The National Academies Press

Currently Skimming:

Appendix F: Additional Information About the Panel's Analyses
Pages 316-340

The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.

From page 316... ... The third part describes the panel's exploration of the use of global regression models for predicting differences between ACS and CCD estimates for the blended reimbursement rate (BRR) using a variety of covariates from the CCD. Read the entire page →
From page 317... ... These tables display the same patterns as those observed in Chapter 4, where the administrative estimates pertain to the most recent year of the reference period for the ACS estimates. Namely, the ACS understates percentage free, percentage free or reduced price, and the BRR and overstates percentage reduced price. Read the entire page →
From page 318... ... SOURCE: Prepared by the panel. multiyear average estimates computed over the same time periods as the ACS estimates, as well as the differences between the ACS multiyear estimates and the CCD estimates for the most recent school year that overlaps the ACS reference period. Read the entire page →
From page 319... ... The differences shown in Table F-3 for the 5-year ACS estimates tend to be relatively small, but are largest (11 percent) for large very high FRPL districts (when compared with CCD estimates for 2009-2010) Read the entire page →
From page 320... ... These tables complement Tables 4-1, 4-2, and 4-3 in Chapter 4, which present results for the very high and high FRPL districts. Each table shows average differences for percentage free, percentage reduced price, percentage free or reduced price, and the BRR. Read the entire page →
From page 321... ... We write Adt = Cdt + bt + Bd + bdt + edt where edt is sampling error with known variance s2dt, and bt + Bd + bdt represents the difference between the CCD and ACS estimates after sampling error is removed. Because the CCD is treated as the gold standard in this discussion, we refer to bt + Bd + bdt as "bias," with bt representing a common time trend in the bias across districts, Bd representing a district-specific bias that is constant over time, and bdt representing the district- and time- specific deviation from the common time trend and constant district-specific bias. Read the entire page →
From page 322... ... and (2) 3-year averages of CCD estimates (so ACS estimates for 2005-2007 are compared with the average of CCD estimates TABLE F-5 Average Differences Between ACS 1-Year Estimates of Enrollment and CCD Estimates Estimand 2005 2006 2007 2008 2009 Very High FRPL Districts Difference from CCD 3,149 3,941 4,628 5,057 4,418 As percentage of CCD 7 9 10 11 12 High FRPL Districts Difference from CCD �184 �211 �297 �111 �131 As percentage of CCD �1 �1 �1 0 0 Low to Moderate FRPL Districts Difference from CCD �767 �1,295 �1,554 �1,650 �1,839 As percentage of CCD �3 �5 �6 �6 �7 NOTES: Calendar year ACS estimates are compared with the CCD estimates for the most recent school year that overlaps the calendar year of the ACS. Read the entire page →
From page 323... ... (4,042) Percentage free �4.7 �7.1 �6.1 �4.1 Percentage reduced price 2.3 2.1 1.8 2.4 Percentage free or reduced price �2.4 �5.0 �4.3 �1.7 BRR, $ �0.06 �0.12 �0.11 �0.05 NOTES: The ACS estimates for 2005-2009 are compared with CCD estimates for the most recent school year that overlaps the reference period of the ACS estimates, namely school year 2009-2010. Read the entire page →
From page 324... ... (1,213) Percentage free �3.2 �4.4 �6.2 Percentage reduced price 1.5 1.2 1.4 Percentage free or reduced price �1.7 �3.2 �4.8 BRR, $ �0.05 �0.08 �0.12 NOTES: The ACS estimates for a 3-year period are compared with CCD estimates for the most recent school year that overlaps the reference period of the ACS estimates. Read the entire page →
From page 325... ... The variance of gdt includes the design variance, but this is not used in building the model. Assumptions about the sampling error and its design variance are introduced below to extrapolate results from large districts to medium and small districts. Read the entire page →
From page 326... ... . 383 384 385 386 387 388 389 390 391 392 393 Method 2 01 Time 5 12345 Dimensions Covariance Parameters 7 Columns in X 18 Columns in Z Per Subject 2 Subjects 393 Max Obs per Subject 10 Number of Observations Number of Observations Read 3930 Number of Observations Used 3930 Number of Observations Not Used 0 Iteration Evaluations �2 Res Log Like Criterion 0 1 825.49213626 1 2 �3590.79275559 0.00012752 2 1 �3591.50965749 0.00000058 3 1 �3591.51280315 0.00000000 Convergence criteria met. Read the entire page →
From page 327... ... BOX F-3 SAS Output Proc Mixed Estimated R Matrix for Large Districts RowCol1Col2Col3Col4Col5Col6Col7Col8Col9Col10 10.01032 0.0017590.0003000.0000518.701E-6�0.00944�0.00161�0.00027�0.00005�7.96E-6 20.0017590.01032 0.0017590.0003000.000051�0.00161�0.00944�0.00161�0.00027�0.00005 30.0003000.0017590.01032 0.0017590.000300�0.00027�0.00161�0.00944�0.00161�0.00027 40.0000510.0003000.0017590.01032 0.001759�0.00005�0.00027�0.00161�0.00944�0.00161 58.701E-6 0.0000510.0003000.0017590.01032 �7.96E-6 �0.00005�0.00027�0.00161�0.00944 6�0.00944�0.00161�0.00027�0.00005�7.96E-6 0.02878 0.0049040.0008360.0001420.000024 7�0.00161�0.00944�0.00161�0.00027�0.00005 0.0049040.02878 0.0049040.0008360.000142 8�0.00027�0.00161�0.00944�0.00161�0.00027 0.0008360.0049040.02878 0.0049040.000836 9�0.00005�0.00027�0.00161�0.00944�0.00161 0.0001420.0008360.0049040.02878 0.004904 10�7.96E-6 �0.00005�0.00027�0.00161�0.00944 0.0000240.0001420.0008360.0049040.02878 327 Read the entire page →
From page 328... ... BOX F-4 SAS Output Proc Mixed Estimated R Correlation Matrix for Large Districts Row Col1 Col2 Col3 Col4 Col5 Col6 Col7 Col8 Col9 Col10 11.0000 0.1704 0.02904 0.0049480.000843 �0.5480 �0.09338�0.01591�0.00271�0.00046 20.1704 1.0000 0.1704 0.02904 0.004948 �0.09338�0.5480 �0.09338�0.01591�0.00271 3 0.02904 0.1704 1.0000 0.1704 0.02904 �0.01591 �0.09338 �0.5480 �0.09338 �0.01591 40.0049480.02904 0.1704 1.0000 0.1704 �0.00271�0.01591�0.09338�0.5480 �0.09338 50.0008430.0049480.02904 0.1704 1.0000 �0.00046�0.00271�0.01591�0.09338�0.5480 6 �0.5480 �0.09338�0.01591�0.00271�0.00046 1.0000 0.1704 0.02904 0.0049480.000843 7 �0.09338 �0.5480 �0.09338 �0.01591 �0.00271 0.1704 1.0000 0.1704 0.02904 0.004948 8 �0.01591 �0.09338 �0.5480 �0.09338 �0.01591 0.02904 0.1704 1.0000 0.1704 0.02904 9 �0.00271�0.01591�0.09338�0.5480 �0.09338 0.0049480.02904 0.1704 1.0000 0.1704 10 �0.00046 �0.00271 �0.01591 �0.09338 �0.5480 0.000843 0.004948 0.02904 0.1704 1.0000 SOURCE: Prepared by the panel. Read the entire page →
From page 329... ... District 0.01032 Covariance Parameter Estimates CovParmSubjectEstimate UN(2,1) District�0.00944 UN(2,2) Read the entire page →
From page 330... ... Time 4 3136 62.50<.0001 SOURCE: Prepared by the panel. BOX F-7 SAS Proc Mixed Output, Least Squares Means Least Squares Means Effect Method Time Estimate Error DF t Value Pr> \|t\| Method* Read the entire page →
From page 331... ... The ratio of the standard deviation to this value is a coefficient of variation. ACS = American Community Survey; ACS1 = ACS 1-year estimates; ACS3 = ACS 3-year estimates; ACS5 = ACS 5-year estimates; BRR = blended reimbursement rate; CCD = Common Core of Data; NA = not applicable. Read the entire page →
From page 332... ... Therefore, the CCD model row is exactly the same for medium and large districts. TABLE F-10 Model Versus Empirical Estimates for Variances of Year-to-Year Changes, Medium Districts Only Standard Deviation Standard Relative to Average Medium Districts Variance ($2) Read the entire page →
From page 333... ... The ratio of the standard deviation to this value is a coefficient of variation. ACS = American Community Survey; BRR = blended reimbursement rate; CCD = Common Core of Data; NA = not applicable. Read the entire page →
From page 334... ... = 9.84. If we plug this value into the linear relationship above and transform to the design standard deviation, we get 0.1153, which is very close to the average design standard deviation across districts and years, 0.1146. Read the entire page →
From page 335... ... FIGF-1.eps to be the observed median enrollment for medium districts and for small districts, respectively. The standard deviation (SD) Read the entire page →
From page 336... ... 400 300 CV2 200 100 0 0.000 0.002 0.004 0.006 0.008 0.010 1/enrollment FIGURE F-2 Squared coefficient of variation of year-to-year change in ACS 5-year estimate of BRR versus inverse of enrollment. NOTES: ACS = American Community Survey; BRR = blended reimbursement FIGF-2.eps rate; CV = coefficient of variation. Read the entire page →
From page 337... ... ACS = American Community Survey; NA = not applicable; RMSE = root mean squared error; SD = standard deviation. SOURCE: Prepared by the panel. Read the entire page →
From page 338... ... The analysis was limited to very high FRPL districts with both 5-year ACS estimates and CCD estimates for 2009-2010 in the panel's evaluation data set prog09.merged.fns. To eliminate outliers that could adversely impact regression results, we excluded any districts that had either a percentage certified for free meals of less than 10 percent or a percentage certified for free or reduced-price meals of less than 20 percent. Read the entire page →
From page 339... ... AIC = Akaike Information Criterion; ACS = American Community Survey; SE ( ACS5 BRR ) FOI = first-order interactions; NA = not applicable; RMSE = root mean squared error. Read the entire page →
From page 340... ... 22.C0910_ChartDistance_FRPL_Rel (index measuring distance to nearby charter-only districts, weighted by number of charter students certified for free or reduced-price meals relative to number in district) 23.C0910_Need (categorical variable for whether percentage of students certi fied for free or reduced-price meals is < 50, 50-74, or 75) Read the entire page →

From page 316...

... The third part describes the panel's exploration of the use of global regression models for predicting differences between ACS and CCD estimates for the blended reimbursement rate (BRR) using a variety of covariates from the CCD.

Appendix F: Additional Information About the Panel's Analyses Pages 316-340

Appendix F: Additional Information About the Panel's Analyses
Pages 316-340