Person-level Weight Sample Clauses
The Person-level Weight clause defines how individual data points or responses are assigned specific weights in a dataset or analysis. In practice, this means that each person's input may be given more or less influence based on predetermined criteria, such as demographic representation or sampling requirements. This clause ensures that the results of analyses or surveys accurately reflect the intended population, correcting for any imbalances or biases in the sample.
Person-level Weight. The person-level weight variable (WGTSP13) was constructed as a composite of separate panel specific weights. A positive person-level weight was assigned to all key members of the U.S. civilian, non-institutionalized population for whom MEPS data were collected, representing the corresponding U.S. population in early 2011. For the Panel 16 Round 1 participants, this weight reflects the original household probability of selection for the NHIS, a factor representing the proportion of the 16 NHIS panel-quarter combinations eligible for MEPS, the oversampling of the subgroups described earlier, ratio-adjustment to NHIS national population estimates at the household level, adjustment for non-participation in MEPS at the household or dwelling unit level, and poststratification to U.S. civilian noninstitutionalized population estimates obtained from March 2011 CPS data at the family and person levels. For both Panels 15 and 16 separately person-level poststratification reflected population distributions across census region, MSA status, race/ethnicity (Hispanic, black/non-Hispanic, Asian, other), sex, and age. Finally, a composite weight was assigned to each responding person and a final poststratification was undertaken across the variables of census region, MSA status, race/ethnicity, sex, and age. Table 3-2 shows the number of persons with person weights for each of the two panels separately, as well as the combined total and the total population estimate represented by the weighted total for all persons with person-level weights. In terms of numbers of persons, there are 15,043 for Panel 15, Round 3 and 20,956 for Panel 16, Round 1. Thus, in total, there are 35,999 sample persons in the file with positive person-level weights (WGTSP13>0). The corresponding estimate for the civilian, noninstitutionalized population based on summing the weights found in the variable WGTSP13 for these 35,999 persons is 305,191,076. Table 3-2. Persons with a person weight for the 2011 Point-in-Time file Panel 15 Panel 16 Combined Population estimate (weighted total of combined sample) Number 15,043 20,956 35,999 305,191,076 Beginning with the 2001 Full Year data, MEPS transitioned to 2000 census-based population estimates for poststratification and raking. In addition, MEPS population estimates underwent some “discontinuities” due to adjustments made to the CPS estimates as of 2003 (CPS being the source of the control figures used for raking and poststratification in MEPS). Those...
Person-level Weight. The person-level weight variable (WGTSP13) was constructed as a composite of separate panel specific weights. A positive person-level weight was assigned to all key members of the U.S. civilian, non-institutionalized population for whom MEPS data were collected, representing the corresponding U.S. population in early 2003. For the Panel 8 Round 1 participants, this weight reflects the original household probability of selection for the NHIS, ratio-adjustment to NHIS national population estimates at the household level, adjustment for non-participation in MEPS at the household level, and poststratification to figures obtained from March 2003 Current Population Survey (CPS) data at the family and person levels. The person-level poststratification reflected population distributions across census region, MSA status, race/ethnicity (Hispanic, black/non-Hispanic, Asian, other), sex, and age. Table 3.2 shows the number of persons with person weights for each of the two panels separately, as well as the combined total and the total population estimate represented by the weighted total for all persons with person-level weights. In terms of numbers of persons, there are 16,413 for Panel 7, Round 3 and 18,028 for Panel 8, Round 1. Thus, in total, there are 34,441 sample persons in the file with positive person-level weights (WGTSP13>0). The corresponding estimate for the civilian, noninstitutionalized population based on summing the weights found in the variable WGTSP13 for these 34,441 persons is 285,077,107. Table 3.2. Persons with a person weight for the 2003 Point-in-Time file Panel 7 Panel 8 Combined Population estimate (weighted total of combined sample) Number 16,413 18,028 34,441 285,077,107 Some may find it helpful to note that MEPS population estimates have undergone some “discontinuities” recently, due to similar “jumps” in CPS estimates, the source of the control figures used for raking and poststratification in MEPS. Those who wish to learn about these recent changes in CPS population estimates may consult the report "Revisions to the Current Population Survey Effective in January 2003" from the January 2003 issue of the monthly Labor Review. This report was authored by ▇▇▇▇ ▇▇▇▇▇▇, ▇▇▇▇▇ ▇. ▇▇▇, ▇▇▇▇▇▇▇ ▇▇▇▇▇▇, ▇▇ ▇▇▇▇▇▇▇, and ▇▇▇▇ ▇▇▇▇▇▇▇, all at the Bureau of Labor Statistics. Differences in the way racial categories are defined are also noted in this report.
Person-level Weight. The person-level weight variable (WGTSP13) was constructed as a composite of separate panel specific weights. A positive person-level weight was assigned to all key members of the U.S. civilian, non-institutionalized population for whom MEPS data were collected, representing the corresponding U.S. population in early 2004. For the Panel 9 Round 1 participants, this weight reflects the original household probability of selection for the NHIS, ratio-adjustment to NHIS national population estimates at the household level, adjustment for non-participation in MEPS at the household level, and poststratification to figures obtained from March 2004 Current Population Survey (CPS) data at the family and person levels. The person-level poststratification reflected population distributions across census region, MSA status, race/ethnicity (Hispanic, black/non-Hispanic, Asian, other), sex, and age. Table 3.2 shows the number of persons with person weights for each of the two panels separately, as well as the combined total and the total population estimate represented by the weighted total for all persons with person-level weights. In terms of numbers of persons, there are 16,530 for Panel 8, Round 3 and 18,267 for Panel 9, Round 1. Thus, in total, there are 34,797 sample persons in the file with positive person-level weights (WGTSP13>0). The corresponding estimate for the civilian, noninstitutionalized population based on summing the weights found in the variable WGTSP13 for these 34,797 persons is 287,368,408.
Person-level Weight. The person level weight variable (WGTSP13) was constructed as a composite of separate panel specific weights. A positive person level weight was assigned to all key members of the U.S. civilian non-institutionalized population for whom MEPS data were collected, representing the corresponding U.S. population in early 2001. For the Panel 6, Round 1 participants this weight reflects the original household probability of selection for the NHIS, ratio-adjustment to NHIS national population estimates at the household level, adjustment for non-participation in MEPS at the household level, and poststratification to figures obtained from March 2001 Current Population Survey (CPS) data at the family and person levels. The person level poststratification reflected population distributions across census region, MSA status, race/ethnicity (Hispanic, black/non-Hispanic, other), sex, and age. Overall, the weighted population estimate based on WGTSP13 for the civilian noninstitutionalized population is 275,740,015. Estimates can be made for this population based on the 34,173 sample persons in the file with positive weights (WGTSP13>0).