does filtering interfere with missing value replacement?

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

jqle
Posts: 2
Joined: Tue Mar 10, 2009 3:09 am

does filtering interfere with missing value replacement?

Postby jqle » Tue Mar 10, 2009 3:28 am

Hi,

I need to filter out cases which contain a certain number (which served as a marker in my imaging program) before replacing the missing values in my dataset. However, I noticed that the missing value will be filtered out as long as I activate the filter function (in this case, Select cases, if D1 NE 6979). After switching on the filter, I conducted a missing value replacement by interpolation (>transform>replace missing value>linear interpolation).

The missing values appeared nicely replaced eventhough those cases with missing values were 'slashed' (indicating that this case is filtered). May I know if it's still workable if my missing values are replaced in this manner (with the filter on)? I am just worried that some hidden steps may have taken place and the missing values are not being replaced via the 'rtg way'.

THANKS in advance! :wink:
statman
Administrator
Posts: 2737
Joined: Tue Jun 12, 2007 12:08 pm
Location: Florida, USA

Re: does filtering interfere with missing value replacement?

Postby statman » Tue Mar 10, 2009 12:00 pm

What SPSS version

Sorry, not sure I understand your question?
See the note below

NOTE: Please read the Posting Guidelines and always tell us your OS, the SPSS version and information about your study and data!

Statman
Statistical Services
Fierce
Moderator
Posts: 391
Joined: Sun Dec 28, 2008 5:55 pm
Location: Belo Horizonte, Brasil

Re: does filtering interfere with missing value replacement?

Postby Fierce » Tue Mar 10, 2009 2:38 pm

I'm not sure either, but some commands (like the ones you used) will work for all data and not only the selected cases. In other words, all missing values should have been replaced correctly. To replace only the selected cases, use the COPY SELECTED CASES TO NEW DATASET and run variables transformations there OR use syntax and run the variable transformations within IF statements.
Don't know if it answers the question, but it may help.
BR

NOTE: "Please read the Posting Guidelines and always tell us your OS, the SPSS version and some information about your data!" - by statman.

Disponível em português (mas prefira inglês para que outros também entendam a solução).
jqle
Posts: 2
Joined: Tue Mar 10, 2009 3:09 am

Re: does filtering interfere with missing value replacement?

Postby jqle » Tue Mar 10, 2009 5:17 pm

Thank you so much, statman & Fierce! Yup. My qns was answered!

For the benefits of others who might be interested in this topic, my qns was: Would the missing values be replaced correctly if the cases with missing values was not part of the cases selected. (I hope this is clearer :oops: )

I am using SPSS 17. I am very happy :D Thanks thanks!

Who is online

Users browsing this forum: No registered users and 2 guests

cron