skills test 3 Flashcards
what is the p value
the probability of getting a result as extreme or more extreme as the observed data if the null hypothesis is true
with what p values do we reject or not reject the null hypothesis
if p is larger than alpha - do not reject
if p is less than alpha - reject
if p is low, the null must go
how do we work out p values
for a sample mean, t.dist or t.dist.2t
for a sample proportion, norm.s.dist and will have to manipulate and multiply by 2 for 2 tails, use the complement rule, etc
what is a type 1 error
when a true null hypothesis is rejected
what is a type 2 error
when a false null hypothesis is not rejected
why are there errors in hypothesis testing
we never have the whole population to actually know the true population parameter. we never know if we could have just got a particularly unusual sample
what is an independent sample
This is where the selection of individuals who make up one sample is independent of the selection of individuals in the other sample.
Eg: an economic wishes to determine whether there is a difference in mean family income for households in 2 socio-economic groups.
Eg: a university wants to compare the mean NCEA results of applicants educated in rural high schools and urban high schools.
what is a paired sample
Each observation in one sample can be matched/paired in a meaningful way with a particular observation in the other sample. These are things such as “repeated measurements” or before and after samples.
Eg: Nike wants to see if there is a difference in durability of 2 shoe sole materials. One type is placed on one shoe, the other type on the other shoe of the same pair. In this scenario the same person wears one of each pair.
Eg: An analyst for Educational Testing Service wants to compare the mean test scores of students before and after taking a review course. Each student in the sample is measured twice.
The sampling distribution of the difference in sample means can be treated as normal if:
- The original populations are both normal
- If either of the populations is not normally distributed the sample size is large enough (n greater than or equal to 30) for the Central Limit Theorem to apply
for stat 101, D0 is always?
0
we also use the large sample hypothesis test for?
small sample tests where the variances of those two populations are unequal
which degrees of freedom do we use when there are two samples
the smaller of n-1
what test do we use to compare variability
F test
the F distribution is always __ skewed?
right
the test statistic for an F distribution is given by?
larger sample variance/smaller sample variance
how do we get df1 and df2 for an F distribution
the numerator of the test statistics sample size - 1 is df1
the denominator of the test statistics sample size - 1 is df2
what excel function do we use for the f test using the critical value method
F.INV.RT, divide alpha by 2 for a 2 tailed test
what excel function do we use for the f test using the p value method
F.DIST.RT, multiply this by 2 for a 2 tailed test