Experimental Design - Parkinson

Question 1

What ARRIVE guidelines relate to reduction?

Answer 1

• Study design
• Sample size
• Allocating animals to experimental groups
• Experimental outcomes
• Statistical methods
• EDA

Question 2

What are the minefields in experimental design?

Answer 2

• Bias
• False positives, negatives, and power
• Incorrect specification of the experimental unit
• Confounding

Question 3

How to control against bias?

Answer 3

Randomization, blinding of assay to treatment, blocking

Question 4

What are false positives?

Answer 4

The change you’ll see a significant result by chance, specified by significance level, p=0.05

Question 5

What are false negatives?

Answer 5

When we carry out an experiment and don’t see an effect. Power = 1-false negatives = 0.9/0.8

Question 6

What is the relationship between false positives and negatives

Answer 6

Up power = Up significance e.g. Power = 0.5, sig = 0.05, power = 0.8, sig = 0.01

Question 7

What is the experimental unit?

Answer 7

The smallest unit which can be independently allocated to a treatment = the replicate

Question 8

What is confounding?

Answer 8

Differences in your experiment are due to something other than your treatment (asymmetrically = bias, symmetrically = increase error variance and lead to bigger sample size needed?)

Question 9

What were the problems in the height example?

Answer 9

Sex confounding nationality, only 1/2 population, not significant

Question 10

How did they fix the problems in the height example?

Answer 10

Blocking gender and using ANOVA

Question 11

Pros and Cons of using homogeneity to control for confounding?

Answer 11

+ Removes a source of confounding
+ Reduce variability
- Reduce scope
- Reduce sample size

Question 12

Pros and cons of using block/factoring to control for confounding?

Answer 12

+ Factors out confounding source
+ Reduce variability
+ Maintain scop
+ Maximize sample size
+ Can identify interactions

Question 13

When do you not integrate sex into research?

Answer 13

• Sex-specific effects E.g. ovarian cancer
  -Different M/F models E.g. Lupus (F), kidney damage induced hypertension (M)

Question 14

How do you integrate sex into research designs

Answer 14

Ensure model works in M and F, and use ANOVA

Question 15

What are the uses of pilot studies?

Answer 15

Optimize treatment (timing, drug conc.), Obtaining treaetment effect and variability, Streamline procedures

Question 16

How to control variability?

Answer 16

Make environmental conditions homogenous,
Block/Factor variability (affects sample size = block, affect scope/suspect interaction = factor)
Randomize the rest

Question 17

How to identify sources of variability?

Answer 17

Raw material, Experimentation, Analysis

Question 18

What does blocking do?

Answer 18

Systematically removes variation from error thereby increasing power
ONLY PARAMETERS NOT INTERESTED IN E.g. batches of animals

Question 19

How to use blocking?

Answer 19

Structure as mini experiment w/ at least 1 replicate of each treatment
W/I block keep conditions homogenous E.g Researcher 1/Batch 1/Day 1 and Researcher 2/Batch2/Day2

Question 20

How to use sequential experiments?

Answer 20

Block and analyze experiment after every block

Question 21

Advantages and disadvantages of sequential experiments?

Answer 21

+ Highly significant effects detected early
+ Unsuccessful treatments stopped early
+ 20-30% saving animals
-Needs Bonferroni correction = reduce power
- Complicated design => statistician

Question 22

How does factoring work?

Answer 22

Systematically removes variation from error in parameters you’re interested in = increases power
Quantifies individual treatment effects and interactions

Question 23

How to use repeated measures?

Answer 23

> 1 treatment on same animal
- less variation
- Non-destructive no carry-over between treatments
- other eye, skin, time series

Question 24

How do you deal with multiple comparisons?

Answer 24

Minimize number of comparisons => avoid omnibus testing, select comparisons carefully, use ANOVA

Question 25

What are the most efficient tests for dealing with multiple comparisons?

Answer 25

Tukey’s HSD, Dunnet’s test, Hsu’s MCB