POP Sampling Flashcards
CI significance
If CI includes 1, not significant.
If CI does not include 1, significant.
Using 95% CI to determine difference between trial groups.
If the 95% CI includes 0, there is no difference between trial groups.
If the 95% CI does not include 0, this is a difference between trial groups and a p value can be calculated to work out whether this is due to chance or another factor.
Interpreting p values
• If <0.05, probability of observed difference due to chance alone is small – statistically significant [reject the null hypothesis & accept the alternative hypothesis].
• If >0.05, probability of observed difference due to chance alone is large – statistically insignificant (no evidence of an effect) [reject the alternative hypothesis & accept the null hypothesis].
Null hypothesis vs Alternative hypothesis.
Null hypothesis – no difference in the population.
Alternative hypothesis – a difference in the population.
Type I vs Type II Errors.
Type I error – no difference, but the test says there is (i.e., false positive).
Type II error – there is a difference, but the test says not (can challenge this error by increasing the population sample size) (i.e., false negative).
Negatively vs positively skewed.
Negatively (left) skewed - Mean < Median < Mode
Positively (right) skewed - Mode < Median < Mean.