We analyze the Spence education game in experimental markets. We compare a signaling and a screening variant, and we analyze the effect of increasing the number of competing employers from two to three. In all treatments, more efficient workers invest more often in education and employers offer higher wages for workers who have invested. However, separation is incomplete, e.g., investment does not pay on average for efficient worker types. Increased competition leads to higher wages in the signaling sessions, not with screening. In the signaling version, we observe significantly more separating outcomes than in the screening version of the game.