lecture 2- ai

Question 1

What is Null-hypothesis significance testing (NHST)?

Answer 1

A statistical method used to determine if there is enough evidence to reject a null hypothesis.

NHST has several known problems, including reliance on arbitrary significance levels.

Question 2

What is effect size?

Answer 2

A quantitative measure of the magnitude of a phenomenon.

Effect size helps in understanding the practical significance of research findings.

Question 3

List three uses of effect size.

Answer 3

• To compare the strength of different studies
• To determine the sample size needed for a study
• To interpret the practical significance of research results

Question 4

What does power refer to in statistical testing?

Answer 4

The probability of correctly rejecting the null hypothesis when it is false.

High power indicates a high likelihood of detecting an effect if one exists.

Question 5

What are the factors influencing power?

Answer 5

• Effect size
• Alpha level
• Sample size

Question 6

What is the importance of increasing power in a study?

Answer 6

It increases the likelihood of detecting an effect and reduces the chance of Type II errors.

A power of at least 80% is usually considered acceptable.

Question 7

What is (Partial) eta squared?

Answer 7

A measure of effect size used in ANOVA that indicates the proportion of variance explained by a factor.

It can also be used for factorial designs.

Question 8

How is classical eta squared defined?

Answer 8

The proportion of the total variance in a dependent variable that is associated with the variance of a given factor in an ANOVA model.

Question 9

What is the difference between classical eta squared and partial eta squared?

Answer 9

Classical eta squared includes all independent variables, while partial eta squared controls for the effects of other variables.

This allows for better comparisons across different experimental designs.

Question 10

What is the interpretation of eta squared values?

Answer 10

• .01 – small
• .06 – medium
• .14 – large

Question 11

Define relative risk.

Answer 11

The ratio of the probability of an event occurring in the treatment group compared to the control group.

Used primarily in medical research.

Question 12

What is an odds ratio?

Answer 12

A measure of association between exposure and an outcome, calculated as the odds of the event occurring in the treatment group divided by the odds in the control group.

Question 13

What is risk difference?

Answer 13

The difference between the proportion of the treatment group that contracts the disease and the proportion of the control group that contracts the disease.

Question 14

What are Type I and Type II errors?

Answer 14

• Type I error: Rejecting the null hypothesis when it is true
• Type II error: Accepting the null hypothesis when it is false

Question 15

When should power be computed?

Answer 15

Before the study’s data are collected.

This involves hypothesizing effect size, alpha level, and planned sample size.

Question 16

What is prospective power?

Answer 16

The power calculated before conducting a study, based on hypothesized parameters.

It helps researchers plan their studies effectively.

Question 17

What is observed power?

Answer 17

The power computed after the study is completed, assuming the sample effect size equals the population effect size.

Question 18

List methods to increase power in a study.

Answer 18

• Use fewer groups
• Use covariates
• Use a repeated measures design
• Use sensitive measures

Question 19

Fill in the blank: A power of at least _______ is usually considered acceptable.

Answer 19

80%

Question 20

True or False: A higher alpha level decreases the power of a study.

Answer 20

False