Univariate Stats Test 2

Front 1

example of an aposteriori test

Back 1

Scheffe's test

Front 2

tests subsequent to ANOVA allow more _______ tests

Back 2

precise

Front 3

sample null hypotheses for tests subsequent to ANOVA

Back 3

m1+m2/m3 = 0

m1 - m2 - m3 = 0

Front 4

How many comparisons are allowed in orthogonal comparisons?

Back 4

as many comparisons as the degrees of freedom within; g-1 comparisons

Front 5

What is the most powerful test subsequent to ANOVA?

Back 5

orthogonal comparisons!

Front 6

t/f Scheffe's test is the least powerful test subsequent to ANOVA.

Back 6

true

Front 7

DF in what comparison are always 1?

Back 7

BONFERRONI, DOODOO HEAD!!

Front 8

is bonferroni mutually orthogonal?

Back 8

negative, soul sistah.

Front 9

t/f: in orthogonal comparisons, you can conduct more comparisons that DF.

Back 9

false, in bonferroni tests you can make more comparisons than DF.

Front 10

what is the familywide error rate?

Back 10

the probability that a family of conclusions will have AT LEAST one type I error.

Front 11

P(no type I error) =

Back 11

1 - alpha

(alpha = .05 unless otherwise specified)

Front 12

P(type I error) =

Back 12

alpha

Front 13

Bonferroni correction:

Back 13

alpha/K

(k is the number of intended comparisons)

Front 14

are Scheffe comparisons orthogonal?

Back 14

nope. (Ea = 0, but Eab does not = 0)

Front 15

Which test has unlimited comparisons?

Back 15

Scheffe's!

Front 16

Scheffe's familywise error rate is

Back 16

= .05

Front 17

the two range tests are ___ and ___; they are aposteriori/apriori

Back 17

newman-keul and Tukey test

APRIORI!

Front 18

how do the two range tests differ?

Back 18

Tukey test g = number of groups in experiment; Newman-Keul g = number of groups in specific comparison; changes critcal q value

Front 19

why are tukey and newman-keul called range tests?

Back 19

they order means from low to high

Front 20

familywise error rate in tukey

Back 20

= alpha, usually .05

Front 21

range tests are based on what kind of distributions?

Back 21

t distributions

Front 22

Why did Newman-Keul adapt the Tukey test?

Back 22

said means with greater differences were tested with more power than those with smaller differences

Front 23

which test is more liberal, tukey or nk?

Back 23

nk tests are more liberal, meaning there is a slightly higher chance of type I error

Front 24

btw subjects independent variables

Back 24

each subj receives only one level of an IV (positive OR neg mood, male OR female)

Front 25

wi subjects indep variables

Back 25

each subject receives all levels of an IV (pos, neg, neutral mood treatments, interacts with m and f)

Front 26

simple effect:

Back 26

measure of an independent variable at a single level of a second independent variable

Front 27

two way interaction:

Back 27

extent to which simple effects are different in direction and/or size

Front 28

confound:

Back 28

varies systematically along levels of an independent variable

Front 29

repeated measures: within subjects IV

Back 29

participants receive each level of an IV

Front 30

repeated measures: fixed effect IV

Back 30

experimenter controls which levels will be used (decides to observe happiness and sadness vs picking two moods out of a hat)

Front 31

repeated measures: random effect IV

Back 31

experimenter allows IV to vary by chance (picks two moods to compare from a hat randomly)

Front 32

are subjects a dependent or independent variable in repeated measure designs?

Back 32

independent variable

Front 33

why is subjects' variability not considered error in repeated measures designs?

Back 33

in repeated measures designs the subjects are considered an independent variable, so the amount they are measured is the fixed variable and subjects and the random effect iv.

Front 34

which is the error term in a repeated measures (within subjects) design?

Back 34

sxa interaction, because we can't explain why the scores are the way they are; the effect of A across S is not systematic

Front 35

repeated measures assumptions:

Back 35

no carry-over effects!

Front 36

how do you account for specific (known) carryover effects?

Back 36

return subjects to baseline (bring eyes back to regular light exposure)

Front 37

how do you account for general carryover effects (effects due to repeated measure)?

Back 37

give subjects breaks or practice trials

Front 38

t/f specific carryover effects are due to repeated measure

Back 38

false; general carryover effects due to repeated measure

Front 39

violations of what assumption increase type one error?

Back 39

the assumption that there will be equal measure to measure correlation in repeated measures designs

Front 40

equal measure to measure correlation:

Back 40

correlation between any pair of trials is approximately the same

Front 41

solutions to violations of equal measure to measure correlation

Back 41

1. MANOVA: treat each level as though its IVs are separate and dependent vars
2. multiply box's epsilon by any within DF, making the test more conservative

Front 42

SS in this test do not substitute for MS

Back 42

scheffe! only substitutes in orthog and bonferroni

Front 43

whats the difference between orthog comps and bonferroni?

Back 43

bonferroni acounts for a familywise error rate (alpha/k) and is more conservative as a result

Front 44

DF = g - 1 for these tests:

Back 44

orthogonal, scheffe

Front 45

df = 1 for this test

Back 45

bonferroni

Front 46

2 way btw grps ANOVA
DF:

Back 46

rows, columns, rxc interaction
Total: N-1
Btw: rc - 1
row: r - 1
col: c - 1
rxc: (r-1)(c-1)
within N - rc

Front 47

within grps ANOVA:
DF:

Back 47

A, SxA
total: N-1
btw S-1
wi: N - s
A: a - 1
AxS: (s-1)(a-1)

Front 48

in wi grps ANOVA, what do these letters stand for?
a:
S:
A:
AxS

Back 48

a: number of scores in each S
S: row total
A: column total
AxS: error within

Front 49

how do you look up critical F in SxA?

Back 49

DF for A, DF for SxA

Front 50

which f ratio do we care about in SxA?

Back 50

A over SxA

Front 51

why is unequal n a problem?

Back 51

1. confound creates a conceptual problem
2. sums of sqaures of the cells do not equal the sums of squares of the main effects and sums of squares of interactions
3. difficulty in deriving orthogonal coefficients

Front 52

box's epsilon =

Back 52

1/ t -1

Front 53

assumptions of repeated measures ANOVA

Back 53

1. there are no carryover effects between trials
2. there is equal measure to measure correlation (correlation btw any pair of trials is approximately equal)

Front 54

simple effect:

Back 54

effect of one IV at a single level of another IV (i.e. does m at positive mood affect score?)

Front 55

main effect:

Back 55

effect of one IV ignoring the others (gender ignoring mood)

Front 56

2 way interaction:

Back 56

extent to which simple effects are different in direction and/or magnitude (ex. does mood affect score for f differently than m?)

Front 57

between subjects independent variables:

Back 57

each subject receives only one level of the IV (pos OR neg)

Front 58

within subjs IV:

Back 58

each subject receives all levels of an IV (3 levels of abc)