I am questionning the Chi-squared table given in 100 Statistical Tests (3rd Edition, 2006) by Gopal K. Kanji, reproduced in part below.

Chi-squared table

Specifically, I think the Two-sided and One-sided levels of significance on the right should be 0.02 and 0.01, instead of 0.01 and 0.005 (regaining consistency with the values below, as well as making the progression from left ot right more regular).

Can somemone give an independent confirmation? (Edit: that is now done in comments).

Can somemone with access to the 1995 edition tell if the table is similar? (Edit: That reference is used in W. Killmann, W. Schindler, A proposal for: Functionality classes for random number generators, Version 2.0, September 18, 2011 (part of AIS-31 at BSI), and in this context the rightmost column in that Chi-squared table is relevant).

  • 1
    $\begingroup$ I checked it with a SPSS function and, yes, you are right - it's a typo in the book. $\endgroup$ – ttnphns Jan 11 '12 at 15:10
  • 1
    $\begingroup$ You can check the table yourself online. E.g., the "6.63" ought to agree with inversecdf(chisquaredistribution(1), 0.995), which returns 7.87944 instead. $\endgroup$ – whuber Jan 11 '12 at 16:29
  • $\begingroup$ Thanks for the two independent confirmations of the error. I still wish I knew what the 1995 edition contained, though. $\endgroup$ – fgrieu Jan 12 '12 at 7:58

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.