Did a reasonable search and did not find confirmation on this, but if I first
split a file and then run a function subsequently, it is the case that the created variables are indeed based on the split? As an example, splitting a file (long format) based on ID to create (save) intraindividual z-scores using the Descriptives function. To be sure, this is the result I get, but just wanted confirmation/further comments on this approach more generally. -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
If you are running a procedure that honors split files (most do) and saves variables such as DESCRIPTIVES or REGRESSION, the calculated variables are based on the split results. You can confirm this, for example, in DESCRIPTIVES by running the command again with splits still on and see that in each split, the standardized variables have mean zero. If you are running a transformation such as COMPUTE, splits are ignored. One exception is SHIFT VALUES, which was designed to respect split boundaries. On Wed, Apr 29, 2020 at 8:22 PM J.D. Haltigan <[hidden email]> wrote: Did a reasonable search and did not find confirmation on this, but if I first |
Thanks Jon. This actually clarifies some nuance for me as the distinction
between procedures and transformations return different results as relates to ignoring/honoring the split. G2Know. -- Sent from: http://spssx-discussion.1045642.n5.nabble.com/ ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
Technically Shift values is a procedure, since it passes the data, but functionally it feels like a transformation. On Wed, Apr 29, 2020 at 8:53 PM J.D. Haltigan <[hidden email]> wrote: Thanks Jon. This actually clarifies some nuance for me as the distinction |
Free forum by Nabble | Edit this page |