Regression Discontinuity Design

Sharp regression discontinuity design can be characterized as a method that is analagous to:

A. localized randomization
B. localized matching
C. localized encouragement design
D. none of the above.

Why are the other answers not as good as the preferred answer?

Sharp regression discontinuity design can be characterized as a method that is analagous to:

A. localized randomization
B. localized matching
C. localized encouragement design
D. none of the above.

Is there another way to characterize regression discontinuity design besides local randomization?

The next figure is from Trochim, W. (1990). "Regression-discontinuity design in health evaluation." In L. Sechrest, E. Perrin and J. Bunker (Eds.). Research Methodology: Strengthening Causal Interpretations of Nonexperimental Data. U.S. Dept. of HHS, Agency for Health Care Policy and Research, Washington, D.C.

What does the figure show: Sharp or fuzzy regression discontinuity design? Where is the discontinuity?

Why would the treatment group be below the cutoff?

Does the treatment effect appear to be positive, negative, or approximately zero?