good nature of information creates a free-rider problem. When deciding how much effort to put in collecting information, individual members of a committee fail to consider the benefits other individuals have from their efforts. There is large experimental literature in economics that shows that free-riding is not a sheer theoretical problem. Psychologists have also found a lot of convincing evidence for the existence of free-riding. However, there are also circumstances in which free-riding does not seem to be much of a problem. On the basis of a meta-analysis, Karau and Williams (1993) have identified four variables that particularly moderate free-riding. The most important variable is the possibility to evaluate individual members’ contributions. The second variable is task valence. Group valence is the variable third in importance. Finally, psychologists have found that the possibility to evaluate group performance reduces free-riding. The main objective of the present paper is to develop simple theoretical models that explain why and how the four variables mentioned above alleviate free-riding problems.