exam 4 Flashcards

(39 cards)

1
Q

predictor

A

1st variable

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

pearson’s r

A

correlation coefficient used for interval/ratio data in linear relationships

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

criterion

A

2nd variable

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

correlation coefficient range

A

-1 to +1 (0=none)

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

correlation coefficient guidelines

A

-.20-+.20 = weak
.20-50 = moderate
.50+ = strong

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

partial correlations

A

measure relationship between two variables while controlling a third variable by holding it constant

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

coefficient of determination (R^2)

A

measure accuracy of correlational predictions (how much variability can be explained by predictor)

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

R^2 guidelines

A

.01 = small effect
.09 = medium effect
.25 = large effect

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

regression

A

allows us to make predictions using the line of best fit

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

Y = bX + a

A

b = slope
a = y-intercept

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

the error of prediction

A

Y - Ŷ

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

total squared error

A

sum(Y − Ŷ)2

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

best fitting line has the smallest…

A

total squared error (least squared error solution)

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

regression test hypotheses are always

A

nondirectional

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

regression test generic hypothesis

A

the equation using X does/does not account for a significant proportion of variance in the Y scores

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

point always falling on regression line

A

(𝑋bar, Ybar)

16
Q

multiple regression

A

regression with 2 or more predictor variables for the single criterion variable

17
Q

beta/standardized b coefficients

A

can be compared, larger for one variable means it predicts more variance of criterion

18
Q

spearman rho coefficient used when

A

both variables measured on ordinal scale
or interval/ratio nonlinear

19
Q

point-biserial coefficient used when

A

one variable continuous (interval/ratio) and other two-level (nominal)

20
Q

phi coefficient Φ used when

A

association between 2 dichotomous variables (nominal)

21
Q

chi square x^2 used when

A

two nominal variables but with dichotomous or multiple group variables
can determine if one variable is independent of other categorical variable
determines if random or by chance

22
Q

contingency table

A

chi square table for frequency in each category

23
Q

NOIR

A

nominal - categories
ordinal - categorize + rank, distance between each is different
interval - rank + equal distances between scores
ratio - rank, equal distance, + true zero

24
use mode when...
categorical data
25
use mean when...
interval and ratio data, symmetrical distributions
26
use median when...
ordinal data or skewed interval/ratio
27
measures of variability
range, variance, standard deviation (most common)
28
central limit theorem
reasonably large sample size (n=30) --> sampling distribution normally distributed
29
type I error (α)
reject null when null is true
30
type II error (β)
fail to reject null when H1 is true
31
omnibus test
looks at effect overall (F), need to conduct post hoc test to look at which levels of IV are differing
32
use z test....
sample of population, mean, and std dev
33
use single sample t test....
sample mean and population mean, no std dev, <30
34
independent 2 sample t test...
random assignment
35
dependent 2 sample t test...
1 sample measured twice, or 2 related samples
36
pearson's r APA results
pearson's r revealed significant relationship btwn .... r (n = #) = #, p = #, x-tailed, R^2 = #. Greater __ were associated with greater __.
37
how to find b + a from spss
b = unstandardized for variable a = unstandardized for constant
38
regression line APA results
the equation using X was/was not found to be a significant predictor of Y. F (df, df) = #, p = #, R^2 = #. More x predicted more y.