I am using Wave IV of Add Health, and I am trying to get one partner among all of the relationships reported. The data is in the long format, and I am trying to get it to wide, but with one partner (so not just converting from long to wide because I am deliberately choosing a specific partner). Generally, I am trying to pick the more serious relationship (if they say they are married but also dating someone else, I want to drop the dating relationship in favor of the marriage).
Specifically, I am using slide 33 of this presentation to guide my process https://www.cpc.unc.edu/projects/add...rence%20slides. I coded it so I am only using people that are in a current relationship. Then I coded if people reported that they are married, cohabiting or dating (also if they are with a partner that is pregnant I included them in dating). I also coded how many times they were married/cohabiting with this partner.
Where I am stuck is if they report more than one dating/cohabiting relationship, how to choose the relationship with the longest duration (which I coded in years). I used the command gsort by the participant ID, partner ID, relationship status, and relationship length. I am looking for a way to drop the partners with the shortest duration.
Related Posts with Coding for one relationship (Add Health WIV). Dropping people with the shortest relationship duration.
Repeated time values - Panel DataHi, I have election data for a country. I am trying to set a panel but as each place has multiple c…
Interpreting a 3-category dummy variables and interaction termsDear All, I run the following model: Suppose I have a dummy variable: small, medium, large; and the…
"rangestat" with "excludeself" and "interval(t -1 0)"Dear Statalisters, I have a question about the results after the rangestat command. Suppose I inpu…
Finding the mode value in aggregated spatial unitsHello, I have a dataset with two geographical units of observations, districts and labor market reg…
Panel data 3 rounds: how do I count number of unique IDs for which a dummy=1 in none, one, two or all of the roundsHello, I am struggling with a simple bit of summary statistics with panel data and wonder if someone…
Subscribe to:
Post Comments (Atom)
0 Response to Coding for one relationship (Add Health WIV). Dropping people with the shortest relationship duration.
Post a Comment