Re: Cluster Analysis

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: Cluster Analysis

Hans Chen
Dear Listers:



I am going to conduct segmentation analysis using the Twestep Cluster in
SPSS, but SPSS warned that there are not enough valid cases to conduct the
specified cluster analysis and this command is not executed. Attached is my
data and syntax for your reference. Could you please show me how to fix the
issue using SPSS or SAS. Thanks for your help!



Hans Chen





TWOSTEP CLUSTER

 /CATEGORICAL VARIABLES = Q1 Q2A Q2B Q3 Q14 Q15 N24 Q25 Q36 N39 Q41 Q47 Q48
Q49 Q52 Q51 N53 Q54

 /DISTANCE LIKELIHOOD

 /NUMCLUSTERS AUTO 15 BIC

 /HANDLENOISE 25

 /MEMALLOCATE 64

 /CRITERIA  INITHRESHOLD (0)  MXBRANCH (8)  MXLEVEL (3)

 /PLOT

       VARCHART COMPARE BYVAR NONPARAMETRIC

 /PRINT COUNT SUMMARY

 /SAVE VARIABLE=TSC_8059 .

AIM TSC_8059

  /CATEGORICAL Q1 Q2A Q2B Q3 Q14 Q15 N24 Q25 Q36 N39 Q41 Q47 Q48 Q49 Q52 Q51
N53 Q54

  /PLOT IMPORTANCE (X=VARIABLE Y=PVALUE)

  /CRITERIA ADJUST=BONFERRONI SHOWREFLINE=NO HIDENOTSIG=NO .



*
Warnings*

* *

There are not enough valid cases to conduct the specified cluster analysis.

This command is not executed.



*
Warnings*

* *

Text: TSC_8059

                An undefined variable name, or a scratch or system variable
was specified in a variable list which accepts only standard variables.  Check
spelling and verify the existence of this variable.

                This command not executed.

=====================
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
Reply | Threaded
Open this post in threaded view
|

Re: Cluster Analysis

Jason Burke
The input cases into the cluster analysis must be complete, unless you
specify otherwise. By default, if any one of the 18 variables features
a missing value, the procedure will omit the case (unless you specify
otherwise).

The warning suggests that there are no cases with non-missing values.

To assess the problem, this syntax may assist you:

COUNT incomplete = Q1 Q2A Q2B Q3 Q14 Q15 N24 Q25 Q36 N39 Q41 Q47 Q48
Q49 Q52 Q51 N53 Q54 (MISSING) .
FREQUENCIES incomplete .

We use SPSS to create a variable (incomplete) which determines if any
of the input variables features a (user-defined or system) missing
value....

HTH,


Jason


On 2/29/08, Hans Chen <[hidden email]> wrote:

> Dear Listers:
>
>
>
> I am going to conduct segmentation analysis using the Twestep Cluster in
> SPSS, but SPSS warned that there are not enough valid cases to conduct the
> specified cluster analysis and this command is not executed. Attached is my
> data and syntax for your reference. Could you please show me how to fix the
> issue using SPSS or SAS. Thanks for your help!
>
>
>
> Hans Chen
>
>
>
>
>
> TWOSTEP CLUSTER
>
>  /CATEGORICAL VARIABLES = Q1 Q2A Q2B Q3 Q14 Q15 N24 Q25 Q36 N39 Q41 Q47 Q48
> Q49 Q52 Q51 N53 Q54
>
>  /DISTANCE LIKELIHOOD
>
>  /NUMCLUSTERS AUTO 15 BIC
>
>  /HANDLENOISE 25
>
>  /MEMALLOCATE 64
>
>  /CRITERIA  INITHRESHOLD (0)  MXBRANCH (8)  MXLEVEL (3)
>
>  /PLOT
>
>       VARCHART COMPARE BYVAR NONPARAMETRIC
>
>  /PRINT COUNT SUMMARY
>
>  /SAVE VARIABLE=TSC_8059 .
>
> AIM TSC_8059
>
>  /CATEGORICAL Q1 Q2A Q2B Q3 Q14 Q15 N24 Q25 Q36 N39 Q41 Q47 Q48 Q49 Q52 Q51
> N53 Q54
>
>  /PLOT IMPORTANCE (X=VARIABLE Y=PVALUE)
>
>  /CRITERIA ADJUST=BONFERRONI SHOWREFLINE=NO HIDENOTSIG=NO .
>
>
>
> *
> Warnings*
>
> * *
>
> There are not enough valid cases to conduct the specified cluster analysis.
>
> This command is not executed.
>
>
>
> *
> Warnings*
>
> * *
>
> Text: TSC_8059
>
>                An undefined variable name, or a scratch or system variable
> was specified in a variable list which accepts only standard variables.  Check
> spelling and verify the existence of this variable.
>
>                This command not executed.
>
> =====================
> 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
>

=====================
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