Type I error: Difference between revisions
imported>Doug Williamson (Standardise page appearance.) |
imported>Doug Williamson (Put 'not' in bold in addition to italics) |
||
| Line 13: | Line 13: | ||
For example in the same situation the significance level might alternatively be quantified as (1 - 5%) = 95%. | For example in the same situation the significance level might alternatively be quantified as (1 - 5%) = 95%. | ||
Meaning that there is 95% confidence that the observed result did ''not'' happen by chance alone. | Meaning that there is 95% confidence that the observed result did '''''not''''' happen by chance alone. | ||
Latest revision as of 09:17, 30 March 2016
An error that occurs in significance testing when the null hypothesis is rejected when it is actually true.
Also known as a 'false positive'.
The probability of a false positive is often known as the significance level.
For example a significance level of 5%: meaning in this case that there would only be a 5% probability that the observed result could have happened by chance alone.
However the significance level is also sometimes expressed as (1 - the probability of a false positive).
For example in the same situation the significance level might alternatively be quantified as (1 - 5%) = 95%.
Meaning that there is 95% confidence that the observed result did not happen by chance alone.
For the avoidance of doubt, it is therefore always best to be explicit about which basis the significance level is being quoted on.
