Table 4.2. Design Effects for RD Designs, by Score Distribution, the Location of the Cutoff Score Value, and Other Key Parameters
Score Distribution |
The Treatment Group Sampling Rate (p) Which Equals the Percentage of the Score Distribution That Lies Above the Cutoff Score (q) |
|||
---|---|---|---|---|
0.50 | .033 or 0.67 | 0.25 or 0.75 | 0.10 or 0.90 | |
Normal | 2.75 | 2.46 | 2.17 | 1.52 |
Uniform | 4.00 | 2.97 | 2.29 | 1.35 |
Truncated Normal | ||||
Values of k1, k2a | ||||
-1.0, 1.0 | 3.65 | 2.91 | 2.30 | 1.41 |
-0.4, 0.4 | 3.94 | 2.99 | 2.29 | 1.38 |
-0.2, 0.2 | 3.98 | 3.00 | 2.29 | 1.37 |
0.0, 2.0 | 3.14 | 3.46 | 3.00 | 1.72 |
0.6, 1.4 | 3.73 | 3.45 | 2.70 | 1.50 |
0.8, 1.2 | 3.92 | 3.26 | 2.49 | 1.43 |
Bimodal: Equal Mixture of Two Normal Distributions |
||||
Value of lb | ||||
2.0 | 5.37 | 2.84 | 2.09 | 1.35 |
1.4 | 3.75 | 2.79 | 2.20 | 1.42 |
0.8 | 2.94 | 2.57 | 2.20 | 1.50 |
0.4 | 2.77 | 2.48 | 2.17 | 1.52 |
Note: See the text for formulas and other assumptions underlying the calculations. a The parameters k1 and k2 are the number of standard deviations from the mean of the full normal distribution that the left and right truncation points fall. b The parameter l is the number of standard deviations to the right (left) of the mean of the overall bimodal distribution where the first (second) component normal distribution is centered. The calculations assume that each normal distribution is weighted equally to create the bimodal mixture distribution (that is, the parameter w=0.5). |