Lecture 11: Bayesian Inference (Alternative)

Question 1

What is the main goal of hypothesis testing in psychology?

Answer 1

To explain variation in behavioural data.

Question 2

What are the two sources of variation in behavioural data?

Answer 2

Systematic and random variation.

Question 3

What does systematic variation result from?

Answer 3

Manipulated or measured variables.

Question 4

What does random variation result from?

Answer 4

Chance or unmeasured variables.

Question 5

What does NHST stand for?

Answer 5

Null Hypothesis Significance Testing.

Question 6

What probability does NHST calculate?

Answer 6

The probability of the data assuming the null hypothesis is true.

Question 7

What is the common threshold for rejecting the null hypothesis in NHST?

Answer 7

p < .05.

Question 8

What is the p-value in NHST?

Answer 8

The probability of the observed (or more extreme) data if H₀ is true.

Question 9

Which form of reasoning does NHST rely on?

Answer 9

Modus tollens.

Question 10

Why is modus tollens problematic in NHST?

Answer 10

It is invalid in probabilistic reasoning.

Question 11

Does a low p-value logically prove the null hypothesis is false?

Answer 11

No.

Question 12

What key question does NHST not answer?

Answer 12

What is the probability that the null hypothesis is true given the data?

Question 13

What do researchers often wrongly infer from a low p-value?

Answer 13

That the alternative hypothesis is true.

Question 14

Does NHST test the alternative hypothesis?

Answer 14

No.

Question 15

What does the HIV test example illustrate?

Answer 15

The importance of base rates in interpreting evidence.

Question 16

What Bayesian concept is shown by the HIV testing example?

Answer 16

Even accurate data can be misleading without considering prior probabilities.

Question 17

What theorem is used in Bayesian inference?

Answer 17

Bayes’ Theorem.

Question 18

What two elements are combined in Bayesian inference?

Answer 18

Prior beliefs and data likelihoods.

Question 19

What does Bayesian inference calculate?

Answer 19

The posterior probability of a hypothesis given the data.

Question 20

What does the coin toss example illustrate?

Answer 20

Bayesian updating of beliefs.

Question 21

What is the starting point in Bayesian analysis?

Answer 21

Prior probabilities.

Question 22

What do we get after updating with new data in Bayesian analysis?

Answer 22

Posterior probabilities.

Question 23

What do Bayes Factors compare?

Answer 23

The relative support for H₀ and H₁.

Question 24

What advantage do Bayes Factors have over p-values?

Answer 24

They provide graded, not binary, evidence.

Question 25

What is a credible interval?

Answer 25

A range of plausible parameter values given the data.

Question 26

What does a credible interval reflect?

Answer 26

Posterior belief.

Question 27

What advantage does Bayesian inference offer over NHST?

Answer 27

It allows testing a range of possible effect sizes.

Question 28

What must researchers justify when using Bayesian methods?

Answer 28

Their choice of priors.

Question 29

What does the subjectivity in priors require from researchers?

Answer 29

Transparency and theoretical justification.

Question 30

What two developments have made Bayesian methods more practical?

Answer 30

Advances in computing power and analytic tools.

Question 31

Why should psychology students learn Bayesian methods?

Answer 31

To understand modern literature and contribute to the field.