Hypothesis Testing 2

Question 1

What is the size of the critical region determined by?

Answer 1

The level of significance

Question 2

P Value

Answer 2

Probability of obtaining a value equal to or more extreme
- NOT the critical region

Question 3

Statistical Significance

Answer 3

Whether or not a sample result led to Ho rejection at the given level of significance (a)

Question 4

Z Calc Formula

Answer 4

(Sample mean - Population Mean)
÷
(Population SD ÷ Square root of sample size)

Question 5

Effect Size

Answer 5

Measure of absolute magnitude of observed effect/difference

Question 6

What is the effect size independent of?

Answer 6

Sample size and standard error of the mean

Question 7

Cohen’s d

Answer 7

(Sample Mean - Population mean) ÷ (Population SD)

Question 8

What is the range of Cohen’s d values for small, moderate, and large effect size?

Answer 8

Small/Weak = 0.20 ≥ x

Moderate = 0.50

Large/Strong = 0.80 ≤ x

Question 9

What happens when the sample population ranges overlap each other?

Answer 9

The sample mean recorded could come from any of the populations that overlap with each other
- errors

Question 10

Type 1 Error

Answer 10

We reject Ho when it is actually found in the population

• the critical region we determined actually contains scores that fall within the common region
• conclude there is an effect, when there actually is not

Question 11

What is the probability of a type 1 error?

Answer 11

The size of the critical region
= alpha

Question 12

How do you control the probability of a type 1 error?

Answer 12

Choose a small critical region (alpha)

• the smaller the alpha, the smaller the likelihood of having type 1 error

Question 13

What is the largest alpha value that can be used?

Answer 13

.05

Question 14

True or False

A type 1 error is like convicting an innocent person in a jury trial.

Answer 14

True

Question 15

Type 2 Error

Answer 15

We fail to reject (accept) Ho when it is actually false

• Hypothesis test failed to detect that the observation falls in the critical region
• conclude there is no effect, when there actually is

Question 16

True or False

A type 2 error is like convicting a guilty person in a jury trial

Answer 16

False

Question 17

Retention Region

Answer 17

Common region

Question 18

What is the probability of the type 2 error?

Answer 18

ß (hard to calculate)

Probability of making a correct decision when Ho is false = 1-ß
(power)

ß = 1 - P correct

Question 19

What is the main method used to reduce the probability of a type 2 error?

Answer 19

Greater sample size
= power

Question 20

What power level should be aimed for when evaluating type 2 errors?

Answer 20

0.8
= 80% probability of not making a type 2 error

Question 21

What are type 1 errors considered?

Answer 21

More serious than type 2 errors
- common in research

• Type 2 can still have serious consequences
  (medicine type 2 is worse)

Question 22

As you decrease alpha , you decrease ____ but increase ___.

Answer 22

• Decrease the probability of making a type 1 error
• Increase the probability of making a type 2 error

Question 23

Power

Answer 23

1- ß