Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

Phi-coefficient error

Starting to use JASP with my students since it is much quicker and more user friendly than other heavyweights - well done chaps!

Whilst running a normal contingency table and ticking Phi and Cramers V - I see that that Cramers V (probably the most robust) is calculated and Phi reports NaN - Value could not be calculated - at least one row or column contains all zeros

Data sheet looks fine to me.

Any ideas?


  • Hi Mark,

    Thanks for your kind words, and thanks for reporting this issue. Could you perhaps send us your .jasp file, so we can reproduce the anomaly? It would be even better if you could post this on our GitHub page, as all the JASP team members then see your report and it can be easily assigned & fixed.


  • Thanks for your reply - I just realised that the error was there only for 2x3 tables - Phi is only calculated for 2x2 tables. Sorry it was me having a very senior moment.


Sign In or Register to comment.