This study investigates the evolution of social groups in two types of networks: one representing scientific collaborations and the other representing mobile phone communications. The researchers developed a new algorithm based on clique percolation to analyze the time-dependent changes in overlapping communities. They found that larger communities tend to persist longer if they can dynamically adjust their membership, suggesting that adaptability is key to their stability. In contrast, smaller communities are more stable when their composition remains unchanged. The study also shows that the time commitment of members to a community can be used to estimate the community's lifetime. The data sets include monthly records of scientific publications and phone call records. The collaboration network is dense with many overlapping communities, while the phone call network has less interconnected communities. The phone call data reflects instant communication, while the publication data reflects long-term collaborations. These differences suggest that the community evolution in both networks reflects general characteristics of community formation. The communities were extracted using the Clique Percolation Method, which identifies communities as overlapping groups of nodes. The study found that the average correlation between community states decreases with time, indicating that communities change over time. Larger communities change more rapidly than smaller ones. The study also found that the lifetime of a community is related to its stationarity, with larger communities having shorter lifetimes and smaller communities having longer lifetimes. The study also found that the probability of a member leaving a community is related to their connection strength to members outside the community. Communities with higher internal connection strength tend to have longer lifetimes. The study concludes that small communities are more stable when their membership is static, while large communities require dynamic membership to persist. The findings suggest that the stability of a community depends on its size and the dynamics of its membership. The study provides insights into the fundamental differences between the dynamics of small groups and large institutions.This study investigates the evolution of social groups in two types of networks: one representing scientific collaborations and the other representing mobile phone communications. The researchers developed a new algorithm based on clique percolation to analyze the time-dependent changes in overlapping communities. They found that larger communities tend to persist longer if they can dynamically adjust their membership, suggesting that adaptability is key to their stability. In contrast, smaller communities are more stable when their composition remains unchanged. The study also shows that the time commitment of members to a community can be used to estimate the community's lifetime. The data sets include monthly records of scientific publications and phone call records. The collaboration network is dense with many overlapping communities, while the phone call network has less interconnected communities. The phone call data reflects instant communication, while the publication data reflects long-term collaborations. These differences suggest that the community evolution in both networks reflects general characteristics of community formation. The communities were extracted using the Clique Percolation Method, which identifies communities as overlapping groups of nodes. The study found that the average correlation between community states decreases with time, indicating that communities change over time. Larger communities change more rapidly than smaller ones. The study also found that the lifetime of a community is related to its stationarity, with larger communities having shorter lifetimes and smaller communities having longer lifetimes. The study also found that the probability of a member leaving a community is related to their connection strength to members outside the community. Communities with higher internal connection strength tend to have longer lifetimes. The study concludes that small communities are more stable when their membership is static, while large communities require dynamic membership to persist. The findings suggest that the stability of a community depends on its size and the dynamics of its membership. The study provides insights into the fundamental differences between the dynamics of small groups and large institutions.