Brush up M&T - studieboek - Design of experiment
13 important questions on Brush up M&T - studieboek - Design of experiment
What is the benefit of randomization?
--> does not eliminate noise!
How can noise be reduced?
- measure individuals twice (before and after) --> paired t-test
- controlling conditions
- precise measurements
- matching similar samples (blocking, cross-over etc)
When is a paired t-test used?
- Higher grades + faster learning
- Never study anything twice
- 100% sure, 100% understanding
What is a cross-over design?
--> variation between animals is eliminated
--> not always possible
--> two groups, A, then B and B, then A
--> carry over can occur (therefor, include wash out period)
--> season effects can occur (timepoint is relevant)
What is a randomized block design?
- used if confounders are observed
- two groups are made based on the blocked variable (female/male) and within the groups there will be randomization)
- blocking should also be used in analysis
What are factorial designs?
- naive: subjects exposed to all combinations
--> expert should be consulted
Why don't we maximize sample size always?
- animal suffering or death (ethics)
- very small differences not interesting (relevance)
What are the standard values tgaken for alpha and ß?
ß = 0.1 or 0.2
Power: pi = 1-ß (0.9vs. 0.8)
How can errors be controlled?
- every statistical test controls alpha correctly
- how is ß controlled?
- ß is relevant only when there is a difference in the means
- the larger the sample size, the smaller ß
What is the formula to calculate the sample size for a t-test?
Z1-ß = z1-0.2 or z1-0.1 = 0.84 or 1.28
Delta = difference between groups
n = sample size
What is the equation for the sample size for a paired t-test?
What is the equation for the sample size for a x2 test?
pi2 = event frequency in group 2
pi mean = (pi1+pi2) / 2
What is the difference between a type I and type II error?
type 2 = error controlled by sample size
Errors are alpha and ß
The question on the page originate from the summary of the following study material:
- A unique study and practice tool
- Never study anything twice again
- Get the grades you hope for
- 100% sure, 100% understanding
![Trustpilot Logo](/images/design_2020/lp/clp-trustpilot-en.png)