Dear all
I wanna apply spss in log reduction. what the best probability should i use it? either p
Azlan -
Regarding the p-value to set as a threshold, I know it has been popular to use 0.05 everywhere, but that is perhaps the biggest perpetuated mistake in the history of statistics, and one I tried to explain decades ago.
-----------
See https://www.researchgate.net/publication/262971440_Practical_Interpretation_of_Hypothesis_Tests_-_letter_to_the_editor_-_TAS
------------
Now the American Statistical Association, after a journal, I've heard, banned the use of p-values, has finally come out with a position.
Press release for the American Statistical Association:
http://www.amstat.org/newsroom/pressreleases/P-ValueStatement.pdf
The fact is that a p-value is a function of sample size, and does not include the important concept of "effect size," and needs a type II error probability or other sensitivity analysis.
So when you ask what threshold p-value to use, that depends on what kind of test you are using (what it measures and how), and on preferably a specific alternative hypothesis, your goals, and the sample size.
I am not familiar with your application, but if you could set this up as a confidence interval, they are also functions of sample size, but much easier to interpret in a practical sense, even if often technically not quite so. I don't know if you can set up what you are doing to find a result within a confidence interval, but if so, it could be far more useful for you.
Cheers - Jim
Article Practical Interpretation of Hypothesis Tests - letter to the...
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS