Linking Instructions. The following steps should be taken to create a file containing persons from any one or more of the twenty years of MEPS HC data.
1. Create a dataset for each year containing the person- and/or event-level records of all persons to be included in the analysis. Keep the unique person identifier (DUPERSID and PANEL), the person-level sampling weight, any classification variables (e.g., sex, race/ethnicity) and response variables (e.g., total expenditure amount, number of prescription drug purchases, etc) to be used in the data analysis.
2. Reconcile the discrepancies in variable names. For all years, most variable names on the annual public use files contain a 2-digit year suffix. For instance, in the 1997 consolidated person-level file (HC-020) the panel variable is called PANEL97, the total annual expenditure amount variable is called TOTEXP97 and the sampling weight variable is called WTDPER97. But in the 2003 dataset (HC-079) these same variables are named PANEL03, TOTEXP03 and PERWT03F, respectively, and in the 1996 dataset (HC-012) the total expenditure and sampling weight variables are named TOTEXP96 and WTDPER96, respectively, and the panel variable is missing (users should assign a value of 1 for each record in HC-012). As illustrated below, the variable names must be made consistent before pooling the data. Note: starting in 2005, the panel variable is called simply PANEL (no year suffix). 1996 (HC-012) 1997 (HC-020) 2003 (HC-079) Combined dataset DUPERSID calculated PANEL WTDPER96 SEX TOTEXP96 DUPERSID PANEL PERWT SEX TOTEXP DUPERSID PANEL97 WTDPER97 SEX TOTEXP97 DUPERSID PANEL PERWT SEX TOTEXP DUPERSID PANEL03 PERWT03F SEX TOTEXP03 DUPERSID PANEL PERWT SEX TOTEXP 3. Create a pooled analysis dataset by combining the individual-year datasets by row; that is, append the records from the 1996 dataset with those from the 1997 and 2003 datasets.
4. Attach the BRR replicate flags to the pooled analysis dataset by column; that is, merge the variables BRR1-BRR128 from this HC-036BRR file to the pooled analysis dataset by DUPERSID and PANEL keeping all records in the pooled analysis dataset and only those records in HC-036BRR dataset that match. Depending on the software being used to manage the datasets, the pooled analysis dataset may need to be sorted by DUPERSID and PANEL prior to merging.
5. To calculate a standard set of 128 BRR replicates, multiply each BRR replicate flag by 2 and by the sample weight (PERWT, if using the example above). That is, ...
Linking Instructions. The following steps should be taken to create a pooled analysis dataset.
1. Create a dataset for each year containing the person- and/or event-level records of all persons to be included in the analysis. Keep the unique person identifier (DUPERSID and PANEL), the person-level sampling weight, any classification variables (e.g., sex, race/ethnicity) and response variables (e.g., total expenditure amount, number of prescription drug purchases, etc.) to be used in the data analysis.
2. Reconcile the discrepancies in variable names. For all years, most variable names on the annual public use files contain a 2-digit year suffix. For instance, in the 1997 consolidated person-level file (HC-020) the panel variable is called PANEL97, the total annual expenditure amount variable is called TOTEXP97 and the sampling weight variable is called WTDPER97. But in the 2003 dataset (HC-079) these same variables are named PANEL03, TOTEXP03 and PERWT03F, respectively, and in the 1996 dataset (HC-012) the total expenditure and sampling weight variables are named TOTEXP96 and WTDPER96, respectively, and the panel variable is missing (users should assign a value of 1 for each record in HC-012). Starting in 2005, the panel variable is simply named PANEL (no year suffix). As illustrated below, the variable names must be made consistent before pooling the data.
3. Create a pooled analysis dataset by simply combining the individual-year datasets (e.g., the records from the 1996 and 1997 files). In other words, the number of records in the pooled file will equal the sum of the record counts for the individual annual files being pooled.
4. Attach the pooled variance structure to the pooled analysis dataset by merging the variables STRA9618 and PSU9618 from this HC-036 file to the pooled analysis dataset by DUPERSID and PANEL keeping all records in the pooled analysis dataset only. Depending on the software being used to manage the datasets, the pooled analysis dataset may need to be sorted by DUPERSID and PANEL prior to merging. This step will add two additional variables to the pooled file (STRA9618 and PSU9618) but have no impact on the number of records.
Linking Instructions. The following steps should be taken to create a pooled analysis dataset.
1. Create a dataset for each year containing the records for all persons to be included in the analysis. Keep the unique person identifier variables (DUPERSID and PANEL), the person-level sampling weight, any classification variables (e.g., sex, race/ethnicity) and response variables (e.g., total expenditure amount, number of prescription drug purchases, etc) to be used in the data analysis.
2. Reconcile the discrepancies in variable names. For all years, most variable names on the annual public use files contain a 2-digit year suffix. For instance, in the 1997 consolidated person-level file (HC-020) the panel variable is called PANEL97, the total annual expenditure amount variable is called TOTEXP97 and the sampling weight variable is called WTDPER97. However, in the 2003 dataset (HC-079) these same variables are named PANEL03, TOTEXP03 and PERWT03F, respectively, and in the 1996 dataset (HC-012) the total expenditure and sampling weight variables are named TOTEXP96 and WTDPER96, respectively, and the panel variable is missing (users should assign a value of 1 for each record in HC-012). Starting in 2005, the panel variable is simply named PANEL (no year suffix). As illustrated below, the variable names must be made consistent before pooling the data.
3. Create a pooled analysis dataset by simply combining the individual-year datasets (e.g., the records from the 1996 and 1997 files). In other words, the number of records in the pooled file will equal the sum of the record counts for the individual annual files being pooled.
4. Attach the pooled variance structure to the pooled analysis dataset by merging the variables STRA9607 and PSU9607 from this HC-036 file to the pooled analysis dataset by DUPERSID and PANEL keeping all records in the pooled analysis dataset only. Depending on the software being used to manage the datasets, the pooled analysis dataset may need to be sorted by DUPERSID and PANEL prior to merging. This step will add two additional variables to the pooled file (STRA9607 and PSU9607) but have no impact on the number of records.
Linking Instructions. The following steps should be taken to create a pooled file over the seven years. 1 In public use files for 2002 and beyond, the survey design variables are named VARSTR AND VARPSU.
1. Create a SAS file for each year containing persons desired from that year of data. Keep the person identifier DUPERSID and any other variables desired for the analysis.
2. Rename similar variables from different years to common names. As part of this process one may have to rename variables from different years so that they have a consistent name. Many variables have names for each year that are of the form nameyy where the name portion is the same, but yy represents the year of the file. To use as the same variable one must rename each year as the same variable. For instance, name96 and name97 might be given the title NAME.
3. Concatenate files together from the seven years.
4. Sort the file of combined years by DUPERSID.
5. Merge the combined file with the HC-036 data file by DUPERSID and keep records that are in both files. Retain all variables.
6. Produce desired estimates as desired variables STRA9603 and PSU9603 in the same manner as the variables VARSTRyy and VARPSUyy are used in variance calculations for individual years.
