hypothesis testing and experimental design Flashcards
the null hypothesis?
the hypothesis that there is no effect in the population
Sampling error
a sample that doesn’t represent the population
sampling from the wrong population = could end up with 2 samples from 2 different populations –> danger of making TYPE 1 error
OR TYPE 2 error
p value
probability that the null is true
Type 1 error
we think there is an effect when there isn’t one
Type 2 error
we think there isn’t an effect but there is one
To help account for sampling error we?
we use a probability test to indicate likelihood samples have actually come from population where null is true
-higher p value (p>0.05), the probability is high that null hypothesis is true = effect unlikely to be found in population
sample from a population where null is true, probability of finding an effect in the population is low
Null hypothesis significance testing
rejecting the null = probability there is no effect in the population is less than .05 (p<0.05)
P value assesses?
standard error assesses?
p value assesses the probability that the null hypothesis is true (and effect will not be found in population)
standard error assesses the effect in relation to estimated error in the sampling methods
- if SE is small its likely we don’t have a lot of deviation between sample and population = probability we have made a sampling error would be small(provided we sampled from the intended population)
(Small SE -> sample close to population -> unlikely sampling error has happened)