cms/lib04/CA01000848/Centricity/Domain/2671/Chapter 9 Review

Formula found on pg. 503
You are not using the equation with standard
deviations 1, 2, and 3. You will use z* values
 Critical values (z*) for 8 different confidence
levels found on pg. 502
 P-hat = # said yes / # in sample
To find answer on graphing calculator: STAT,
TEST, 1 PROP Z-INT. Enter values for x (# said
yes), n (# in sample), and the confidence level.
Type 1: It’s right, but you say it’s wrong.
 (You reject, but it was true)
Type 2: It’s wrong, but you say it’s right.
 (You fail to reject, but it was false)
Ho: Null Hypothesis: ≤, ≥, and =
Ha: Alternative Hypothesis: <, >, and ≠
To decide which “tail” test a problem uses,
look at Ha:
Ha is <: It’s a left tail test
Ha is >: It’s a right tail test
Ha is ≠: It’s a two tail test
Identifying p, p-hat, and q
 p is the population proportion (the percentage
that they claim is true). This is the number that is
mentioned in the null and alternative hypotheses.
 q is the complement of the population proportion
 p-hat is the sample proportion (the percentage
that they have found during a survey)
You will need the Critical Values Table that
uses the significance value and tail test value
(found in white book on page 353…and in
your notes). This will give you the critical
value to serve as the barrier for your graph.
Shade the appropriate area(s) on your graph
based on this barrier. The shaded area is your
rejection region.
Z-score formula (found in white book on page
374…and in your notes)
This value will lie either in the “reject Ho” or
“fail to reject Ho” region.
If it lies exactly on the critical value line, you
“fail to reject Ho”.
The statement you make is in terms of the
original claim (not necessarily the null
hypothesis)…read the problem carefully!
Expected Counts – Remember how to
calculate these:
(row total * column total ) / table total

similar documents