design of experiment

Question 1

what does it mean when a group is confounded

Answer 1

a group that is not comparable from the outset

Question 2

what does replication do

Answer 2

reduces varaince of estimators of treatment

Question 3

what does randomisation do

Answer 3

removes bias from comparisons and gives validity to the claim that the experiment is answering the question you set out to answer

Question 4

what is power

Answer 4

the probabilty of detecting a difference taht exists between popualtion

Question 5

what does power depend on

Answer 5

• sample size
• parameters

Question 6

what is the power most commonly used

Answer 6

80%

Question 7

what does power increase with

Answer 7

• absolute value of standardised difference

Question 8

what does the pace of power increase depend on

Answer 8

sample size

Question 9

what is a type I error (alpha)

Answer 9

when there is dignificance but not assocaition

Question 10

what is a type II error (beta)

Answer 10

when there is an assocation but no significance

Question 11

what is statistical powrer

Answer 11

statistical power= 1- beta

Question 12

what happens to SD with sample size increase

Answer 12

it becomes more accurtate

Question 13

what happens to SE with a sample size increases

Answer 13

decreases

Question 14

what would you mf-ms expect the ratio of (mf-ms)/SE with a null hypothesis to be

Answer 14

±2

Question 15

what does power depend on

Answer 15

• sample size
• population SD
• the difference between population means