Here's how the relatively small experiment was assembled (and here's the abstract): 47 female college student participants were told that several male students had already commented on their Facebook profiles. The female participants were divided into three groups and were A) told that the guys said they liked their profile "a lot" B) told that the guys liked it an "average amount" or C) given an "uncertain" scenario where they didn't know what the males thought. The females, in turn, then had rank their level of attraction for the males based on their "emotional response" to the information.
Naturally, the study found that participants in the "uncertain" scenario "reported thinking about the men the most." This group found the men who expressed no opinion about them to be the most attractive. So, yes, according to this small sample study this would give credence to the "playing hard to get" romance strategy. But, as Intelligent Life notes, even as the study confirms "conventional wisdom" it also reveals something perhaps a bit more unexpected: "uncertainty is only effective [to heighten romance] for as long as it's distracting."
Put bluntly, as More Intelligent Life writes: "It's a crafty opening gambit, but hardly an essential one."