407x Filetype PPTX File size 1.55 MB Source: nancepage.weebly.com
A significance test is a formal procedure
for comparing observed data with a claim
(hypothesis).
Significance The claim is a statement about a parameter.
Test Tests ask if sample data give good evidence
against a claim.
If an observed statistic is “far” away from a
hypothesized claim about a parameter, we
have some evidence that the claim is wrong.
Statistics vary from sample to sample.
A significance test answers the question, “
What are the chances we would observe a
Significance sample statistic at least this extreme,
assuming the claim about the parameter is
Test true?”
If the chances are low, there is evidence to
suggest the claim is wrong.
Null Hypothesis,
Claim about the parameter you are trying
to find evidence against.
Always a statement of equality.
Let’s assume the parameter is equal to
_____.
Hypotheses Alternative Hypothesis,
Claim about the parameter you are trying
to find evidence for.
Tests whether or not the actual value of
the parameter is greater than, less than,
or simply not equal to the value in the
null hypothesis.
The alternative hypothesis is one-sided if it
states that a parameter is larger than the
One- or null hypothesis value or if it states that the
two-sided? parameter is smaller than the null value.
If it two-sided if it states that the parameter
is different from the null hypothesis value.
Suppose you suspect a “chute” of playing
cards is not fair. The chute supposedly
contains 10 standard decks shuffled together.
You are interested in knowing whether there
are more hearts than usual. To test this, you
deal 12 cards at random and calculate the
proportion of hearts in your hand.
Example a) Describe the parameter of interest in this
setting. parameter: the true proportion of
hearts in the “chute” of cards.
b) Write the appropriate null and alternative
hypothesis for this situation.
no reviews yet
Please Login to review.