Lec 7 Flashcards

1
Q

what tests look for relationships/associations?

A

bivariate correlation
interclass correlation coefficient regression
internal consistency
kappa
factor analysis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

term? items in a test or survey are consistent with each other

A

internal consistency

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

inter vs intra rater reliability

A

inter - different raters
intra - same rater

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

type of validity?
a score on one outcome measure can predict another outcome measure

A

predictive

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

type of validity?
2 different outcome measures get similar results

A

concurrent

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

type of validity?
the gold standard measure and another outcome measure get similar results

A

criterion

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

type of validity?
subgroups of items in an outcome measure explain factors of the same complex construct (ex: satisfaction, coordination)

A

construct

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

what stat is commonly used in methodological research?

A

Pearson correlation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

what type of data can be used for Pearson correlation?

A

interval/ratio

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

what does the r stat measure?

A

relationship b/w 2 variables and how strong it is

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

r varies from ____ to _____

A

+1 to -1
(0 = no correlation)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

T/F: Pearson r predicts the slope

A

F!!!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

r=___ is a perfect correlation

A

+/- 1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Pearson correlation is very sensitive to ____
what does this mean?

A

number of participants
doesn’t take many participants to get a significant result

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

an r of ____ is a small effect

A

+/- 0.1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

an r of ____ is a medium effect

A

+/- 0.3

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

an r of ____ is a large effect

A

+/- 0.5

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

what is the coefficient of determination?

A


how much of the variability in one variable can be predicted by the other variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

T/F: statistical significance does not always imply a meaningful relationship (r)

A

T!!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

T/F: if there is a perfect correlation, there will always be a perfect agreement

A

F!!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

T/F: just because there is a significant correlation does not mean there is a cause an effect relationship

A

T

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

what is point-biserial correlation?

A

rpb
run b/w 2 levels of a categorical variable (ex: 2 groups) and in I/R level variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

T/F: a point-biserial correlation is run exactly the same as a Pearson correlation in SPSS

A

T

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

what is the non-parametric equivalent to the Pearson correlation?

A

Spearman coefficient (ρ or rs)

24
how is ordinal data different from I/R?
ordinal data does NOT have even interval between the values
25
the larger the difference in ranked scores in a Spearman correlation, the ____ the association
smaller
26
the Spearman correlation value (rs) is calculated from the _____
summed D^2 = difference b/w scores squared
27
T/F: D^2 of the Spearman correlation is reported in the manuscript
F!
28
what is internal consistency?
how closely related the items in an outcomes measure are as a group
29
what is Cronbach's α? how is it calculated?
internal consistency statistic calculated from averaging all the possible pairwise correlations b/w items; also takes into account number of items and variance
29
Cronbach's α range
0-1
30
what Cronbach's α is an acceptable value?
0.7
31
what Cronbach's α is a good value?
0.8
32
what Cronbach's α is an excellent value?
0.9
33
an internal consistency score of closer to 1 indicates ____
pair wise correlations are similar
34
what is the measure of internal consistency?
Cronbach's alpha (α)
35
what is the standard error of the mean (SEM)?
the average distance between the sample mean (M) and the population mean (μ)
36
what is the equation for the standard error of the mean (SEM)?
σ/√n
37
what is the standard error of measurement?
test reliability how much error expected from a measurement
38
what is the equation for standard error of measurement?
SEm = SD√(1-r) r = test-retest reliability
39
why is standard error of measurement important?
need to determine is change in outcome measure scores are a true change or if it is just error/chance
40
what is interclass correlation coefficient (ICC)?
reliability coefficient (v. flexible) measure of agreement association
41
interclass correlation coefficient (ICC) is most often used to calculate _____
inter-rater
42
interclass correlation coefficient (ICC) ranges from
0-1 1 = most consistent
43
difference b/w Pearson and interclass correlation coefficient (ICC)
Pearson - only 2 sets of scores; measures association NOT agreement ICC - many sets of score; association AND agreement measured
44
ICC(2,1) what is the form? what is the model?
form = 1 model = 2
45
what is the more conservative model of ICC (ICC will be lower)?
2
46
model of ICC? each participant is assessed by a different set of randomly selected raters (rare)
1
47
model of ICC? each participant is assessed by each rater, and raters have been randomly selected and represent all similar raters (common)
2
48
model of ICC? each participant is assessed by each rater , but the raters are the only raters of interest
3
49
what does the form number for ICC depends on?
number of measurements (ex: 1 = single measurement/1 measurement from each rater, etc.)
50
SPSS name for model 1 ICC
one-way random
51
SPSS name for model 2 ICC
two-way random
52
SPSS name for model 3 ICC
two-way mixed
53
Pearson correlations can assess ___ raters
only 2
54
what does G-theory assess?
analyzes reliability used with unlimited factors (facets) and their contributions to the error of a measure
55
what does G-theory estimate?
relative contribution of identified error sources to the total measurement error
56
T/F: overall generalizability coefficient (G) is reported in articles.
T
57
how should the overall generalizability coefficient (G) be interpreted?
as an r closer to 1 = better