Regression 'Warning' in SPSS: What does it mean?

[gemma7]

I have been performing linear regressions in SPSS. Before my output is reported I have received this 'warning box':

"For models with dependent variable (4 items), some variables have impossible tolerances. The original correlation matrix may not be positive definite. Pairwise deletion may be inappropriate."

I know how to check the tolerance levels of the IV's (by looking at the collinearity statistics) but this doesn't report the tolerance levels for the DV.

My question is: What does this warning actually mean in terms of the interpretation of my regression? Is it still valid for me to report these regression results, and do I need to report this warning in the report?

Thank you to anyone who can help me with this!!! :)

[TheAnalysisFactor]

Hi Gemma7,

I've never seen that warning. Can you give more information about what you're doing? Maybe post your code and describe your variables?

It seems you've selected pairwise deletion to deal with missing data? Which data are missing and how much?

Karen

[bukket]

Hi, I have to use SPSS for one of my assignments and whenever I try to make a linear regression it gives the following warning:

''Text: Export
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.

Unrecognized text appears on the REGRESSION command. The only recognized subcommands are: Global options: DESCRIPTIVES MATRIX MISSING WIDTH; Case selection/weight: REGWGT SELECT; Variable list: VARIABLES; Equation options: CRITERIA NOORIGIN ORIGIN STATISTICS; Dependent variable(s): DEPENDENT; Equ. methods: METHOD BACKWARD ENTER FORWARD REMOVE STEPWISE TEST; Residuals: RESIDUAL CASEWISE PARTIALPLOT SAVE SCATTERPLOT OUTFILE. Text found: capita.
*WARNING* REGRESSION syntax scan continues. Further diagnostics from this command may be misleading - interpret with care. ''

I took my data from an excel document but there seems nothing wrong when I look at the variable view in SPSS. I want to have a regression with gdp as the dependent and expors per capita the independent variable. I could only attachment the excel file to give a better idea because the forum does not accept .sav files. I will be glad if anyone can find a solution to my problem which has probably a silly reason which I can not understand..

Thanks in advance :))