The DHS Program User Forum: Dataset use in SPSS

Home » Data » Dataset use in SPSS » different samples

Show: Today's Messages :: Show Polls :: Message Navigator

different samples [message #9050]

Tue, 02 February 2016 09:11

giulia.mandrelli
Messages: 5
Registered: February 2016

Member

In Namibia 2013 data set(dhs survey) I have in the variable "Child's age in months" 1965 valid answers, while in the variable "height/age" only 1558. How can I do if I want a sample of 1558 also in the "child's age in months" variable? I would like to have the same valid answers in all the variables.
Thank you very much in advance
Giulia

Report message to a moderator

Re: different samples [message #9052 is a reply to message #9050]

Tue, 02 February 2016 14:40

Reduced-For(u)m
Messages: 292
Registered: March 2013

Senior Member

An easy way to do it is to generate an indicator variable for "valid height/age" measure, and then just keep those observations ("drop if HAZvalid==0" in Stata). Almost surely all the valid height/age measures have valid ages, and so you would just be trimming away the height/age measures that were invalid (outside acceptable range) or weren't taken (for whatever reason).

But basically all you should have to do is drop the missing HAZ values, and then everyone with a valid HAZ should have a valid age in months.

Report message to a moderator

Re: different samples [message #9057 is a reply to message #9052]

Tue, 02 February 2016 16:21

giulia.mandrelli
Messages: 5
Registered: February 2016

Member

Thank you very much for the reply, but how can I drop the HAz?? I don't know how to do it in SPSS

Report message to a moderator

Re: different samples [message #9058 is a reply to message #9057]

Tue, 02 February 2016 20:38

Reduced-For(u)m
Messages: 292
Registered: March 2013

Senior Member

Perhaps this will help: https://kb.iu.edu/d/afay

"The SELECT command with the SYSMIS() function can drop all missing cases from the current SPSS data set."

It goes on to give the syntax and an example - I think that is probably the right command, but I don't really use SPSS. If you can't make it work, you should google around or check out the SPSS help files on-line.

Report message to a moderator