Question

Can i reject H3 based on the fact that the p-value for H2 is less than that of H1?

I have three groups of data A, B and C. I also have three null hypotheses H1, H2 and H3. H1 states that A is not significantly different from C and I failed to reject it with a p-value of 0.9419 (alpha = 0.05). H2 states that B is not significantly different from C and I failed to reject it with a p-value of 0.3366 (alpha = 0.05). H3 states that B is more accurate than A when compared to C (C being the model for accuracy). Can i reject H3 based on the fact that the p-value for H2 is less than that of H1?
Oct 25, 2011

Topics

0 / 0  ·  2 Answers  ·  64 Views

All Answers (2)

  • Achilleas Kostoulas · The University of Manchester
    No, not really. What you have proved is that H1 is very unlikely to be true, and H2 is somewhat likelier to be true but still very improbable. Hence A is in all probability different from C; B is also very likely different from C. It doesn't follow that A and B are in any way related. Your p values are not measures of how different A and B are from C, just how likely it is that they differ.
    Oct 26, 2011
  • Sando George · University of Guyana
    I see... thank you for your input.
    Oct 26, 2011

Question Followers (3)