0
$\begingroup$

I've a question regarding insignificant p-value.

If I got p-value greater than 0.05, does that mean my result is biological insignificant?

Can anyone help me to clear this up?

$\endgroup$
  • 1
    $\begingroup$ I think your question is related to the meaning of p-value. If you still don't understand after reading this you can ask specific questions =) $\endgroup$ – Max Apr 20 '18 at 19:02
  • 1
    $\begingroup$ $p$-values tell you nothing about biological significance. $\endgroup$ – Erick Wong Apr 20 '18 at 19:03
0
$\begingroup$

Suppose you are testing $H_0: \mu = 100$ against $H_a: \mu \ne 100,$ where the value of $\mu$ may be of importance in biology.

If the p-value of a test exceeds 5%, then there is not sufficient evidence to reject $H_0.$ Then you would say there is not evidence of a statistically significant change in $\mu$ (from 100) according to the data used for the test.

Even if the p-value is smaller than 5% and you can claim a statistically significant change, that does not mean that the effect is of practical importance in biology, That is for an biologist to say. (I think this is an example of what @EricWong says in his Comment.)

Maybe you have a huge amount of data and you can claim $\mu \approx 100.5$ instead of $\mu = 100.$ The change by 0.5 might be 'real', but a biologist might say that's not an important difference. (She might say, "I'd be really interested if you told me the change is 5, but people would laugh if I wrote a paper claiming a difference of 0.5.")

If a study is wisely planned with about the right number $n$ of observations, then statistical significance and practical importance will amount to about the same thing. That takes more careful advance planning than one often sees in practice.

$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.