Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Research Design One > Exam One - correlation and regression > Flashcards

Exam One - correlation and regression Flashcards

1
Q

What is the goal of correlation?

A

evaluate the extent of a relationship between two continuous variables

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

What two things does the correlation coefficient tell us?

A

the strength and direction of the relationship

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

What letter is the correlation coefficient represented by?

A

lowercase r

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

______ = perfect inverse relation

A

-1

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

_______ = no relationship

A

zero

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

________ = perfect direct relationship

A

+1

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

coefficient of determination is represented by…

A

r^2

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

definition of coefficient of determination (r^2)

A

percentage of the total variance in Y scores that can be explained by X scores

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

______ r^2 values mean you can make more accurate estimates based on knowledge of one variable

A

higher

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

P value is inversely related to…

A

sample size and r value

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

magnitude of r relevance scale

A

0-0.25 = little to no relation
.25-.5 = fair relation
.5-.75 moderate/good
>.75 good/excellent

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

correlations (should/should not) be interpreted as being important if p-value is <0.05

A

should not

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

t or f? correlation determines causation

A

FALSE

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

coefficient r measures a linear or curvilinear relationship?

A

linear

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

values that are correlated are not…

A

necessarily similar

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

what is the goal of regression?

A

prediction

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

definition of regression

A

measurement variables used to develop an equation to predict one based on the other

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

in linear regression equation X is the ….

A

independent or predictor variable

19
Q

in linear regression equation Y is the…

A

dependent variable or criterion variable

20
Q

distance between the regression line and data point are called…

A

residuals

21
Q

residuals are the…

A

extent of error for each observation relative to the repression line (prediction)

22
Q

in equation Y = 64.72 + 1.39 (x) what is the dependent variable?

A

y

23
Q

in equation Y = 64.72 + 1.39 (x) what is the independent variable?

A

x

24
Q

in equation Y = 64.72 + 1.39 (x) what is the y-intercept?

A

64.72

25
Q

in equation Y = 64.72 + 1.39 (x) what is the slope?

A

1.39

26
Q

What units does r^2 use?

A

no units, just % of variability

27
Q

_____ r^2 implies a more accurate prediction equation

A

higher

28
Q

SEE

A

standard error of the estimate - magnitude of the expected error in Y based on the predictors

29
Q

what units does SEE use?

A

the units of Y

30
Q

_______ SEE implies a more accurate prediction equation

A

lower

31
Q

95% CI of the prediction =

A

predicted value +/- (1.96*SEE)

32
Q

multiple linear regression

A

method to predict the criterion variable based on more than one predictor variable

33
Q

What two methods are possible for multiple linear regression?

A
  • “enter” method
  • forward, backward, or stepwise methods
34
Q

enter method of multiple linear regression

A

this will use all variables in the prediction equation whether they make a meaningful contribution or not

35
Q

forward, backward, or stepwise methods of linear regression

A

uses statistical criteria to max prediction accuracy using the fewest possible variables

36
Q

_______- people per predictor variable are necessary when using multiple linear regression

A

7-15 (liberal)

37
Q

The accuracy of multiple linear regression is evaluated by…

A

the coefficient of determination r^2 and the SEE

38
Q

r^2

A

proportion of the variability in Y accounted for by all the predictors

39
Q

SEE

A

expected error in our prediction of Y, expressed in units of Y

40
Q

logistic regression

A

similar to linear regression, predicted variables are evaluated for the extent they accurately predict a CATEGORICAL outcomee

41
Q

1 predictor: ____ logistic regresion

A

simple

42
Q

> 1 predictor: _______ logistic regression

A

multiple

43
Q

use of logistic regresion?

A

predicting categorical clinical outcomes

Research Design One (20 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions