SPSS for regression analysis
October 31, 2016

SPSS for regression analysis

Mykola Osipov | TrustRadius Reviewer
Score 2 out of 10
Vetted Review
Verified User

Modules Used

  • IBM SPSS Modeler

Overall Satisfaction with IBM SPSS

I used a trial version of IBM SPSS, and I needed just the regression analysis part of SPSS. I wanted to find a proper model to fit my data, but I did not find SPSS helpful.

Pros

  • All I can say is that the regression analysis part of SPSS is very poor. The GUI is amazing, but the content is useless!

Cons

  • SPSS can be used for building simple regressions, only.
  • Researchers need the system to find models that fit observed data the best, and they do not know what model or its structure to choose. Researchers should enter the data and press the button, and SPSS should tell them, like "Hey buddy, these models... fit your data the best, and the standard errors are..", or, "Hey buddy, I'm sorry, I was not able to find any suitable model for your data..."
  • Not positive, not negative, I have just found SPSS to be not suitable for my tasks
I did not find any other product, suitable for my tasks, so I developed my own, called The Synthesis of Optimal Models (SOM).
As I said, SPSS is excellent for regressions like: y=ax+b, y=ax2 +bx+c, or y=sin(x). However, I was not able to build multidimensional models y=f(x1, x2, x3,...,xm).

Comments

  • Tom Woodward | TrustRadius Reviewer
    I'm an IBM Employee so take this with a grain of salt. First I'm not an SPSS user but I do know that there are a number of SPSS products, Modeler, Modeler Gold, Statistics, etc. Some 17 different packages and options I believe. I also know that simply googling "spss statistics regression multidimensional " returns dozens and dozens of returns; that says to me what this reviewer said didn't exist has been part of Statistics for many years, that he may not have been using the right version, or didn't have the necessary skills. trustradius really should not lump all the different products in one general bucket, and if they do the reviewers should be explicit in which offering they're discussing. Simply saying an "SPSS Trial" may be misleading and inappropriate.

More Reviews of IBM SPSS Statistics