We ran a large field experiment in which a charity solicited 50,000 of its donors, and tailored suggested donation amounts based on their past behavior. Two anchoring mechanisms were manipulated, the size of the first amount on the donation grid relative to the previous donation (the left end of the grid), and the rate of the increase of the amounts on the rest of the donation grid (the right end). As expected, both anchoring mechanisms have positive influence on donation amounts, but only the size of the first amount on the donation grid negatively influences likelihood of donation.
PROKOPEC, S. and DE BRUYN, A. (2009). The Impact of Anchors on Donors' Behavior: A Field Experiment. In: Advances in Consumer Research, Volume 37. Association for Consumer Research (ACR).