For reverse coding variables in SPSS, you must copy your variables,
RECODE them and then adjust their value labels. Comparing the recoded variables with the copies ensures the results are correct. This tutorials quickly walks you through.
Read more...
SPSS TUTORIALS 
THIS TUTORIAL HAS 3 COMMENTS:
By Dobromir Stoyanov on March 15th, 2019
Hello, I think that your website is quite useful for researchers. I have one question however, that I could not find answer to on your website, so I will be glad to hear your opinion on it. I want to run logliner analysis on SPSS Statistics 25 but I have more than 10 variables.= 1 outcome variable and 11 predictors/factors. The big number of factor variables can be explained by the fact that a big part of them are dummy variables from a multiple response question. Is there an alternative way to run the analysis. Thank you.
By Ruben Geert van den Berg on March 15th, 2019
Loglinear analysis on so many variables is insane. And you'll need a huge sample size to satisfy its assumptions. Besides, there's no "outcome variable" or "predictors" in loglinear analysis. There's just variables as with correlations.
This sounds as if some kind of regression analysis is what you're looking for.
By iftikhar hussain on May 9th, 2019
Hi,
Many thanks for such a clear writing. I suggest you should add printable version facility as well. It will help many like me. Good job