Mean Replacement
Locked
 
|
I am trying to perform a mean replacement, but I can't quite get the SPSS syntax to work. I must have some kind of error, but I'm not seeing it. Do you see it?
COMPUTE PTR_Join IF (SYSMIS(PTR_Join)) =Mean.10(PTR_CG1 PTR_CG2 PTR_CG3 PTR_CG4 PTR_CG5 PTR_CG6 PTR_CG7 PTR_CG8 PTR_CG9 PTR_CG10 PTR_CG11 PTR_CG12 PTR_CG13 PTR_CG14 PTR_CG15 ). EXECUTE. ===================== 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 |
|
|
I am trying to perform a mean replacement, but I can't quite get the SPSS
syntax to work. I must have some kind of error, but I'm not seeing it. Do you see it? COMPUTE PTR_Join IF (SYSMIS(PTR_Join)) =Mean.10(PTR_CG1 PTR_CG2 PTR_CG3 PTR_CG4 PTR_CG5 PTR_CG6 PTR_CG7 PTR_CG8 PTR_CG9 PTR_CG10 PTR_CG11 PTR_CG12 PTR_CG13 PTR_CG14 PTR_CG15 ). EXECUTE. -- 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 |
|
|
In reply to this post by lsynhorst
Your syntax can't work. Try FRE PTR_Join. DO IF (SYSMIS(PTR_Join)). - COMPUTE PTR_Join_rec = Mean.10(PTR_CG1, PTR_CG2, PTR_CG3, PTR_CG4, PTR_CG5, PTR_CG6, PTR_CG7, PTR_CG8, PTR_CG9, PTR_CG10, PTR_CG11, PTR_CG12, PTR_CG13, PTR_CG14, PTR_CG15).END IF. FRE PTR_Join_rec. Mario Giesel Munich, Germany
Am Mittwoch, 10. Oktober 2018, 16:19:15 MESZ hat lsynhorst <[hidden email]> Folgendes geschrieben:
I am trying to perform a mean replacement, but I can't quite get the SPSS syntax to work. I must have some kind of error, but I'm not seeing it. Do you see it? COMPUTE PTR_Join IF (SYSMIS(PTR_Join)) =Mean.10(PTR_CG1 PTR_CG2 PTR_CG3 PTR_CG4 PTR_CG5 PTR_CG6 PTR_CG7 PTR_CG8 PTR_CG9 PTR_CG10 PTR_CG11 PTR_CG12 PTR_CG13 PTR_CG14 PTR_CG15 ). EXECUTE. ===================== 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 |
|
|
In reply to this post by lsynhorst
When using IF, you don't use the COMPUTE command. Also, separate variables with commas. Try this (tested).
IF (SYSMIS(PTR_Join)) PTR_Join =Mean.10(PTR_CG1, PTR_CG2, PTR_CG3, PTR_CG4, PTR_CG5, PTR_CG6, PTR_CG7, PTR_CG8, PTR_CG9, PTR_CG10 PTR_CG11, PTR_CG12, PTR_CG13, PTR_CG14, PTR_CG15 ). EXECUTE.
Brian Dates
From: SPSSX(r) Discussion <[hidden email]> on behalf of lsynhorst <[hidden email]>
Sent: Wednesday, October 10, 2018 10:19:14 AM To: [hidden email] Subject: Mean Replacement I am trying to perform a mean replacement, but I can't quite get the SPSS syntax to work. I must have some kind of error, but I'm not seeing it. Do you see it?
COMPUTE PTR_Join IF (SYSMIS(PTR_Join)) =Mean.10(PTR_CG1 PTR_CG2 PTR_CG3 PTR_CG4 PTR_CG5 PTR_CG6 PTR_CG7 PTR_CG8 PTR_CG9 PTR_CG10 PTR_CG11 PTR_CG12 PTR_CG13 PTR_CG14 PTR_CG15 ). EXECUTE. ===================== 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 |
|
In reply to this post by lsynhorst
Why is the variable SYSMIS?
Is it the result of a previous COMPUTE or READ where the computer was unable to follow your instructions? It is advantageous in the long run to preserve the distinction between user-missing and system-missing. ----- Art Kendall Social Research Consultants -- 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
Art Kendall
Social Research Consultants |
Administrator
|
In reply to this post by bdates
If the variables are contiguous in the file, as their names suggest they
might be, you can shorten that to: IF SYSMIS(PTR_Join) PTR_Join =Mean.10(PTR_CG1 TO PTR_CG15). bdates wrote > When using IF, you don't use the COMPUTE command. Also, separate variables > with commas. > > Try this (tested). > > IF (SYSMIS(PTR_Join)) PTR_Join =Mean.10(PTR_CG1, PTR_CG2, PTR_CG3, > PTR_CG4, PTR_CG5, PTR_CG6, PTR_CG7, PTR_CG8, PTR_CG9, PTR_CG10 PTR_CG11, > PTR_CG12, PTR_CG13, PTR_CG14, PTR_CG15 ). > > EXECUTE. > > > Brian Dates > ________________________________ > From: SPSSX(r) Discussion < > SPSSX-L@.UGA > > on behalf of lsynhorst < > lsynhorst2@ > > > Sent: Wednesday, October 10, 2018 10:19:14 AM > To: > SPSSX-L@.UGA > Subject: Mean Replacement > > I am trying to perform a mean replacement, but I can't quite get the SPSS > syntax to work. I must have some kind of error, but I'm not seeing it. Do > you see it? > COMPUTE PTR_Join IF (SYSMIS(PTR_Join)) =Mean.10(PTR_CG1 PTR_CG2 PTR_CG3 > PTR_CG4 PTR_CG5 PTR_CG6 PTR_CG7 PTR_CG8 PTR_CG9 PTR_CG10 PTR_CG11 PTR_CG12 > PTR_CG13 PTR_CG14 PTR_CG15 ). EXECUTE. > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@.UGA > (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 > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@.UGA > (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 ----- -- Bruce Weaver [hidden email] http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." NOTE: My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. -- 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
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
|
|
While the computation for mean replacement is straightforward, the consequences of using the subsequent data should be considered. In statistical calculations, the mean-substituted data may be misleading. The MVA procedure (Analyze > Missing Value Analysis) provides four more sophisticated methods for imputation. Even better, the MULTIPLE IMPUTATION procedure can be used to retain the uncertainty in the imputed values when they are used in supported procedures. Working with MI data can be a little more confusing, but it does the best job of handling the missing data. On Wed, Oct 10, 2018 at 11:51 AM Bruce Weaver <[hidden email]> wrote: If the variables are contiguous in the file, as their names suggest they |
|
|
It is certainly important to know why data are missing.
*IFF*these are items in a summative scale (e.g., an attitude) scale and the OP has reason to assert the items are missing at random then 10 out of 15 items used as the measure is palatable. If it were my work I would first get a handle on the reasons for items being missing. I would also want to get an handle on the reasons for missingness. First, assign user-missing codes for items and then do multiple response analyses on user-missing patterns. ----- Art Kendall Social Research Consultants -- 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
Art Kendall
Social Research Consultants |
|
|
In reply to this post by lsynhorst
Did you see the previous responses to your post?
----- Art Kendall Social Research Consultants -- 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
Art Kendall
Social Research Consultants |
|
|
In reply to this post by Art Kendall
Here, here, Art and Jon. I didn't want to get in the mechanism for missingness, but the issues of ignorability and nonignorability play a large role. Jon also made a good point of the difficulties with mean substitution
as opposed to more sophisticated methods which don't compromise the results and their interpretation.
Brian
From: SPSSX(r) Discussion <[hidden email]> on behalf of Art Kendall <[hidden email]>
Sent: Wednesday, October 10, 2018 3:28:22 PM To: [hidden email] Subject: Re: Mean Replacement It is certainly important to know why data are missing.
*IFF*these are items in a summative scale (e.g., an attitude) scale and the OP has reason to assert the items are missing at random then 10 out of 15 items used as the measure is palatable. If it were my work I would first get a handle on the reasons for items being missing. I would also want to get an handle on the reasons for missingness. First, assign user-missing codes for items and then do multiple response analyses on user-missing patterns. ----- Art Kendall Social Research Consultants -- 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 |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |