This may or may not help but I find it to be very effective in discriminating angles--it is a strategy not listed in prep-books (some may be)
1) mentally mesure each angle relative to the X-axis. Having a reference point for comparison really helps.
2) for angles that are intentionally similar, look at the leg lengths--clearly, the choice with the longer legs will be the smaller angle for angles that seem the same.
3) for angles that are very very very close (which, on the actual DAT they are) look for parallel lines. Say you have choices a), b), c), and d). Most of the time it is easy to discriminate the exremes, but for those that are tough, stay cognizant of whether of not the different choices have legs that are parallel to each other. Say the top leg of the angle in choice a) is parallel with the top leg on the angle in choice c). Ask yourself if the other legs will intersect. By mentally noting what you would have to do to make the other legs parallel to each other (or not intersect) will tell you which angle is bigger or smaller relative to the other. The leg doing the intersecting is the leg that is on the smaller angle. The test makers do things like putting legs parallel to each other to give the test-taker another path with which to solve the problem. In summary, with angles that are really close in measurement, look for:
1) leg-length disparity-----the longer legs are on the shorter angle for angles that seem ridiculously close.
2) parallel legs-----what leg is doing the intersecting and ask yourself what you would have to do to make the other legs parallel as well.
Hope it helps a little.