I am having problems with import of Excel files into SPSS. Some work just fine while others won't pass. All experimentation with "clean" new files usually work fine. Dates, integer or variables with decimals, everything seems to work. But when I try files from other sources such as people within the organization using the same network and Excel installation, they often get stuck. But a problematic file can pass if some variables are deleted. My last attempts suggest that files where there are decimals get stuck, while variables with integer values pass. Another solution seems to be to copy the original data and paste them as values rather than with the underlying formatting.
So solvable as it may be, Excel import should still be easy. I can't recall anything similar from versions before ver 24. Anyone who have experienced anything similar? Or who could say if the problem is with SPSS or if the likely cause is in the OS or our Excel installation? Robert ===================== 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
Robert Lundqvist
|
A number of changes were made to the Excel import in release 24. If you are using the dialog interface to read Excel files, then by default two new options are selected: Percentage of values that determine data type Ignore hidden rows and columns If you deselect (uncheck) these options, Excel should work like it did in previous versions. There are also new options for ignoring leading and trailing blanks for string variables, but those options are not selected (checked) by default. I suspect the culprit might be the option to determine data type based on a specified percentage of values (DATATYPEMIN subcommand). On Thu, Jan 5, 2017 at 3:38 AM, Robert Lundqvist <[hidden email]> wrote: I am having problems with import of Excel files into SPSS. Some work just fine while others won't pass. All experimentation with "clean" new files usually work fine. Dates, integer or variables with decimals, everything seems to work. But when I try files from other sources such as people within the organization using the same network and Excel installation, they often get stuck. But a problematic file can pass if some variables are deleted. My last attempts suggest that files where there are decimals get stuck, while variables with integer values pass. Another solution seems to be to copy the original data and paste them as values rather than with the underlying formatting. |
In reply to this post by Robert L
I have often found the Excel files are not actually MS Excel files, but in another format, disguised as Excel. They may be generated from a third source. Often there are illegal variable names, or non-ASCII characters (even if they look all right). Opening the file in Excel and using 'Save As...' and selecting .xls or .xlsx format has usually solved the import problem, in some cases offending variable names had to be changed.
PR
|
In reply to this post by Robert L
I got the suggestion to check if fixpack 1 version 24 had been installed, which it wasn't. And the problems now have disappeared as far as I can see. Thanks!
Robert Lundqvist
|
Free forum by Nabble | Edit this page |