We have conducted a series of experiments in order to study the influence of framing modifications on results of traditional experiments of the Tragedy of the commons, that is, Public Goods Games (PGG). Until date almost all Public Goods Games experiments have been performed with small or mid-sized groups (maximum 100 people), to extract conclusions that possibly apply to current societies (with are quite bigger than 100 people). In this work, we test if small group results still hold for bigger group sizes. Moreover, when experiments are conducted in the lab, people make decisions one after the other in relative small periods of time. We study how time-separated decisions (longer decision intervals) influence cooperation. Finally, in view of the intrinsic difficulties in giving feedback to participants when the group is large, we also analysed the influence of the way information about group contributions is exposed to the participants. In terms of protocol, the experiment allowed us to estimate the dropout rate of people playing this kind of experiments online.