Can anyone plz tell me the difference between p-value & alpha in statistical significance testing.?
Question:
the problem is that even SPSS(statistical software) test the significance of correlation coefficients with the help of p-value i.e. comparing p-value with alpha of 0.05.
how this could be...if both are different...since the alpha is the type-I error and p-value is the probability of obtaining the result as extreme as observed coditioning the null hypothesis is true...
Answer:
Alpha is the percentage value that p has to "beat" for a test to be statistically significant. You normally get to choose your alpha, and the standard choice is .05.
Alpha = Type I error = probability of rejecting a true null hypothesis
Therefore we want this error as small as possible.
p-value = the probability of rejecting null hypothesis given the observed test statistics
In plain English, p-value is the smallest alpha at which null hypothesis would be rejected given the observed data.
In essence, the p-value is trying to find the probability of making type I error with the given data. Hence one can compare this p-value with the pre-determined alpha value (the maximum probability that we're willing to accept in wrongly rejecting null hypothesis).
More Questions & Answers...