The dataset contains information about 333 golf players in the PGA Tour on the period 1996-2006. The dataset has ben used to understand how players' previous performance moderates the effect of prize on tournament self-selection. The dataset specifies which tournaments each player entered in each season as well as information on the player, player's previous performance and the tournament. In total, the dataset encompasses 54,915 self-selection decisions for a total of 395 PGA Tour events. The selection criteria for tournaments and players is the following: The PGA Tour consists of 45 to 49 tournaments, depending on the season, with usually one tournament being held per week. With few exceptions, 150 players enter each PGA Tour tournament. In contrast, there are 250 players per season who own a PGA Tour card, which gives them the right to play in its tournaments. Since there are less spots available in a tournament (150 players) than PGA Tour cards (250 players), the PGA Tour has designed, through a set of exemption rules, a priority ranking that determines who enters the tournaments. At the top of the exemption ranking are the fully exempt players, who have the right to enter any tournament of their choice. The dataset focuses exclusively on players who have fully exempt status during a given season. In total, 333 fully exempt players for the 1996-2006 period. In addition the dataset only has information about regular tournaments, which are tournaments owned and operated by the PGA Tour. They stand in contrast with the non-regular tournaments, which are also part of the PGA Tour schedule, but are owned by other entities other than the PGA Tour. Excluding non-regular tournaments means that the sample, composed only of regular tournaments, is rather homogeneous in terms of prestige. This allows to control for the potential effect that non-monetary incentives such as tournaments’ reputation may have on self-selection, thus reducing noise in the data. The second reason for excluding non-regular tournaments is that each of them has unique entry criteria, so that even fully exempt players are not necessarily exempt in non-regular tournaments. Moreover, the data is reduced to solely regular tournaments that have been held on the same golf course during the period under study. Accordingly, three regular tournaments with changing venues were removed. The final dataset is left with 34 to 37 tournaments per season, depending on the year. The period of study starts in 1996, this is because it is not possible to identify the players who were fully exempt for the season prior to that. It ends in 2006, because one year later the PGA Tour entered in a new era with the introduction of the FedEx Cup.