midterm 1

Question 1

benefits of multiple levels for IV

Answer 1

more precision regarding a theory, investigating non-linear effects. reduces number of experiments conducted and number of participants needed. allows for intrapolation and extrapolation

Question 2

formula for IV inflation when multiple tests

Answer 2

1-(1-a)^c, where c is the number of comparisons (tests). 0.05 is the range for acceptable type 1 error

Question 3

bonferroni correction

Answer 3

divide desired alpha by number of comparisons

Question 4

problem with bonferroni correction

Answer 4

reduced likehihood of type 1 error thus increases likelihood of type 2 error

Question 5

what test do you use when comparing a single score to the population

Answer 5

z score

Question 6

what test do you use when comparing a sample mean and standard deviation is known

Answer 6

z test

Question 7

what test do you use when comparing a sample mean and the standard deviation is unknown

Answer 7

one sample t-test (like the normal one)

Question 8

what test do you use when you have 2 levels of 1 iv

Answer 8

either independent or paired samples (for within subjects) t-test (or related samples).

Question 9

what test do you use when there are 3+ levels

Answer 9

one way or one way repeated measures anova

Question 10

what test do you use when there are more than 1 IV

Answer 10

factorial anova

Question 11

what does between group variance measure

Answer 11

systematic treatment effects, and chance effects

Question 12

what does within group variance measure

Answer 12

chance effects

Question 13

how does a larger df change the distribution of F

Answer 13

smaller critical values, i.e. less spread to the right

Question 14

what does tukey assume

Answer 14

same n for all groups

Question 15

pros of repeated anova

Answer 15

participants are their own controls, meaning more power and less error. more participants!

Question 16

cons of repeated anova

Answer 16

order effects and practice effects. may guess hypothesis. more time in lab. limits possibilities (cannot be too obvious)

Question 17

what is different from between subjects and repeated measures anova

Answer 17

mathematically removed individual difference variance from F ratio denominator

Question 18

should you include p for tukey

Answer 18

yes bitch

Question 19

how do more groups add more precision

Answer 19

more experimental, control, or placebo conditions, allows you to see if an effect is moderated or affected by another.

Question 20

requirements for factorial anova

Answer 20

2 or more ivs, 2 or more levels of each iv, quantitative DV

Question 21

why are two t tests not good instead of factorial anova

Answer 21

interaction effects (i.e. moderation)

Question 22

can you have nonmanipulated IVs for factorial design

Answer 22

yes

Question 23

mixed model factorial design

Answer 23

each participant participates in one level of the IV and in both levels of the other IV

Question 24

types of interactions

Answer 24

spreading, crossover

Question 25

spreading

Answer 25

effect exists at one level of IV, weaker or nonexistant at other level

Question 26

crossover

Answer 26

no main effects because of the little weird x you formed in the notes

Question 27

put interaction null hypothesis into words

Answer 27

the effect of factor a influences the effect of factor b

Question 28

for factorial anova rows A, B, AxB, what are these variances part of?

Answer 28

between groups. varability across group means, variability between groups

Question 29

interaction null hypothesis

Answer 29

at level b, a1 = a2

Question 30

simple effect

Answer 30

at a given level of factor A, levels of factor B differ.