Exhibit I.b.2. Scenario I, Missing Data Not Dependent on Pretest or Post-test: Data Missing for 40% of Students
Imact Estimate | Standard Error of Imact Est. | 90% Cl | ||||||||
---|---|---|---|---|---|---|---|---|---|---|
Data | Pre-test Data Available? | Estimate | True Imact | Bias | Bias Level | Standard Estimate | Unbiased Estimate | Bias | Bias Level | % of Samples in which 90% Cl Contains .20 |
No Missing Data | No | 0.203 | 0.200 | 0.003 | Low Bias | 0.085 | 0.088 | -0.002 | Low Bias | 0.892 |
Yes | 0.203 | 0.200 | 0.003 | Low Bias | 0.062 | 0.062 | 0.000 | Low Bias | 0.886 | |
A. Pre-Test (X) Data Missing | ||||||||||
Case Deletion | No | 0.202 | 0.200 | 0.002 | Low Bias | 0.081 | 0.080 | 0.001 | Low Bias | 0.901 |
Yes | ||||||||||
Dummy Variable Method | No | 0.203 | 0.200 | 0.003 | Low Bias | 0.073 | 0.075 | -0.001 | Low Bias | 0.893 |
Yes | ||||||||||
Mean Value Imputation | No | 0.203 | 0.200 | 0.003 | Low Bias | 0.073 | 0.079 | -0.006 | Low Bias | 0.859 |
Yes | ||||||||||
Single, Non-stochastic | No | 0.201 | 0.200 | 0.001 | Low Bias | 0.055 | 0.081 | -0.026 | Low Bias | 0.736 |
Regression Imputation | Yes | |||||||||
Single, Stochastic | No | 0.201 | 0.200 | 0.001 | Low Bias | 0.062 | 0.079 | -0.017 | Low Bias | 0.780 |
Regression Imputation | Yes | |||||||||
Multiple Stochastic | No | 0.205 | 0.200 | 0.005 | Low Bias | 0.069 | 0.074 | -0.005 | Low Bias | 0.871 |
Regrssion Imputation (n=5) | Yes | |||||||||
EM Algorithm with Multiple Imputation (n=5) | No | 0.200 | 0.200 | 0.000 | Low Bias | 0.072 | 0.076 | -0.004 | Low Bias | 0.875 |
Yes | ||||||||||
B. Post-Test (Y) Data Missing | ||||||||||
Case Deletion | No | 0.202 | 0.200 | 0.002 | Low Bias | 0.112 | 0.112 | 0.000 | Low Bias | 0.886 |
Yes | 0.202 | 0.200 | 0.002 | Low Bias | 0.081 | 0.080 | 0.001 | Low Bias | 0.901 | |
Mean Value Imputation | No | 0.202 | 0.200 | 0.002 | Low Bias | 0.065 | 0.112 | -0.047 | Low Bias | 0.679 |
Yes | 0.202 | 0.200 | 0.002 | Low Bias | 0.055 | 0.097 | -0.042 | Low Bias | 0.677 | |
Single, Non-stocastic | No | 0.202 | 0.200 | 0.002 | Low Bias | 0.065 | 0.112 | -0.047 | Low Bias | 0.679 |
Regression Imputation | Yes | 0.201 | 0.200 | 0.001 | Low Bias | 0.048 | 0.082 | -0.034 | Low Bias | 0.674 |
Single, Stochastic | No | 0.205 | 0.200 | 0.005 | Low Bias | 0.083 | 0.129 | -0.046 | Low Bias | 0.694 |
Regression Imputation | Yes | 0.203 | 0.200 | 0.003 | Low Bias | 0.060 | 0.095 | -0.034 | Low Bias | 0.702 |
Multiple, Stochastic | No | 0.195 | 0.200 | -0.005 | Low Bias | 0.107 | 0.114 | -0.007 | Low Bias | 0.863 |
Regression Imputation (n=5) | Yes | 0.196 | 0.200 | -0.004 | Low Bias | 0.077 | 0.084 | -0.007 | Low Bias | 0.853 |
EM Algorithm with Multiple Imputation (n=5) | No | 0.204 | 0.200 | 0.004 | Low Bias | 0.085 | 0.083 | 0.002 | Low Bias | 0.878 |
Yes | ||||||||||
Weighting - Simple | No | |||||||||
Yes | ||||||||||
Weighting - Sophisticated | No | |||||||||
Yes | 0.202 | 0.200 | 0.002 | Low Bias | 0.081 | 0.080 | 0.001 | Low Bias | 0.899 | |
Fully Specified Regression Models | No | 0.202 | 0.200 | 0.002 | Low Bias | 0.081 | 0.080 | 0.001 | Low Bias | 0.900 |
w/ Treatment-Covariate Interactions | Yes | |||||||||
Note: When pre-test scores are available, they are used as a covariate in the analysis model. In addition, we used pre-test scores to impute values and create weights. Bias estimates were computed as described in Chapter 4 and repeated at the beginning of this appendix. The level of the bias is characterized as "High Bias" or "Low Bias" based on the criteria established in Chapter 4. 90% CI refers to the 90-percent confidence interval around the impact estimate. For more details on the simulations, see Chapter 4 and Appendix C. |