I have some qualitative data and having trouble preparing the data before I perform the analysis in SPSS. My understanding of Friedman's anova is that the data must be ranked from lowest to highest.

My experimental design was repeated measures using a single variable (interaction technique) with three levels (a,b,c for example). The experiment involved participants ranking there agreement to 13 questions (from the IBM Usability Satisfaction Questionnaire). Effectil my SPSS spreadsheet has 39 columns (3 levels * 13 questions). Here participants had to rank there agreement on a scale between 1 - 5 with 1 being Strongly Agree and 5 being Strongly Disagree. Hence questions ranked with a value of 1 are more positive.

Considering this I am confused how I enter the data into SPSS. If 1 is better should I invert this value and represent 1 as a ranking of 5 when entering into SPSS, and invert 5 -> 1 etc etc?

I have tried both approaches and get different results. Which is correct?

Any help would be greatly appreciated as I'm rather confused.

  • 2
    $\begingroup$ 1) Friedman analysis is for quantitative data; 2) You don't have to convert data to ranks: the program will do it for you internally; 3) Result (significance) is not affected by reversion of the rating scale $\endgroup$ – ttnphns May 24 '12 at 13:53
  • $\begingroup$ Many thanks for your reply, the significance and the chi square result were affected when the results were inverted? $\endgroup$ – Bear May 28 '12 at 20:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.