chp 4 Flashcards
in most cases, the null hypothesis is something researchers hope to prove ____
false
results found as extreme as they are is measured by
p value
expecting that true groups in an experiment are identical is known as
null hypothesis
the p value proves the null hypothesis that results are due to ____ alone
chance
why do we want less than 0.05 value?
arbitrary; bc we want to REJECT null hypothesis
- say there IS a difference and not by chance
how is statistically significance measured
p value - shows there is a difference
define null hypothesis
no difference among groups and that observed differences are due to random variation (chance)
the smaller the p value the ___ the evidence is to dispute the null hypothesis
stronger
numeric representation of the degree to which random variation alone could account for the differences observed between groups
p value
when the null hypothesis is rejected, the result is said to be _____
statistically significant
when the p value is greater than 0.05, the null hypothesis can be ____
accepted
– saying it did happen by chance
the incorrect rejection of the true null hypothesis
type 1
- false positive
- saying there is a difference but actually not one
incorrectly retaining a false null hypothesis
type 2
false negative
-saying there is no difference when there IS one
t/f p values convey info regarding size
false
what is the CI if there is a large sample
narrow confidence interval
in a small sample the SEM is ___ and the CI is ____
larger; wider
a test is called ____ tailed if the null hypothesis is rejected for values of the test statistic falling into either tail of its sampling distribution
two tailed
test is ___ tailed if the null hypothesis is rejected only for the values of the test statistic falling into one specified tail
one
which tail test is easier to achieve significance, why?
one tail
- bc in two tail you have more possibility (less confident) it will fall in either side..
- -> p value is HALF the two- tailed P value
which tail test will you find a type 2 error
two tail
- more difficult to achieve significance
in a one tail test, if results went other direction than predicted this means it ___ significant
not statistical significant
which tail test should you have a prediction for direction
one tail test
one tail has a chance for type ___ error
one
which tail test is more often used?
two tail test
are two tailed test more or less conservative
more (have to divide them into two so 0.25 each side)
what is the threshold denoted by
alpha