This page is updated every other Friday to provide answers to frequently asked questions regarding the use of NCHAT data. If the information on this page does not meet your needs, please email [email protected] and we will be happy to answer your questions!
How do I access NCHAT data?
Both public and restricted data can be accessed at ICPSR.
You must apply to use the restricted access data. This can be started by clicking the “Access Restricted Data” button at the ICPSR link above. For more information, please watch our webinar on accessing the data.
How do I merge the different data sets?
Currently, the main respondent survey file (public and restricted), partner survey file (restricted), main respondent time diary (public and restricted), and partner time diary (restricted) are available.
To merge the data sets, use MATCH_ID. We suggest using the main respondent survey data set as the “master file” and to merge the other data sets to this file. Data may need to be reshaped (e.g., long to wide) before merging so that the MATCH_ID uniquely identifies the data by being on one line (instead of multiple lines like in the time diary data).
An example using the Partner data:
You can use the MATCH_ID variable from the main respondent survey data and the partner survey data to merge partner responses to the main respondent data. The main respondent and their partner should have the same MATCH_ID. The variable names in the main respondent data and the partner data are the exact same. To handle this, you must first rename the partner variables before merging, except the MATCH_ID, with something like, "variable_PARTNER." Once this is done, you can merge the partner data to the main respondent dataset. The data should be in wide format after merging so that both partners are on the same line. And then if necessary, you can reshape the data from wide to long (for multilevel analyses for example).
How do I know that I’ve merged the data correctly?
Double check your sample sizes:
- There are 3,642 main respondents.
- There are 1,515 partners who participated in the main survey.
- There are 1,932 main respondents who participated in the time diary.
- There are 826 partners who participated in the time diary.
- 665 MATCH_IDs have complete data from all four data sets.
How do I weight the data in Stata?
Use the following code for weighting the main respondent data:
svyset Weight_mainrespondent [pweight=Weight_mainrespondent]
Use the following code to weight the partner data and if using dyadic data:
svyset Weight_couple [pweight=Weight_couple]
Can I use IPUMS Contextual Determinants of Health (CDOH) with the NCHAT data?
Yes! We are in the process of depositing the data with ICPSR. NCHAT users will be able to merge IPUMS CDOH data with the NCHAT data in DSDR's Virtual data enclave. When completing your application for restricted data, users must request the specific measures from IPUMS CDOH in their application, when they become available. Visit cdoh.ipums.org for more information.