Alternative Hypothesis
Decision rule
- Tells us precisely when H0 should be rejected
- Species when a sample mean is a rare outcome
Decision Rule
' For our electric IQ example:
-If the z score for the sample mean is greater than 1.96 or
less than -1.96, r
If He is false and there is a LARGE effect. . .
* The hypothesized and true distributions don't overlap much
lf H{3 is false and there is 3 LARGE effect . . .
* The probability of a Type ll error is small
*- Type ll error
-Retaining a false null hypothes
Two-Tailed Test
- Directional test
- Uppertail critical; lower-tail critical
One-Tailed Test
' Upper tail critical: *- For a one-tailed test, if the sample mean falls in the direction
that you predict (Upper or lower tail),
-the probability that you rejec
_.e.' I? GU
W5? ~00?
Third Property. Sample size and the Standard Error lav-5
- The larger the sample snze, the smaller the standard error of
the mean i
batik9"- Q ak Example Gt ti 1 9: ML: 51 m.
- Population standard deviation = 100 {Tm n: "31 *1: '
-r:r"
3|
'i
i
ll
1
E.
r=
. .-=:.
uhJ" _ .-
M. 1 a
1-." I ' -" I3|-
.- : ' I J
IL. .I'" .
f.
-w v
,. "h
F. a!" .-'
I i. I
i f .I':
"E ith
99 Variance and Standard Deviation
'3 . 1"
for Sample:
variability _ _ , I . ' -
-Sample _ variance and standard
lity of selecting someone with an IQ below
' What is the probabi Probability of an M below 132
- 5. Determine probability below 132:
.9772
Example 2
- What percentage of people have le below 91?
' i. C)
Percentage of I05 below 91 l U
* 1 Sketch normal GLA
Sampling Distributions
You should know everything there is to know about . . .
- Sampling distributions
Mean of sampling distribution
Sample size
*Central limit theorem
Inferential Statistics
- Experimental method in the behavioral sciences
-Make a hypoth
-+1 on tax bls
- If we had compared raw scores, we might have concluded
that Sen. Klott voted consistently conservativer
- Review: What can 2 scores do for us?
- 1. The z score better describes the place of the raw score in occu rre nce were innite
Probab