Merge the intertemporal data

[Sa Li]

Dear scholars:
i have a question about how to process the data. My research topic is the long-term effects of child malnutrition.And i am focusing on the adult outcomes. For example, my research period is the 1991-2015. My research age group is 0-18 years. What I do now is shown as follows. What i confuse is that from child-level, I got data from 1991-2011, it is panel data. But from adult outcomes, i only have 2015, it is cross-sectional. Therefore, i am really not sure whether i am doing is right.
Merge the datasets
Child-level adult-level
Year 1991-2011 2015
age 0-18 years 18-43

Also, actually, i am not sure whether what i did is right. Because the survey i used has ten waves from 1989-2015. So this is what i want to focus.

I read some literature, for example, "to investigate if stunting or wasting among children aged 2–5 years is associated with cardiovascular disease risk or educational achievement during adolescence", this one use the data measured in 1999-2000 and repeated in 2004-2005. What makes me fee confused is that how to merge two panel data sets.

REALLY appreciated to get your reply. Thanks in advance.

[Sa Li]

Also, this survey is collect in ten waves, not every year including 1991, 1993,1997,2000, 2004, 2006,2009,2011, 2015。

[Jorge L. Guzman]

Hi Sa Li,
Thanks for your question.
The first thing that comes to mind is the ID of each individual.
I am sure that if you are following people throughout time, there must be an ID per person. For example in the year 2000, the person with the ID #007 was 10 years old and in 2010 he was 20. When you merge the database from the year 2000 and the year 2010 you must do a merge 1:1 ID using database year2010.
That would be intuition if you want to follow people througout time. Another way of looking at this is you are putting/pasting one database right next to the other.
Another way of doing it, would be to append the data. If you have the same ID and year and you are doing a panel, you would have to append it and then use the commands xtset and so on. Another way of explaining append is...you are putting one database on top of the other.
Those two approaches should work in principle.
Kind regards,
Jorge

[Sa Li]

Thank you so much for your reply. It helps a lot. Let me explain it for you to see whether i understand right. I want to include all the waves, what i did right now is to merge 1991-2001's child level information with 2015 adult information. In other words, what i did right now is to put adult-level's information in 2015 right next to child-level information from 1991-2011.

And，if i focus on the child-level information from 1991-2006, adult-level from 2009-2015. In this case, i use append command, to put adult-level's information under the child-level right?

Thank you so much.