Announcement

Collapse
No announcement yet.
X
Collapse
  • Page of 1
  • Filter
  • Time
  • Show
Clear All
new posts
  • #1

    Panel Data - adding information from observations of parents to children

    Hi all,

    I have a question regarding a long panel set I am using.

    I have variables for household (IDHH), individual ID (HHPBASE), the survey (whether it was round 1 or round 2), PBASE (the id number of the individual within the household), RO8 (father's ID), RO9 (mother's ID), and ED6 (years of education) as below - there are a lot more variables in this and I will want to map other variables, not just years of education, but for illustration purposes I have provided the following from the data set:

    Code:
    * Example generated by -dataex-. To install: ssc install dataex
clear
input str10 IDHH double HHPBASE int(SURVEY PBASE RO8 RO9 ED6)
"010201010"  101020101001 1  1  2  .  0
"010201010"  101020101002 1  2  1  .  0
"010201010"  101020101003 1  3  .  1  8
"010201010"  101020101005 1  5  .  1  7
"010201010"  101020101006 1  6  .  1  2
"010201020"  101020102001 1  1  2  . 10
"0102010201" 101020102001 2  1 75 75 12
"010201020"  101020102003 1  3  4  1 15
"0102010201" 101020102003 2  3  3  1 15
"010201020"  101020102004 1  4  3  .  0
"0102010201" 101020102004 2  4  2  .  8
"010201020"  101020102005 1  5  .  3  7
"0102010201" 101020102005 2  5  .  2 12
"010201020"  101020102006 1  6  .  3  3
"0102010201" 101020102006 2  6  .  2  8
"010201020"  101020102007 1  7  .  3  0
"0102010201" 101020102007 2  7  .  2  5
"010201020"  101020102008 1  8  9  1 15
"0102010201" 101020102008 2  8  8  1 15
"010201020"  101020102009 1  9  8  . 12
"0102010201" 101020102009 2  9  7  . 12
"010201020"  101020102010 1 10  .  8  2
"0102010201" 101020102010 2 10  .  7  6
"010201020"  101020102011 1 11  .  8  0
"0102010201" 101020102011 2 11  .  7  2
"010201020"  101020102012 1 12 13  1 15
"0102010201" 101020102012 2 12 13  1 15
"010201020"  101020102013 1 13 12  . 12
"0102010201" 101020102013 2 13 12  .  7
"010201020"  101020102014 1 14  . 12  0
"0102010201" 101020102014 2 14  . 12  0
"010201020"  101020102015 1 15  .  .  0
"0102010201" 101020102015 2 15  .  .  0
"010201030"  101020103001 1  1  2  .  9
"0102010301" 101020103001 2  1  2 69 10
"010201030"  101020103002 1  2  1  .  0
"0102010301" 101020103002 2  2  1  .  0
"010201030"  101020103003 1  3  .  1  9
"0102010301" 101020103003 2  3  .  1 11
"010201030"  101020103004 1  4  .  1  6
"0102010301" 101020103004 2  4  .  1 12
"010201030"  101020103005 1  5  .  1  5
"0102010301" 101020103005 2  5  .  1  9
"010201030"  101020103006 1  6  .  1  1
"0102010301" 101020103006 2  6  .  1  6
"010201030"  101020103007 1  7  .  1  0
"0102010301" 101020103007 2  7  .  1  4
"010201040"  101020104001 1  1  2  .  0
"0102010401" 101020104001 2  1  2 75  0
"010201040"  101020104002 1  2  1  .  0
"0102010401" 101020104002 2  2  1  .  0
"010201040"  101020104003 1  3  .  1  6
"0102010401" 101020104003 2  3  .  1  5
"010201040"  101020104004 1  4  .  1  1
"0102010401" 101020104004 2  4  .  1  7
"010201040"  101020104005 1  5  .  1  0
"0102010401" 101020104005 2  5  .  1  6
"010201050"  101020105002 1  2  3  1  0
"0102010501" 101020105002 2  2  2 75  8
"010201050"  101020105003 1  3  2  .  0
"0102010501" 101020105003 2  3  1  .  0
"010201050"  101020105006 1  6  .  2  0
"0102010501" 101020105006 2  6  4  1  0
"010201050"  101020105007 1  7  .  2  8
"0102010501" 101020105007 2  7  7  1  9
"010201050"  101020105008 1  8  .  2  5
"0102010501" 101020105008 2  8  .  1 12
"010201050"  101020105009 1  9  .  2  4
"0102010501" 101020105009 2  9  .  1  9
"010201050"  101020105010 1 10  .  2  0
"0102010501" 101020105010 2 10  .  1  4
"010201060"  101020106001 1  1 75  .  0
"0102010601" 101020106001 2  1 75 75  0
"010201060"  101020106012 1 12  .  1  0
"0102010601" 101020106012 2 12  3  1  0
"010201060"  101020106013 1 13 14  1  0
"0102010601" 101020106013 2 13  7  1  0
"010201060"  101020106014 1 14 13  .  0
"0102010601" 101020106014 2 14  6  .  0
"010201060"  101020106015 1 15  .  1  0
"0102010601" 101020106015 2 15 11  1  0
"010201070"  101020107001 1  1  2  .  0
"0102010701" 101020107001 2  1  2 75  0
"010201070"  101020107002 1  2  1  .  0
"0102010701" 101020107002 2  2  1  .  0
"010201070"  101020107003 1  3  .  1  0
"0102010701" 101020107003 2  3  .  1  0
"010201070"  101020107004 1  4  .  1  7
"0102010701" 101020107004 2  4  .  1  8
"010201070"  101020107006 1  6  .  1  0
"0102010701" 101020107006 2  6  .  1  0
"010201070"  101020107007 1  7  .  1  0
"0102010701" 101020107007 2  7  .  1  0
"010201070"  101020107008 1  8  .  1  1
"0102010701" 101020107008 2  8  .  1  9
"010201070"  101020107009 1  9  .  1  1
"0102010701" 101020107009 2  9  .  1  9
"010201080"  101020108001 1  1  2  .  0
"0102010801" 101020108001 2  1  2  .  0
"010201080"  101020108002 1  2  1  .  0
end
label values SURVEY SURVEY
label def SURVEY 1 "IHDS1 1", modify
label def SURVEY 2 "IHDS2 2", modify
label values RO8 RO8
label def RO8 75 "Died 75", modify
label values RO9 RO9
label def RO9 69 "Other nonres 69", modify
label def RO9 75 "Died 75", modify
label values ED6 ED6
label def ED6 0 "none, <1 0", modify
label def ED6 1 "1st class 1", modify
label def ED6 2 "2nd class 2", modify
label def ED6 3 "3rd class 3", modify
label def ED6 4 "4th class 4", modify
label def ED6 5 "5th class 5", modify
label def ED6 6 "6th class 6", modify
label def ED6 7 "7th class 7", modify
label def ED6 8 "8th class 8", modify
label def ED6 9 "9th class 9", modify
label def ED6 10 "Secondary 10", modify
label def ED6 11 "11th Class 11", modify
label def ED6 12 "High Secondary 12", modify
label def ED6 15 "Bachelors 15", modify
    What I am attempting to do here is to, for each observation, add the education levels of the mother and father. Ultimately, I will want to run a regression where the outcome variable is child's education, and I want to control for parent's education. Of course, I understand that this will reduce the number of observations, but that is fine as I am interested in the younger generations cohabiting with their parents, in which case the data should be present for all individuals.

    What would be the best way to do this?

    Thanks in advance!
    Sanchit
    Tags: None
  • #2
    Well, either I'm misunderstanding your explanation of the data or there is something seriously wrong with your data. Let's look at IDHH 010201010 in round 1. PBASE 1 has PBASE 2 as his father, but PBASE 2 has PBASE 1 as his father. So PBASE 1 and PBASE 2 are each the father of the other. That can't possibly be right.

    Also your variables R08 and R09 also contain special values referring to things like "Died 75" and "Other nonres 69." That is one of the most bizarre conflations of unrelated information in the same variable I have ever seen and I cannot believe you will get through much analysis of this data without all sorts of strange errors cropping up as a result of that. I strongly recommend replacing those by missing values and, if the information about Died 75 or Other nonres 69 is important, create new variables to indicate those conditions.

    Anyway, assuming that the data are messed up and that your description of it is correct, the following code, when supplied with correct data, would work
    Code:
    * Example generated by -dataex-. To install: ssc install dataex
clear*
input str10 IDHH double HHPBASE int(SURVEY PBASE RO8 RO9 ED6)
"010201010"  101020101001 1  1  2  .  0
"010201010"  101020101002 1  2  1  .  0
"010201010"  101020101003 1  3  .  1  8
"010201010"  101020101005 1  5  .  1  7
"010201010"  101020101006 1  6  .  1  2
"010201020"  101020102001 1  1  2  . 10
"0102010201" 101020102001 2  1 75 75 12
"010201020"  101020102003 1  3  4  1 15
"0102010201" 101020102003 2  3  3  1 15
"010201020"  101020102004 1  4  3  .  0
"0102010201" 101020102004 2  4  2  .  8
"010201020"  101020102005 1  5  .  3  7
"0102010201" 101020102005 2  5  .  2 12
"010201020"  101020102006 1  6  .  3  3
"0102010201" 101020102006 2  6  .  2  8
"010201020"  101020102007 1  7  .  3  0
"0102010201" 101020102007 2  7  .  2  5
"010201020"  101020102008 1  8  9  1 15
"0102010201" 101020102008 2  8  8  1 15
"010201020"  101020102009 1  9  8  . 12
"0102010201" 101020102009 2  9  7  . 12
"010201020"  101020102010 1 10  .  8  2
"0102010201" 101020102010 2 10  .  7  6
"010201020"  101020102011 1 11  .  8  0
"0102010201" 101020102011 2 11  .  7  2
"010201020"  101020102012 1 12 13  1 15
"0102010201" 101020102012 2 12 13  1 15
"010201020"  101020102013 1 13 12  . 12
"0102010201" 101020102013 2 13 12  .  7
"010201020"  101020102014 1 14  . 12  0
"0102010201" 101020102014 2 14  . 12  0
"010201020"  101020102015 1 15  .  .  0
"0102010201" 101020102015 2 15  .  .  0
"010201030"  101020103001 1  1  2  .  9
"0102010301" 101020103001 2  1  2 69 10
"010201030"  101020103002 1  2  1  .  0
"0102010301" 101020103002 2  2  1  .  0
"010201030"  101020103003 1  3  .  1  9
"0102010301" 101020103003 2  3  .  1 11
"010201030"  101020103004 1  4  .  1  6
"0102010301" 101020103004 2  4  .  1 12
"010201030"  101020103005 1  5  .  1  5
"0102010301" 101020103005 2  5  .  1  9
"010201030"  101020103006 1  6  .  1  1
"0102010301" 101020103006 2  6  .  1  6
"010201030"  101020103007 1  7  .  1  0
"0102010301" 101020103007 2  7  .  1  4
"010201040"  101020104001 1  1  2  .  0
"0102010401" 101020104001 2  1  2 75  0
"010201040"  101020104002 1  2  1  .  0
"0102010401" 101020104002 2  2  1  .  0
"010201040"  101020104003 1  3  .  1  6
"0102010401" 101020104003 2  3  .  1  5
"010201040"  101020104004 1  4  .  1  1
"0102010401" 101020104004 2  4  .  1  7
"010201040"  101020104005 1  5  .  1  0
"0102010401" 101020104005 2  5  .  1  6
"010201050"  101020105002 1  2  3  1  0
"0102010501" 101020105002 2  2  2 75  8
"010201050"  101020105003 1  3  2  .  0
"0102010501" 101020105003 2  3  1  .  0
"010201050"  101020105006 1  6  .  2  0
"0102010501" 101020105006 2  6  4  1  0
"010201050"  101020105007 1  7  .  2  8
"0102010501" 101020105007 2  7  7  1  9
"010201050"  101020105008 1  8  .  2  5
"0102010501" 101020105008 2  8  .  1 12
"010201050"  101020105009 1  9  .  2  4
"0102010501" 101020105009 2  9  .  1  9
"010201050"  101020105010 1 10  .  2  0
"0102010501" 101020105010 2 10  .  1  4
"010201060"  101020106001 1  1 75  .  0
"0102010601" 101020106001 2  1 75 75  0
"010201060"  101020106012 1 12  .  1  0
"0102010601" 101020106012 2 12  3  1  0
"010201060"  101020106013 1 13 14  1  0
"0102010601" 101020106013 2 13  7  1  0
"010201060"  101020106014 1 14 13  .  0
"0102010601" 101020106014 2 14  6  .  0
"010201060"  101020106015 1 15  .  1  0
"0102010601" 101020106015 2 15 11  1  0
"010201070"  101020107001 1  1  2  .  0
"0102010701" 101020107001 2  1  2 75  0
"010201070"  101020107002 1  2  1  .  0
"0102010701" 101020107002 2  2  1  .  0
"010201070"  101020107003 1  3  .  1  0
"0102010701" 101020107003 2  3  .  1  0
"010201070"  101020107004 1  4  .  1  7
"0102010701" 101020107004 2  4  .  1  8
"010201070"  101020107006 1  6  .  1  0
"0102010701" 101020107006 2  6  .  1  0
"010201070"  101020107007 1  7  .  1  0
"0102010701" 101020107007 2  7  .  1  0
"010201070"  101020107008 1  8  .  1  1
"0102010701" 101020107008 2  8  .  1  9
"010201070"  101020107009 1  9  .  1  1
"0102010701" 101020107009 2  9  .  1  9
"010201080"  101020108001 1  1  2  .  0
"0102010801" 101020108001 2  1  2  .  0
"010201080"  101020108002 1  2  1  .  0
end
label values SURVEY SURVEY
label def SURVEY 1 "IHDS1 1", modify
label def SURVEY 2 "IHDS2 2", modify
label values RO8 RO8
label def RO8 75 "Died 75", modify
label values RO9 RO9
label def RO9 69 "Other nonres 69", modify
label def RO9 75 "Died 75", modify
label values ED6 ED6
label def ED6 0 "none, <1 0", modify
label def ED6 1 "1st class 1", modify
label def ED6 2 "2nd class 2", modify
label def ED6 3 "3rd class 3", modify
label def ED6 4 "4th class 4", modify
label def ED6 5 "5th class 5", modify
label def ED6 6 "6th class 6", modify
label def ED6 7 "7th class 7", modify
label def ED6 8 "8th class 8", modify
label def ED6 9 "9th class 9", modify
label def ED6 10 "Secondary 10", modify
label def ED6 11 "11th Class 11", modify
label def ED6 12 "High Secondary 12", modify
label def ED6 15 "Bachelors 15", modify

format HHPBASE %12.0f

rename RO8 father_id
rename RO9 mother_id

isid IDHH PBASE SURVEY

frame copy default second
frlink m:1 IDHH father_id SURVEY, frame(second IDHH PBASE SURVEY) gen(flink)
frlink m:1 IDHH mother_id SURVEY, frame(second IDHH PBASE SURVEY) gen(mlink)

gen father_educ = frval(flink, ED6)
gen mother_educ = frval(mlink, ED6)
label values father_educ mother_educ ED6
    Note: The renaming of those variables is not necessary: you remove the -rename- commands and replace the references to father_id, and mother_id by their original names in the rest of the code. I just do it because I find it very difficult to work with variable names like that and keep straight which is which. Since there is a fair amount of action going on here that requires keeping mother and father straight, I find it better to use variable names that state which is which. But you don't have to do that if you don't want to.
    Last edited by Clyde Schechter; 30 Mar 2020, 19:41.
    • 2 likes

    Comment

    • #3
      Originally posted by Clyde Schechter View Post
      Well, either I'm misunderstanding your explanation of the data or there is something seriously wrong with your data. Let's look at IDHH 010201010 in round 1. PBASE 1 has PBASE 2 as his father, but PBASE 2 has PBASE 1 as his father. So PBASE 1 and PBASE 2 are each the father of the other. That can't possibly be right.

      Also your variables R08 and R09 also contain special values referring to things like "Died 75" and "Other nonres 69." That is one of the most bizarre conflations of unrelated information in the same variable I have ever seen and I cannot believe you will get through much analysis of this data without all sorts of strange errors cropping up as a result of that. I strongly recommend replacing those by missing values and, if the information about Died 75 or Other nonres 69 is important, create new variables to indicate those conditions.

      Anyway, assuming that the data are messed up and that your description of it is correct, the following code, when supplied with correct data, would work
      Code:
      * Example generated by -dataex-. To install: ssc install dataex
clear*
input str10 IDHH double HHPBASE int(SURVEY PBASE RO8 RO9 ED6)
"010201010" 101020101001 1 1 2 . 0
"010201010" 101020101002 1 2 1 . 0
"010201010" 101020101003 1 3 . 1 8
"010201010" 101020101005 1 5 . 1 7
"010201010" 101020101006 1 6 . 1 2
"010201020" 101020102001 1 1 2 . 10
"0102010201" 101020102001 2 1 75 75 12
"010201020" 101020102003 1 3 4 1 15
"0102010201" 101020102003 2 3 3 1 15
"010201020" 101020102004 1 4 3 . 0
"0102010201" 101020102004 2 4 2 . 8
"010201020" 101020102005 1 5 . 3 7
"0102010201" 101020102005 2 5 . 2 12
"010201020" 101020102006 1 6 . 3 3
"0102010201" 101020102006 2 6 . 2 8
"010201020" 101020102007 1 7 . 3 0
"0102010201" 101020102007 2 7 . 2 5
"010201020" 101020102008 1 8 9 1 15
"0102010201" 101020102008 2 8 8 1 15
"010201020" 101020102009 1 9 8 . 12
"0102010201" 101020102009 2 9 7 . 12
"010201020" 101020102010 1 10 . 8 2
"0102010201" 101020102010 2 10 . 7 6
"010201020" 101020102011 1 11 . 8 0
"0102010201" 101020102011 2 11 . 7 2
"010201020" 101020102012 1 12 13 1 15
"0102010201" 101020102012 2 12 13 1 15
"010201020" 101020102013 1 13 12 . 12
"0102010201" 101020102013 2 13 12 . 7
"010201020" 101020102014 1 14 . 12 0
"0102010201" 101020102014 2 14 . 12 0
"010201020" 101020102015 1 15 . . 0
"0102010201" 101020102015 2 15 . . 0
"010201030" 101020103001 1 1 2 . 9
"0102010301" 101020103001 2 1 2 69 10
"010201030" 101020103002 1 2 1 . 0
"0102010301" 101020103002 2 2 1 . 0
"010201030" 101020103003 1 3 . 1 9
"0102010301" 101020103003 2 3 . 1 11
"010201030" 101020103004 1 4 . 1 6
"0102010301" 101020103004 2 4 . 1 12
"010201030" 101020103005 1 5 . 1 5
"0102010301" 101020103005 2 5 . 1 9
"010201030" 101020103006 1 6 . 1 1
"0102010301" 101020103006 2 6 . 1 6
"010201030" 101020103007 1 7 . 1 0
"0102010301" 101020103007 2 7 . 1 4
"010201040" 101020104001 1 1 2 . 0
"0102010401" 101020104001 2 1 2 75 0
"010201040" 101020104002 1 2 1 . 0
"0102010401" 101020104002 2 2 1 . 0
"010201040" 101020104003 1 3 . 1 6
"0102010401" 101020104003 2 3 . 1 5
"010201040" 101020104004 1 4 . 1 1
"0102010401" 101020104004 2 4 . 1 7
"010201040" 101020104005 1 5 . 1 0
"0102010401" 101020104005 2 5 . 1 6
"010201050" 101020105002 1 2 3 1 0
"0102010501" 101020105002 2 2 2 75 8
"010201050" 101020105003 1 3 2 . 0
"0102010501" 101020105003 2 3 1 . 0
"010201050" 101020105006 1 6 . 2 0
"0102010501" 101020105006 2 6 4 1 0
"010201050" 101020105007 1 7 . 2 8
"0102010501" 101020105007 2 7 7 1 9
"010201050" 101020105008 1 8 . 2 5
"0102010501" 101020105008 2 8 . 1 12
"010201050" 101020105009 1 9 . 2 4
"0102010501" 101020105009 2 9 . 1 9
"010201050" 101020105010 1 10 . 2 0
"0102010501" 101020105010 2 10 . 1 4
"010201060" 101020106001 1 1 75 . 0
"0102010601" 101020106001 2 1 75 75 0
"010201060" 101020106012 1 12 . 1 0
"0102010601" 101020106012 2 12 3 1 0
"010201060" 101020106013 1 13 14 1 0
"0102010601" 101020106013 2 13 7 1 0
"010201060" 101020106014 1 14 13 . 0
"0102010601" 101020106014 2 14 6 . 0
"010201060" 101020106015 1 15 . 1 0
"0102010601" 101020106015 2 15 11 1 0
"010201070" 101020107001 1 1 2 . 0
"0102010701" 101020107001 2 1 2 75 0
"010201070" 101020107002 1 2 1 . 0
"0102010701" 101020107002 2 2 1 . 0
"010201070" 101020107003 1 3 . 1 0
"0102010701" 101020107003 2 3 . 1 0
"010201070" 101020107004 1 4 . 1 7
"0102010701" 101020107004 2 4 . 1 8
"010201070" 101020107006 1 6 . 1 0
"0102010701" 101020107006 2 6 . 1 0
"010201070" 101020107007 1 7 . 1 0
"0102010701" 101020107007 2 7 . 1 0
"010201070" 101020107008 1 8 . 1 1
"0102010701" 101020107008 2 8 . 1 9
"010201070" 101020107009 1 9 . 1 1
"0102010701" 101020107009 2 9 . 1 9
"010201080" 101020108001 1 1 2 . 0
"0102010801" 101020108001 2 1 2 . 0
"010201080" 101020108002 1 2 1 . 0
end
label values SURVEY SURVEY
label def SURVEY 1 "IHDS1 1", modify
label def SURVEY 2 "IHDS2 2", modify
label values RO8 RO8
label def RO8 75 "Died 75", modify
label values RO9 RO9
label def RO9 69 "Other nonres 69", modify
label def RO9 75 "Died 75", modify
label values ED6 ED6
label def ED6 0 "none, <1 0", modify
label def ED6 1 "1st class 1", modify
label def ED6 2 "2nd class 2", modify
label def ED6 3 "3rd class 3", modify
label def ED6 4 "4th class 4", modify
label def ED6 5 "5th class 5", modify
label def ED6 6 "6th class 6", modify
label def ED6 7 "7th class 7", modify
label def ED6 8 "8th class 8", modify
label def ED6 9 "9th class 9", modify
label def ED6 10 "Secondary 10", modify
label def ED6 11 "11th Class 11", modify
label def ED6 12 "High Secondary 12", modify
label def ED6 15 "Bachelors 15", modify

format HHPBASE %12.0f

rename RO8 father_id
rename RO9 mother_id

isid IDHH PBASE SURVEY

frame copy default second
frlink m:1 IDHH father_id SURVEY, frame(second IDHH PBASE SURVEY) gen(flink)
frlink m:1 IDHH mother_id SURVEY, frame(second IDHH PBASE SURVEY) gen(mlink)

gen father_educ = frval(flink, ED6)
gen mother_educ = frval(mlink, ED6)
label values father_educ mother_educ ED6
      Note: The renaming of those variables is not necessary: you remove the -rename- commands and replace the references to father_id, and mother_id by their original names in the rest of the code. I just do it because I find it very difficult to work with variable names like that and keep straight which is which. Since there is a fair amount of action going on here that requires keeping mother and father straight, I find it better to use variable names that state which is which. But you don't have to do that if you don't want to.
      Hi Clyde,

      Thank you so much for this, it worked perfectly. I did not know of frames and it has been very helpful.

      I just wanted to clarify that I got the labels a little mixed up, RO8 actually refers to the spouse, hence why two people would have each other marked as spouse, my bad completely. Have definitely renamed my variables to use as I go forward!

      Kind regards,
      Sanchit

      Comment

      • #4
        Dear Prof. Clyde,

        I am using Stata version 14.2, so I am unable to use frame command/function. Could you please rewrite the code in #2 using merge command? Thank you.

        Comment

        • #5
          Code:
          rename RO8 father_id
rename RO9 mother_id

isid IDHH PBASE SURVEY

clonevar link = PBASE
clonevar parent_educ = ED6
tempfile second
save `second'
drop link parent_educ

clonevar link = father_id
merge m:1 IDHH link SURVEY using `second', keepusing(parent_educ) ///
    keep(master match) nogenerate
rename parent_educ father_educ
sort IDHH PBASE SURVEY

drop link
clonevar link = mother_id
merge m:1 IDHH link SURVEY using `second', keepusing(parent_educ) ///
    keep(master match) nogenerate
rename parent_educ mother_educ

label values father_educ mother_educ ED6
          • 1 like

          Comment

          • #6
            Thank you so much for your kind help, Prof. Clyde.

            Comment

            Previous Next
            Working...
            X