Practical Considerations in Raking Survey Data

More documents

Recommendations

Info

Table 3. Raking Using Marginal Percentage Controls and No “Input” Weight (first and last iteration shown). The FREQ Procedure Cumulative Cumulative VAR1 Frequency Percent Frequency Percent --------------------------------------------------------- 1 3 27.27 3 27.27 2 5 45.45 8 72.73 3 3 27.27 11 100.00 Cumulative Cumulative VAR2 Frequency Percent Frequency Percent --------------------------------------------------------- 1 5 45.45 5 45.45 2 6 54.55 11 100.00 Raking VAR1, iteration - 1 Marginal Marginal Calculated Control Calculated Control Difference VAR1 margin Total Difference % % in % 1 3 20 17 27.273 20.000 7.273 2 5 35 30 45.455 35.000 10.455 3 3 45 42 27.273 45.000 -17.727 ========== ======== ========== ======== 11 100 100.00 100.00 Raking VAR2, iteration - 1 Marginal Marginal Calculated Control Calculated Control Difference VAR2 margin Total Difference % % in % 1 42.667 60 17.3333 42.667 60.000 -17.333 2 57.333 40 -17.3333 57.333 40.000 17.333 ========== ======== ========== ======== 100.000 100 100.00 100.00 Raking VAR1, iteration - 5 Marginal Marginal Calculated Control Calculated Control Difference VAR1 margin Total Difference % % in % 1 20.000 20 0.000256716 20.000 20.000 -0.000 2 35.001 35 -.000834329 35.001 35.000 0.001 3 44.999 45 0.000577612 44.999 45.000 -0.001 ========== ======== ========== ======== 100.000 100 100.00 100.00 Raking VAR2, iteration - 5 Marginal Marginal Calculated Control Calculated Control Difference VAR2 margin Total Difference % % in % 1 60.000 60 0.000205597 60.000 60.000 -0.000 2 40.000 40 -.000205597 40.000 40.000 0.000 ========== ======== ========== ======== 100.000 100 100.00 100.00 **** Program terminated at iteration 5 because all Calculated Percents differ from Marginal Percents by less than 0.001 36
Table 4: Example of Weight Trimming During Raking OBSERVATIONS IN ORIGINAL DATASET TO BE TRUNCATED CUTOFF: MEDIAN+6*IQR CONDITION: MEDIAN+6*IQR +1 weight_to_ id truncate mean median IQR cutoff condition 715 477.576 144.250 132.491 51.0592 438.847 439.847 651 509.018 144.250 132.491 51.0592 438.847 439.847 1085 690.762 144.250 132.491 51.0592 438.847 439.847 770 515.720 144.250 132.491 51.0592 438.847 439.847 OBSERVATIONS TO BE TRUNCATED AFTER ITERATION = 1 CUTOFF: MEDIAN+6*IQR CONDITION: MEDIAN+6*IQR + 1 truncated_ id weight mean median IQR cutoff condition 1085 451.059 144.250 133.108 51.7302 443.490 444.490 OBSERVATIONS TO BE TRUNCATED AFTER ITERATION = 2 CUTOFF: MEDIAN+6*IQR CONDITION: MEDIAN+6*IQR + 1 THERE ARE NO WEIGHTS TO TRUNCATE RERAKING-TRUNCATION PROCESS CONVERGED IN 2 ITERATIONS WITH CONDITION MEDIAN+6*IQR+1. 37
Page 1 and 2: Practical Considerations in Raking
Page 3 and 4: 1. Introduction A survey sample may
Page 5 and 6: 2. Basic Algorithm The procedure kn
Page 7 and 8: The adjustment factors, (0) Tj+ / m
Page 9 and 10: The authors’ experience indicates
Page 11 and 12: Table 2 illustrates the use of the
Page 13 and 14: Control totals usually do not come
Page 15 and 16: 7. Examining and Diagnosing Slow Co
Page 17 and 18: categories. Taking the meaning of v
Page 19 and 20: median in order to achieve converge
Page 21 and 22: 12. Raking Surveys that Screen for
Page 23 and 24: trimming high weight values one gen
Page 25 and 26: References Bishop YMM, Fienberg SE,
Page 27 and 28: Table 1. A 2 x 2 Table for Which Ra
Page 29 and 30: Raking AREA - 1 VARIABLE2, iteratio
Page 31 and 32: Figure 1. Convergence of a Raking P
Page 33 and 34: Figure 3. Convergence of Variables
Page 35: Figure 5. Prediction of the Number

Practical Considerations in Raking Survey Data

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?