Lecture 11

Question 1

what does alpha value indicates (relating to H0)?

Answer 1

alpha value indicates the proportion of time we incorrectly reject the H0 when it’s in fact true

Question 2

what is the curse of multiplicity

Answer 2

when studies report the results of multiple statistical tests raising the probability that at least ONE FALSE REJECTION ERROR (reject h0 = statistically significant difference) even if there is no underlying effect.

If the critical alpha level for a single test is set at .05, this means the probability of erroneously attributing statistical significance to a result when the null is true is .05. But if two or three tests are run, the probability of achieving at least one statistically significant result rises to .10 and .14 respectively.

Question 3

2 types of alpha values

Answer 3

• per comparison alpha value

- familywise alpha value

Question 4

what is per comparison alpha value

Answer 4

value to be assigned to alpha for each NHST

Question 5

what is familywise alpha value

Answer 5

value used to control the possible false rejection

Question 6

how to address curse of multiplicity?

Answer 6

Bonferroni correction to the PER COMPARISON value

Question 7

how do we do bonferroni correction

Answer 7

1. we decide the value for familywise alpha value
2. divide this value by the number of NHSTs being undertaken

alpha per comparison = alpha family wise / k

(k= no of tests undertaken)

Question 8

“we should never make correction, unless the research is has no theory or previous guide” true or false

Answer 8

false - extreme idea

Question 9

“we should always make some kind of correction to account for the possibility of false rejection errors

Answer 9

false - extreme idea

Question 10

is family wise alpha value larger than 0.05 acceptable?

eg: 0.10 or 0.15

Answer 10

yes

Question 11

is it still acceptable if you dont have all the same per comparison alpha values?

eg: 0.03, 0.01, 0.01

Answer 11

if they sum to the same family wise alpha value then it’s OK