Common use of Variance Estimation Clause in Contracts

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for MEPS estimates, analysts need to take into account the complex sample design of MEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these methodologies. Replicate weights have not been developed for the MEPS data. Instead, the variables needed to calculate appropriate standard errors based on the ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the use of the ▇▇▇▇▇▇-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR and VARPSU on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned computer software packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. Specifically, 125 variance estimation strata were created, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 for the 2008 PIT file. Beginning with the 2002 PIT data base 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS estimatessurvey data, analysts need one needs to take into account the complex sample design of MEPS for both person-level person and family- family level analyses. Several methodologies have been developed for estimating standard errors for surveys with Various approaches can be used to develop such estimates of variance including a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these ▇ series method for variance estimation or various replication methodologies. Replicate weights have not been developed for the MEPS data. Instead, We will describe the variables needed to calculate appropriate standard errors based on the implement a ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files▇ series estimation approach. Software packages that permit the use of the Using a ▇▇▇▇▇▇-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method▇ Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR VARST13 and VARPSU PSU13 on this MEPS data file (updated versions of corresponding variables provided in previously released MEPS public use files) serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned a computer software packages will package such as SUDAAN should provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. Specifically, 125 variance estimation strata were created, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 for the 2008 PIT file. Beginning with the 2002 PIT data base 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS estimatessurvey data, analysts need one needs to take into account the complex sample design of MEPS for both person-level and family- family-level analyses. Several methodologies have been developed for estimating standard errors for surveys with Various approaches can be used to develop such estimates of variance including a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these ▇ Series method for variance estimation or various replication methodologies. Replicate weights have not been developed for the MEPS data. Instead, We will describe the variables needed to calculate appropriate standard errors based on the implement a ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files▇ Series estimation approach. Software packages that permit the use of the Using a ▇▇▇▇▇▇-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method▇ Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR and VARPSU on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned a computer software packages will package such as SUDAAN should provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. InitiallyIn the past, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were have been developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old onePUFs. Thus, in order to make data from future years can be pooled and the pooling of data across variance strata and PSU variables provided can be used without modification for variance estimation purposes for estimates covering multiple years of MEPS more straightforward, the numbering system for the variance strata has changeddata. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. Specifically, 125 There are 203 variance estimation strata were createdstrata, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 for the 2008 PIT file. Beginning with the 2002 PIT data base 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for MEPS estimates, analysts need to take into account the complex sample design of MEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these methodologies. Replicate weights have not been developed for the MEPS data. Instead, the variables needed to calculate appropriate standard errors based on the ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the use of the ▇▇▇▇▇▇-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR and VARPSU on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned computer software packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. InitiallyPrior to 2002, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the roundsyear. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The current set of variance strata and PSUs were re-established for the 2008 PIT data, and have been carried over to the subsequent PIT files including the PIT 2014 dataset. Specifically, 125 165 variance estimation strata were created, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 1001- 1165 for the 2008 2014 PIT file. Beginning with the 2002 PIT data base database, 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases databases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRRHC- 036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for MEPS estimates, analysts need to take into account the complex sample design of MEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these methodologies. Replicate weights have not been developed for the MEPS data. Instead, the variables needed to calculate appropriate standard errors based on the ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the use of the ▇▇▇▇▇▇-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR and VARPSU on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned computer software packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. InitiallyPrior to 2002, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the roundsyear. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The current set of variance strata and PSUs were re-established for the 2008 PIT data, and have been carried over to the subsequent PIT files including the PIT 2013 dataset. Specifically, 125 165 variance estimation strata were created, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 1001- 1165 for the 2008 2013 PIT file. Beginning with the 2002 PIT data base database, 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases databases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRRHC- 036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for MEPS estimates, analysts need to take into account the complex sample design of MEPS for both person-level and family- family-level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these methodologies. Replicate weights have not been developed for the MEPS data. Instead, the variables needed to calculate appropriate standard errors based on the ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the use of the ▇▇▇▇▇▇-series - series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR and VARPSU on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned computer software packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. InitiallyIn the past, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were have been developed to be compatible with all future PUF until PUFs, subject to changes in the NHIS sample design changed. As discussed, this (there has been a design change took place for the 2006 NHIS which will affect MEPS for the first time in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one). Thus, in order to make the when pooling of data across years (2002 and forward), the variance strata and PSU variables provided can be used without modification for variance estimation purposes for estimates covering multiple years of MEPS more straightforward, the numbering system for the variance strata has changeddata. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. Specifically, 125 There are 203 variance estimation strata were createdstrata, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 for the 2008 PIT file. Beginning with the 2002 PIT data base 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for MEPS estimates, analysts need to take into account the complex sample design of MEPS for both person-level and family- family-level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these methodologies. Replicate weights have not been developed for the MEPS data. Instead, the variables needed to calculate appropriate standard errors based on the ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the use of the ▇▇▇▇▇▇-series - series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR and VARPSU on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned computer software packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussedThus, this change took place in 2006, effectively changing when pooling data across years 2002 through the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on component of 2007 files, the old one. Thusvariance strata and PSU variables provided can be used without modification (except for the renumbering necessary of the 2007 variance strata and PSU values, in order to make the pooling of data across as discussed below) for variance estimation purposes for estimates covering multiple years of MEPS more straightforwarddata. There were 203 variance estimation strata, the numbering system for the each stratum with either two or three variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit valuesestimation PSUs. For the 2007 PIT data a new set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. Specifically, 125 variance estimation strata were created, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 With MEPS Panels 11 and 12 based on two NHIS independent samples, 328 variance strata are available for variance estimation purposes for the 2008 2007 PIT filedata. Beginning with the 2002 PIT data base 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases under the old design. These were Those numbered 1-203125 are associated with Panel 12 while those numbered 126 through 328 are associated with Panel 11. This numbering system was used to make analyses across multiple years of MEPS data straightforward. Those using data from Panel 12 and MEPS panels in the future can use the variance strata numbering from Panel 12 onward. For those using data analyses where data pooling across calendar years is limited to 2002 from Panel 11 and laterearlier, the numbering of from 1 to 203 can be used for years 2006 and earlier. To include Panel 11 data from 2007 MEPS databases, subtracting 125 from the values identifying the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of appearing on the 2007 databases will produce the variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS that corresponded to those from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRRearlier years.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for MEPS estimates, analysts need to take into account the complex sample design of MEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these methodologies. Replicate weights have not been developed for the MEPS data. Instead, the variables needed to calculate appropriate standard errors based on the ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the use of the ▇▇▇▇▇▇-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR and VARPSU on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned computer software packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The current set of variance strata and PSUs were re-established for the 2008 PIT data, and have been carried over to the subsequent PIT files including the PIT 2011 dataset. Specifically, 125 165 variance estimation strata were created, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 1001- 1165 for the 2008 2011 PIT file. Beginning with the 2002 PIT data base database 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases databases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRRHC- 036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS estimatessurvey data, analysts need one needs to take into account the complex sample design of MEPS for both person-level and family- family-level analyses. Several methodologies have been developed for estimating standard errors for surveys with Various approaches can be used to develop such estimates of variance including a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these ▇ Series method for variance estimation or various replication methodologies. Replicate weights have not been developed for the MEPS data. Instead, We will describe the variables needed to calculate appropriate standard errors based on the implement a ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files▇ Series estimation approach. Software packages that permit the use of the Using a ▇▇▇▇▇▇-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method▇ Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR VARST13 and VARPSU PSU13 on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned a computer software packages will package such as SUDAAN should provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. InitiallyIn the past, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were have been developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old onePUFs. Thus, in order to make the pooling of data across multiple from future years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new set of variance strata can be pooled and PSUs were developed for use with data collected under the new NHIS sample design. Specifically, 125 variance estimation strata were created, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 for the 2008 PIT file. Beginning with the 2002 PIT data base 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables provided can be used without modification for all respondents participating in MEPS, along with the standard MEPS person ID variables variance estimation purposes for linking to other MEPS files. This one file contains records for each person who is on any estimates covering multiple years of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRRdata.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for MEPS estimates, analysts need to take into account the complex sample design of MEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these methodologies. Replicate weights have not been developed for the MEPS data. Instead, the variables needed to calculate appropriate standard errors based on the ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the use of the ▇▇▇▇▇▇-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR and VARPSU on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned computer software packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. InitiallyPrior to 2002, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the roundsyear. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF PUFs until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The current set of variance strata and PSUs were re-established for the 2008 PIT data, and have been carried over to the subsequent PIT files including the PIT 2015 dataset. Specifically, 125 165 variance estimation strata were created, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 1001- 1165 for the 2008 2015 PIT file. Beginning with the 2002 PIT data base database, 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases databases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRRHC- 036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for MEPS estimates, analysts need to take into account the complex sample design of MEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these methodologies. Replicate weights have not been developed for the MEPS data. Instead, the variables needed to calculate appropriate standard errors based on the ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the use of the ▇▇▇▇▇▇-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR and VARPSU on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned computer software packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The current set of variance strata and PSUs were re-established for the 2008 PIT data, and have been carried over to the subsequent PIT files including the PIT 2012 dataset. Specifically, 125 165 variance estimation strata were created, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 1001- 1165 for the 2008 2012 PIT file. Beginning with the 2002 PIT data base database, 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases databases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRRHC- 036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for MEPS estimates, analysts need to take into account the complex sample design of MEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these methodologies. Replicate weights have not been developed for the MEPS data. Instead, the variables needed to calculate appropriate standard errors based on the ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the use of the ▇▇▇▇▇▇-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR and VARPSU on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned computer software packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The current set of variance strata and PSUs were re-established for the 2008 PIT data, and have been carried over to the subsequent PIT files including the PIT 2010 dataset. Specifically, 125 165 variance estimation strata were created, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 1001- 1165 for the 2008 2010 PIT file. Beginning with the 2002 PIT data base 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRRHC- 036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for estimates based on MEPS estimatessurvey data, analysts need one needs to take into account the complex sample design of MEPS for both person-level and family- family-level analyses. Several methodologies have been developed for estimating standard errors for surveys with Various approaches can be used to develop such estimates of variance including a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these ▇ Series method for variance estimation or various replication methodologies. Replicate weights have not been developed for the MEPS data. Instead, We will describe the variables needed to calculate appropriate standard errors based on the implement a ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files▇ Series estimation approach. Software packages that permit the use of the Using a ▇▇▇▇▇▇-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method▇ Series approach, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR and VARPSU on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned a computer software packages will package such as SUDAAN should provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. InitiallyIn the past, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2002 2003 Point-in-Time PUF, the variance strata and PSUs were have been developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old onePUFs. Thus, in order to make data from future years can be pooled and the pooling of data across variance strata and PSU variables provided can be used without modification for variance estimation purposes for estimates covering multiple years of MEPS more straightforward, the numbering system for the variance strata has changeddata. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. Specifically, 125 There are 203 variance estimation strata were createdstrata, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 for the 2008 PIT file. Beginning with the 2002 PIT data base 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRR.)

Variance Estimation. MEPS has a complex sample design. To obtain estimates of variability (such as the standard error of sample estimates or corresponding confidence intervals) for MEPS estimates, analysts need to take into account the complex sample design of MEPS for both person-level and family- level analyses. Several methodologies have been developed for estimating standard errors for surveys with a complex sample design, including the ▇▇▇▇▇▇-series linearization method, balanced repeated replication, and jackknife replication. Various software packages provide analysts with the capability of implementing these methodologies. Replicate weights have not been developed for the MEPS data. Instead, the variables needed to calculate appropriate standard errors based on the ▇▇▇▇▇▇-series linearization method are included on this point-in-time file as well as all other MEPS public use files. Software packages that permit the use of the ▇▇▇▇▇▇-series linearization method include SUDAAN, Stata, SAS (version 8.2 and higher), and SPSS (version 12.0 and higher). For complete information on the capabilities of each package, analysts should refer to the corresponding software user documentation. Using the ▇▇▇▇▇▇-series linearization method, variance estimation strata and the variance estimation PSUs within these strata must be specified. The variables VARSTR and VARPSU on this MEPS data file serve to identify the sampling strata and primary sampling units required by the variance estimation programs. Specifying a “with replacement” design in one of the previously mentioned computer software packages will provide estimated standard errors appropriate for assessing the variability of MEPS survey estimates. It should be noted that the number of degrees of freedom associated with estimates of variability indicated by such a package may not appropriately reflect the number available. For variables of interest distributed throughout the country (and thus the MEPS sample PSUs), one can generally expect to have at least 100 degrees of freedom associated with the estimated standard errors for national estimates based on this MEPS database. Initially, MEPS variance strata and PSUs were developed independently from year to year, and the last two characters of the strata and PSU variable names denoted the rounds. However, beginning with the 2002 Point-in-Time PUF, the variance strata and PSUs were developed to be compatible with all future PUF until the NHIS design changed. As discussed, this change took place in 2006, effectively changing the MEPS design beginning with calendar year 2007, where Panel 12 was based on the new NHIS design while Panel 11 was based on the old one. Thus, in order to make the pooling of data across multiple years of MEPS more straightforward, the numbering system for the variance strata has changed. Those strata associated with the new design will have four digit values while those associated with the old design will have three digit values. For the 2007 PIT data a new temporary set of variance strata and PSUs were developed for use with data collected under the new NHIS sample design. The current set of variance strata and PSUs were re-established for the 2008 PIT data, and were carried over to the PIT 2009 data. Specifically, 125 165 variance estimation strata were created, each stratum with either two or three variance estimation PSUs. These have been numbered 1001-1125 1165 for the 2008 2009 PIT file. Beginning with the 2002 PIT data base 203 variance strata were formed for use in developing variance estimates for all subsequent years and data bases under the old design. These were numbered 1-203. For data analyses where data pooling across calendar years is limited to 2002 and later, the numbering of the variance strata and variance PSUs now permits this with no further actions needed. If pooled analyses involve data in calendar years earlier than 2002, a pooled linkage file has been created to permit assignment of variance strata and PSU values for any person sampled under the old NHIS sample design (the one used for the NHIS from 1995-2005, and thus associated with MEPS samples for MEPS Panels 1-11). This person-level file contains variance stratum and PSU variables for all respondents participating in MEPS, along with the standard MEPS person ID variables for linking to other MEPS files. This one file contains records for each person who is on any of the MEPS full-year consolidated files. It is found on PUF Number HC-036. (A Balanced Repeated Replicate or BRR version of this file is also available. See PUF Number HC-036BRR.)