As social foraging groups increase in size, individual levels of scanning decrease while time spent feeding increases. Traditionally, anti-predatory considerations provide the preeminent explanation for this scanning-group-size effect, but contest and scramble competition could also account for the results. The anti-predatory hypothesis predicts that the reduced need of vigilance in groups allows subjects to decrease their scanning and consequently increase their feeding rate. Contest competition tends to decrease the group foraging benefits especially for subordinate individuals. However, the scramble competition hypothesis predicts that competition drives individuals to increase their feeding rate by decreasing their handling time per peck and, therefore, is consistent with an increased rate of feeding in the inter-scan intervals. I used this rationale to explore the contribution of each factor to the scanning-group-size effect of house sparrows ( Passer domesticus ) observed while they fed from an artificial seed patch during the winter. The inter-scan pecking rate of individuals, excluding scanning, increased with group size while handling time per peck decreased. Dominants pecked more quickly than subordinates at group sizes greater than 12, subordinates hopped more than dominants and spent more time handling each seed for group sizes greater than 18. These results suggest a contribution of contest and scramble competition to the scanning-group-size effect. Dominants appeared to be more sensitive to changes in group size than subordinates and both competitive and anti-predatory considerations might help to explain this result.