Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Identifying households where members fulfill a condition

    Hi all,

    I would like study the impact of a raising of the school leaving age reform on later life outcomes which are measured at the household level. Those born since April 1 1933 were affected by this reform. Only year of birth is publicly available in the data.

    As the effect of the policy may differ if only one partner were affected vs. both, I would like to run the analyses separately by these groups. How can I properly identify households where both partners were affected by the reform i.e. both partners were born after 1933. More generally, how do I ensure that all individuals within a household fulfill a particular condition?

    In the below data extract, I've provided the individual ID, household ID, year of birth & status of the respondent (finstat_fin). finstat_fin identifies whether someone is a core member or a partner of a core member. For e.g. : C1CM is a cohort 1 core member. C1CP is the partner of a cohort 1 core member. C1NP4 is a new partner of a cohort 1 core member, identified at wave 4. C1YP is a younger partner of a cohort 1 core member and C4OP is an older partner of a cohort 4 core member.

    Code:
    * Example generated by -dataex-. To install: ssc install dataex
    clear
    input long idauniq int idahhw1 str12 finstat_fin float yob_fin
    119190 10001 "C1CM" 1935
    118967 10001 "C1CM" 1925
    107161 10002 "C1CM" 1950
    105327 10002 "C1YP" 1955
    118162 10003 "C1CM" 1921
    120124 10003 "C1CM" 1923
    106163 10004 "C1CM" 1932
    118542 10005 "C1CM" 1951
    118861 10005 "C1CM" 1952
    120442 10006 "C1CM" 1947
    108230 10007 "C1CM" 1935
    106289 10008 "C1CM" 1936
    117769 10009 "C1CM" 1940
    119916 10010 "C1CM" 1944
    117816 10011 "C1CM" 1947
    117652 10011 "C1CM" 1947
    112899 10012 "C1CM" 1940
    120884 10013 "C1CM" 1933
    118229 10014 "C1CM" 1933
    117527 10014 "C1CM" 1933
    end
    Thanks in advance for your help with this, really appreciated.

    Many thanks
    Karen

  • #2
    If you sort by year of birth, then you can look at whether the year of birth corresponding to the last observation in a household is less than or equal to 1933.

    Code:
    bys idahhw1 (yob_fin): gen wanted= yob_fin[2]<=1933

    Comment


    • #3
      Thank you so much Andrew for your help, much appreciated.

      Comment

      Working...
      X