IBM Support

PI56608: SPLIT FILE LAYERED FOR A STRING VARIABLE WITH SLOVAK CHARACTERS ON SLOVAK LOCALE: REGRESSION OUTPUT TABLES ARE INCORRECT

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as fixed if next.

Error description

  • You have installed IBM SPSS Statistics 23 FP2 on a Slovakian
    Windows 7 64 bit operating system. You have a file with a string
    variable that has Slovak characters in it. You use this variable
    to split the file with SPLIT FILE LAYERED command.
    After that you run a REGRESSION procedure:
    
    SET PRINTBACK = on.
    SET UNICODE=ON.
    SET LOCALE="sk_SK".
    SHOW UNICODE LOCALE.
    
    GET  FILE= 'C:\Temp\sample.sav'.
    SORT CASES  BY split_string.
    SPLIT FILE LAYERED BY split_string.
    REGRESSION
      /MISSING LISTWISE
      /STATISTICS COEFF OUTS R ANOVA
      /CRITERIA=PIN(.05) POUT(.10)
      /NOORIGIN
      /DEPENDENT y
      /METHOD=ENTER x.
    
    The output shows all tables for Regression but you notice that
    one of the rows for a Slovak value of the string variable is
    doubled on all tables and one row for another value is missing
    on all tables.
    This was reported to IBM SPSS development. Please note: This
    issue is not replicable on English operating systems using
    Slovak locale.
    As a workaround you can either use below syntax or recode the
    string variable into a numeric variable and use the numeric
    variable as split file layered variable. This will work without
    any problem on Slovak OS as well.
    
    SORT CASES  BY split_string.
    SPLIT FILE SEPARATE BY split_string.
    REGRESSION
      /MISSING LISTWISE
      /STATISTICS COEFF OUTS R ANOVA
      /CRITERIA=PIN(.05) POUT(.10)
      /NOORIGIN
      /DEPENDENT y
      /METHOD=ENTER x.
    

Local fix

  • This issue will be addressed in a future main release of IBM
    SPSS Statistics.
    

Problem summary

  • ****************************************************************
    * USERS AFFECTED:                                              *
    * This affects users of SPSS Statistics on a Slovakian locale. *
    ****************************************************************
    * PROBLEM DESCRIPTION:                                         *
    * SPLIT FILE LAYERED FOR A STRING VARIABLE WITH SLOVAK         *
    * CHARACTERS ON SLOVAK LOCALE: REGRESSION OUTPUT TABLES ARE    *
    * INCORRECT                                                    *
    ****************************************************************
    * RECOMMENDATION:                                              *
    * Please upgrade to IBM SPSS Statistics 25.0 to resolve this   *
    * issue.                                                       *
    ****************************************************************
    

Problem conclusion

Temporary fix

Comments

APAR Information

  • APAR number

    PI56608

  • Reported component name

    SPSS STATISTICS

  • Reported component ID

    5725A54ST

  • Reported release

    N00

  • Status

    CLOSED FIN

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt / Xsystem

  • Submitted date

    2016-02-04

  • Closed date

    2017-08-09

  • Last modified date

    2017-08-09

  • APAR is sysrouted FROM one or more of the following:

  • APAR is sysrouted TO one or more of the following:

Fix information

Applicable component levels

  • RN00 PSY

       UP

  • RO00 PSY

       UP

[{"Business Unit":{"code":"BU048","label":"IBM Software"},"Product":{"code":"SSCQ88K","label":"Statistics Desktop"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"23.0","Edition":"","Line of Business":{"code":"","label":""}}]

Document Information

Modified date:
09 August 2017