Linking variables from SA 1998 HR, PR & IR files to Adult Health respondents [message #16333] |
Fri, 14 December 2018 12:05 ![Go to next message Go to next message](/theme/default/images/down.png) |
kafuia
Messages: 2 Registered: May 2014
|
Member |
|
|
I am trying to link variables from the household recode, person's recode, and individual recode (women's survey) files to the Adult Health (AH) file in the South Africa 1998 DHS. The Adult Health survey included male and female household members aged 15 years and over. However, the adult health file contains the ea number, cluster number, household number and a variable called 'qline' as identifiers. Qline is described as 'line number of woman'. There is not a variable called 'respondent's line number' in that file, such as v003, hv003, hvidx or index to the household schedule (idx4) that appear in the other files. Is the variable qline in fact supposed to be the same as the 'respondent's line number' and was just mis-labelled? Or does it only represent the line number of the woman in the household who participated in the women's survey? If so, how can we link variables from the HR and PR files to respondents in the AH (adult health) file, and how can we link data from individuals in the files to the data for members listed in the household schedule? Is there another variable that can be used to link the individuals across files so that analyses can include variables from different files?
Thank you
|
|
|
|