Skip to main content

TUTORIALS

  1. Go to the Interactive Graphs page under the Harmonized Data tab. From the menu, select your topic of interest, any or all countries or years, and up to two separate subpopulations. Please note that if you choose an individual-level topic, then you can select from individual-level subpopulations, and if you select a household-level topic, then you can select from household-level subpopulations. Once you have made your selections, click "Plot".

  2. Once the graph has appeared, you can click a subpopulation in the legend to hide or show the estimates for that subpopulation, drag over an area to zoom in, and hover over a line or bar to see the estimate (& confidence interval) for the subpopulation. You can print or download the graph by clicking on the three horizontal lines in the top right. All applicable notes about the data, including the Harmonized dataset and weights used, appear in a gray box below the graph.

  3. You can view the data in a table format by clicking on the "Table" tab on the top middle. You can download the table by clicking the export buttons on the top right beneath the tabs.

  4. You can also view the data in a map format by clicking on the "Map" tab on the top middle. You can click to select different subpopulations, you can zoom in or out, and hover over the country to view an estimate. You can print or download the map by clicking on the three horizontal lines in the top right. Please note that the map will show estimates for all countries with available data for that particular year.

  1. Go to the Publications page under the Additional Resources tab. Here, you can filter your search results by keyword, research topics (e.g., Demographics, Cognition, Health), specific studies (e.g., HRS, MHAS, ELSA), and years (ranging from 1993 to 2023).

  2. Once you have selected your search criteria, click the "Search” button to show the results of your search. You can use the "Previous” and "Next” buttons at the bottom of the screen to navigate through the list of results.

  1. Go to the Search page under the Health & Retirement Studies tab. Here, you can filter your search results by keyword, specific studies (e.g., HRS, MHAS, ELSA), survey years (ranging from 1992 to 2019), survey types (e.g., Core Interview, End of Life Interview), and survey modules (e.g., Demographics, Cognition, Health).

  2. Once you have selected your search criteria, click the "Search" button to show the results of your search. You can use the "Previous" and "Next" buttons at the bottom of the screen to navigate through the list of results.

There are two ways to find relevant Harmonized variables:

1a. Go to the Search page under the Harmonized Data tab. Here, you can filter your search results by keyword, specific studies (e.g., HRS, MHAS, ELSA), and survey years (ranging from 1992 to 2020).

1b. Once you have selected your search criteria, click the "Search” button to show the results of your search. You can use the "Previous” and "Next" buttons at the bottom of the screen to navigate through the list of results.

2a. Go to the Variable Concordance page under the Harmonized Data tab. Here, you can filter your search results by topic (e.g., Demographics, Cognition, Health), subtopics, variable names, and specific studies (e.g., HRS, MHAS, ELSA).

2b. Once you have selected your search criteria, a list of relevant harmonized variables with their associated descriptions will populate the bottom half of the screen. Any differences in how these harmonized variables are built across studies (i.e., scaling, number of items) are discussed in the notes section at the bottom of the screen.

We provide Stata Creation Code for Harmonized SHARE, and Harmonized KLoSA.

  1. Go to the Get Data page under the Health & Retirement Studies tab. Here, you can follow the link and the data access instructions to download the original survey data. The survey website requires users to register, and once registered, you can access the survey data.

  2. Unzip and store all data files on your computer.

    • For SHARE: From SHARE Research Data Center, download the 8.0.0 release Stata data for Waves 1, 2, 3, 4, 5, 6, 7, and 8. Download the Longitudinal Weights Release 8.0.0 for Stata and download the All Waves Coverscreen Release 8.0.0 for Stata.

    • For KLoSA: From KEIS, download the combined KLoSA 1-8th wave (Stata) file, download the combined KLoSA 1-7th wave imputation file, and download the individual Stata exit data files for waves 2-8. Unzip the KLoSA 1-8th wave (Stata) file and the KLoSA 1-7th wave imputation file and place all files in the same folder as the other KLoSA data files.

  3. Go to the Get Data page under the Harmonized Data tab. Here, you can download the Stata Creation Code which is a program file in Stata. Edit the lines in Stata Creation Code where it lists the folder location of the survey data and the folder location where you would like to save the Harmonized dataset.

Here is an example for changing the KLoSA creation code, change the following lines:
local stataraw "||KLoSA_data_folder||"
local output "||save to folder||"
to:
local stataraw "C:/YourRawDataFolderPath/KLoSA_STATA_2023_e"
local output "C:/YourSaveDataFolderPath"

Next, you can run the edited Stata Creation Code in Stata and be aware that the Harmonized dataset would possibly take several hours to generate.

Currently, all Harmonized datasets are available in Stata. The Harmonized HRS, Harmonized MHAS, Harmonized CRELES, KLoSA data, Harmonized TILDA, and Harmonized LASI are also available in SAS. The Harmonized HRS, Harmonized MHAS, Harmonized ELSA and ELSA data, SHARE data, Harmonized CRELES, KLoSA data, Harmonized TILDA, and Harmonized LASI are also available in SPSS. The Harmonized ELSA and ELSA data are also available in tab-delimited format. SHARE data and the Harmonized LASI are also available in R. Limited KLoSA data and the Harmonized JSTAR are also available in a Microsoft Excel format.

Any Harmonized dataset which is generated using Stata Creation code or is downloaded in Stata format can be converted for use in other statistical packages using a program such as Stat/Transfer. If you do not have access to Stat/Transfer, you may be able to read the .dta dataset into your stat package using an "import" or "get" function. When reading a .dta dataset into another package, it is best to first save the dataset in a Stata version 13 format using the command saveold.

An example of the saveold command in Stata:

if using Stata 14 or later:
saveold "Your_data_path/Data", version(13)
if using Stata 13 or earlier:
save "Your_data_path/Data"

Back to Top