Pseudoreplication

Question 1

What type of pseudoreplication is present?

Answer 1

Temporal and satirical pseudoreplication

Question 2

What is temporal pseudoreplication?

Answer 2

Repeated measures over time treated as independent

Question 3

What is sacrificial pseudoreplication?

Answer 3

Subsamples from each experimental unit treated as true replicates

Question 4

What are the risks of including non-independent replicates in your analysis?

Answer 4

Inflate the sample size drastically, which inflates the degrees of freedom and the statistical power, and risks a type I error where the null hypothesis is rejected even if a true effect doesn’t actually exist.

Question 5

How do you deal with temporal pseudoreplication in the analysis?

Answer 5

Include time as a repeated-measures random factor

Question 6

How do you deal with sacrificial pseudoreplication?

Answer 6

Either average the subsamples within each experimental unit (less powerful) or treat the r as nested random effects

Question 7

What are the different random and fixed effects in a repeated measures ANOVA?

Answer 7

Treatment (fixed effect)
Block (random effect)
Channel nested within block (random effect)
Time (fixed repeated-measures factor)
Subsamples nested within channel (random effect)

Question 8

Why is temporal pseudoreplication bad?

Answer 8

• Correlation between observations
• share individual characteristics
• Underestimates variability and can lead to inflated type 1 errors

Question 9

Why is temporal pseudoreplication bad?

Answer 9

• Inflates the sample size
• Underestimates variability
• Violates the assumption of independence
• Correlation between individuals