Chapter 7/8 - Hypothesis Testing

Question 1

The average on the test was 27.93/42 (66.5%) with a standard deviation of 5.8 points (the median was 69%).

Answer 1

Test Midterm #1

Question 2

where to find z-scores not at exactly 0 or -1 or -2 or +1 or +2.
The in-between scores!

Answer 2

Question 3

EXAMPLE #1:

Answer 3

EXAMPLE #2

Question 4

the 1-tail and 2-tail hypothesis testing:

Answer 4

Explaining why we need to do Hypothesis Testing in the first place.

Question 5

Answer 5

Question 6

The ASSUMPTIONS:

Answer 6

THE 6 STEPS for doing a Hypothesis testing:

Question 7

STEP 1: ASSUMPTIONS, POP., SAMPLE?

Answer 7

STEP 2: STATE THE HYPOTHESIS

Question 8

STEP 3: MEAN, STANDARD DEVIATION

Answer 8

STEP 4 : FIGURE OUT CRITICAL VALUES

Question 9

STEP 4: 2-TAILED HYPOTHESIS

Answer 9

STEP 4: 1-TAILED HYPOTHESIS

Question 10

STEP 5: CALCULATE Z-SCORE

Answer 10

STEP 6: DECISION

Question 11

Ex #2 : Z-TESTING

Answer 11

Ex #2 :Step 1

Question 12

Ex #2 - Step 2

Answer 12

Ex #2 - Step 3

Question 13

Ex #2 : Step 4

Answer 13

Ex #2: Step 5

Question 14

Ex #2 : Step 6

Question 15

P-HACKING:

Manipulating data or analyses to artificially get significant p-values.

WHEN analyses are being chosen based on what makes the p-value significant, not what’s the best analysis plan.

Answer 15

• to calculate the p-value, we look at the Null Hypothesis
• In the Null Hypothesis Significance Testing (NHST) Framework, we either reject or fail to reject the Null.
• This is a Binary Decision. This leads to 4 possible scenarios:

Question 16

Confidence Intervals:

Answer 16

Question 17

Answer 17

Question 18

Answer 18

Question 19

The range of scores contained in a 90% confidence interval will THIS the range of scores contained in a 95% confidence interval.

a) larger than
b) smaller than
c) the same size as
d) smaller than or the same size as

Answer 19

B) smaller than

Question 20

Answer 20

Question 21

Answer 21

Question 22

Answer 22

What if we could remove the sample size from the equation?
Yes, this is possible with Cohen’s d estimating the effect size

What it measures: How many standard deviations above or below the Null Hypothesis your value lies.

Helpful because it allows standardization across studies. Since the value will be the same no matter how many participants are in the study.

Question 23

Remember, Cohen’s

Answer 23

Question 24

Because effect size is not affected by the sample size, there are only 2 things that will affect the value

1. Difference between the MEANS

Answer 24

2. Effect size depends on the VARIABILITY in the populations

Question 25

Answer 25

Question 26

QQQQQQQ- what does this slide mean?

Answer 26

Question 27

Why use STATISTICAL POWER?

Let’s say there’s a new drug and the pharm company wants to know if there is a difference with the drug (will it improve symptoms).

What is the probability that I will reject the Null Hypothesis (if my theory is right and symptoms improve)?

to answer this question: look at the statistical power, which is the likelihood we will reject the Null if the null hypothesis is false

Answer 27

Question 28

Answer 28

Question 29

1 of 5 factors that INCREASE POWER:

Answer 29

2 of 5 factors that INCREASE POWER:

Question 30

3 & 4 factors that INCREASE POWER:

Answer 30

5 of 5 factors that INCREASE POWER:

Question 31

Answer 31