Lecture 1_Review

Question 1

What is our intent when applying MLR?

Answer 1

to explain (or predict) individual differences in an outcome variable (DV), by linking them to individual differences on one or more predictor variables (IVs)

Question 2

The linkage between differences in an outcome variable (DV), and differences in one or more predictor variables (IVs) is centered on what core factor?

Answer 2

the covariability between the variables

Question 3

What are 3 measures of variability?

Answer 3

• Sum of squared deviations (SS).
• Variance
• Standard Deviation

Question 4

What is the Sum of Squares (SS)?

Answer 4

The basic measure of variability.
= the sum of the squared deviations of the individual scores from the mean of the scores.
= ∑ (Xi - X̅̅)^2

Question 5

What is Variance?

Answer 5

• the average squared deviation of the scores from their mean.
= SS / N (sample), or SS / N-1 (population estimate)
= ∑ (Xi - X̅̅)^2 / N

Question 6

What is Standard Deviation

Answer 6

• “the amount, on average, that scores differ from their mean”
= the square root of the variance
= [∑ (Xi - X̅̅)^2 / N]^(1/2)

Question 7

Type I error

Answer 7

claiming there is an effect/ difference/ association when there is not one (a false positive).
• probability = α(alpha).

Question 8

Type II error

Answer 8

overlooking an effect/ difference/ association that is really there (a false negative).
• probability = β(beta).

Question 9

Power

Answer 9

the probability of finding an effect/ difference/ association if it is there to be found.
• probability = (1 −β).

Question 10

α (alpha)

Answer 10

p(being wrong if we decide to reject H0 | H0 is true)

≤ 0.05 (by convention)

Question 11

β (beta)

Answer 11

p(being wrong if we decide to retain H0 | H0 is false)

≤ 0.20 (by convention)

Question 12

Power (1 –β)

Answer 12

p(being right if we decide to reject H0 | H0 is false)

Question 13

(1 –α)

Answer 13

p(being right if we decide to retain H0 | H0 is true)

Question 14

What do the probabilities associated with hypothesis testing apply to?

Answer 14

These probabilities are the probabilities about our decisions under different assumptions.
• Not the probabilities that H0 is true or false

Question 15

treatment effects

Answer 15

the source of variance occurring between treatment groups leading to H1 being true

Question 16

Error Variance

Answer 16

variance due to unsystematic factors (occurs both within and between groups)

Question 17

F statistic

Answer 17

= (Between group variability) / (within group variability)
= [(error variance) + (treatment effects)] / (error variance)
F = 1 [if H0 is true]
F > 1 [if H0 false]

Question 18

When to reject H0?

Answer 18

If F exceeds a critical value (based dfs, and for specific values of α)
• chance that we are wrong is ≤ α.