1
PSY207
B
Spring 2011
Psychological Statistics
VIII. Hypothesis Testing
Quick Overview
A. zTest
B. Outcomes of hypothesis testing
You should know about
•The four possible outcomes of hypothesis testing
• Type I error
• Type II error
• Alpha
•How to compute effect size and power
You should know about
•What happen when H
0
is false
• Large effect
• Small effect
• Hypothesized sampling distribution
• True sampling distribution
•What happen when H
0
is true
• Hypothesized sampling distribution
• True sampling distribution
A. zTest
• Used when you’re testing a hypothesis with the population
parameters (μ and
σ
) already known
• Steps
1. State the hypothesis (H
0
)
2. Set decision criteria (depending on H
1
)
3. Obtain sample statistics (z statistic)
4. Make a decision: retain or reject H
0
A. zTest
• Used when you’re testing a hypothesis with the population
parameters (μ and
σ
) already known
• Steps
1. State the hypothesis (H
0
)
2. Set decision criteria (depending on H
1
)
3. Obtain sample statistics (z statistic)
4. Make a decision: retain or reject H
0
E.g., For SAT, μ = 500,
σ
= 100
a. Assuming a program claims to boost SAT scores, if you ‘re going to test
this claim, what should your H
0
and H
1
be?
H
0
: μ
≤
500, H
1
: μ > 500
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document2
A. zTest
• Used when you’re testing a hypothesis with the population
parameters (μ and
σ
) already known
• Steps
1. State the hypothesis (H
0
)
2. Set decision criteria (depending on H
1
)
3. Obtain sample statistics (z statistic)
4. Make a decision: retain or reject H
0
E.g., For SAT, μ = 500,
σ
= 100
a. Assuming a program claims to boost SAT scores, if you ‘re going to test
this claim, what should your H
0
and H
1
be?
b. Assuming that you hypothesize that lack of sleep reduces SAT scores, if
you ‘re going to test this hypothesis, what should your H
0
and H
1
be?
H
0
: μ
≥
500, H
1
: μ < 500
A. zTest
• Used when you’re testing a hypothesis with the population
parameters (μ and
σ
) already known
• Steps
1. State the hypothesis (H
0
)
2. Set decision criteria (depending on H
1
)
3. Obtain sample statistics (z statistic)
4. Make a decision: retain or reject H
0
E.g., For SAT, μ = 500,
σ
= 100
a. Assuming a program claims to boost SAT scores, if you ‘re going to test
this claim, what should your H
0
and H
1
be?
b. Assuming that you hypothesize that lack of sleep reduces SAT scores, if
you ‘re going to test this hypothesis, what should your H
0
and H
1
be?
c. Assuming that you hypothesize that being excited when tested
influences SAT scores, but you’re not sure of the direction of the influence.
If you ‘re going to test this hypothesis, what should your H
0
and H
1
be?
H
0
: μ = 500, H
1
: μ
≠
500
A. zTest
• Used when you’re testing a hypothesis with the population
parameters (μ and
σ
) already known
• Steps
1. State the hypothesis (H
0
)
2. Set decision criteria (depending on H
1
)
3. Obtain sample statistics (z statistic)
4. Make a decision: retain or reject H
0
In the previous example, assuming the
level you pick is .05
a. H
0
: μ
≤
500, H
1
: μ > 500 > z
crit
= 1.65
b. H
0
: μ
≥
500, H
1
: μ < 500 > z
crit
= 1.65
c. H
0
: μ = 500, H
1
: μ
≠
500 > z
crit
= ±1.96
A. zTest
This is the end of the preview.
Sign up
to
access the rest of the document.
 Spring '07
 Pfordesher
 Statistical hypothesis testing, Statistical significance, Type I and type II errors, Statistical power, Effect size

Click to edit the document details