Significance

Significance is a statistical term indicating your confidence in your conclusion that a real difference exists or that a relationship actually exists, i.e., that the result is unlikely to be due simply to chance.

If your hypothesis states a direction of this difference – use a One-Tailed significance test, otherwise use a Two-Tailed significance test.

Note:   Significant does not imply important, interesting, or meaningful!

  Similarly, not all observations that are not statistically significant are unimportant, uninteresting, …

---

Transcript

Well, in professor Smith's lecture, he talked about significance.  But when statisticians use the word "significance", they don't mean "is important or not".  Statistical significance concerns your confidence that your conclusion is actually representing the real difference, i.e., the result is unlikely to be due simply to chance.  So I want a 95% confidence in this, I need to make sure that there's enough significance that (yes) this result isn't caused by chance - it's caused by a particular thing that I'm studying.  We also need to look at: Is the distribution one-sided or two-sided?  And use the appropriate test with this. That as my wife constantly reminds me, when you use the word "significant" in a statistical context, it doesn't mean it is important or interesting or meaningful.  And similarly, not all observations that are not statistically significant are unimportant or uninteresting. So, when you use the word "significant", be clear are using it to talk about statistical significance or not.