a significant correlation is one which
a) indicates a valid difference
b) is especially meaningful
c) can be generalized to the population
d) does not deviate from zero
Is it correct d) ?
2
-
$\begingroup$ @Chinny84 ,I thought option d) , because null hypothesis : rho = 0 (no relationship) , its rejected ...so there is relationship between two variables, so option d)?? $\endgroup$– e4e5Commented May 10, 2017 at 3:45
-
$\begingroup$ e4e5 - Thank you for providing a possible answer, I was just trying to get you to put down a rationale reason and then we can update if incorrect. Then you get more out of it than me just answering :). $\endgroup$– Chinny84Commented May 10, 2017 at 14:04
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
By a significant correlation we mean that the observed correlation in the sample is not due to chance, at that level of significance. The observed correlation in the sample is a representative of the correlation in the population. In other words, the correlation observed in the sample can be generalized to the population. Option (c) is correct.
