An alternative to statistical power analysis that has much support among statisticians is the use of confidence intervals, for example the 90% confidence interval .
the outcome of a significance test is the dichotomous decision whether or not to reject the null hypothesis.
often the null hypothesis is not reject, a statement of no effect, which leaves us with very little information.
most scientists are interested in the size of the effect, which is best interpreted using confidence intervals.
therefore, even when the effect is not considered statistically significant, the data can provide useful information when interpreted using confidence intervals.