Regression Flashcards

1
Q

covariance

A

how 2 variables covary with each other. uses raw score units. unable to tell strength this way. s

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

correlation

A

scale-free degree of linear relationship between 2 varibles. r

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

scatterplot

A

strength of association, direction of it, shape of it

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

pearson correlation

A

magnittude and direction of linear relationship

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

change in magnitude

A

outliers, extreme #’s inflating the mean, curvlinear, errori in x or y

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

correlation does not equal

A

causality

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

correlation coefficient

A

strength of relationship between 2 variables. r

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

Y=mx + b in regression form

A

Y= a +bx, Y=bo + b1x

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

bo

A

regression constant, intercept

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

b1

A

regression coefficient, slope

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

when does regression line pass through y axis?

A

when x=0

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

regression line always passes through

A

x bar, y bar

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

means of predicted y

A

= means of observed y

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

Error

A

actual - predicted Y

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

least squares criterion

A

slope and intercept to minimal distance between actual and predicted y. decreases sum of square residuals

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

improves ability to predict Y from using predictors (x)

A

regression

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

R

A

corrleation betwen observed and predicted = absolute value of correlation

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

Rsquared

A

proportion of variance in Y that is accounted for in linear relationship in x. biased, overestimates population

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

adjusted R squared

A

unbiased estimate of population

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

stand error orf estimate

A

error present in predicting y from x. decrased SEE is more accurate

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

constant unstandardized B

A

Regression constant, bo, intercept

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

underneath the constant in unstandardized B

A

slope, regression coefficient

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

unstandardized coeff

A

for every unit increase in x, there is a [ ] increase in y

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

R2 = .43

A

approxing 43% of the variance in y is accounted for by its linear relationship with x

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

for every 1 unit increase in x, y increases by [what factor]

A

slope

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

multiple regression

A

looking at multiple predictors

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

1st order model

A

x has linear relationship with y and does not interact but can correlate

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

holding, controlling, partialling out

A

studing x1 on y, but x2 can affect that relationship (can be corr with x1, y or both, so we remove the effect of x2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

effect size R^2

A

proportion of y that is accounted for in model

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

F test tells what is significantly different than zero

A

R, Rsquared

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

T test tells what is significantly different than zero

A

regression coeffecition (slope) when controlling for effects of other variables

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

APprox [ ] percent of the variance in y is accounted for in its linear relationship with x

A

adjusted r^2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
33
Q

a 1 unit increase in x1 is associated with a [-] in loans

A

decrease [slope]

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
34
Q

unique contributors

A

looking at variance over one predictor over and above another

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
35
Q

holding x2 constant, the predicited y increases by [. ] for every aditional unit of x1

A

positive slope

36
Q

can pull regression line one way or another

A

outliers

37
Q

cooks, leverage,

A

looks at outliers. >2 needs to delete

38
Q

partial correlation

A

removing the effects of 1 or more varianbles (x2) from both variables (x1, y)

39
Q

how to know variance with part or partial

A

square

40
Q

part (semi-partial) correlation

A

removing the effects of one variable (x2) from 1 varianble (x1) but not the other (y)

41
Q

variance accounted for in. x2, over and above the variance accounted for in x1

A

part

42
Q

variables need to be controlled by

A

theory

43
Q

common cause hypothesis

A

if a and b correlate, they share a common cause. Partialling out the effects of the common cause variable will yeild zero

44
Q

common cause example

A

not wearing seatbelts in 80’s caused astronaughts to die in space

45
Q

mediation

A

if a and b correlate, b/c a causes b through 1 ore more mediator variables, then the correlation of a & b partional out the effects of. the mediator should equal zero

46
Q

mediator efect on significance

A

can drastically decrease or make not significant

47
Q

medeator causes what

A

outcome and explains underyling mechanism of relationshp

48
Q

mediator example

A

grades cause happiness. grades lead to greater self esteem which causes happiness

49
Q

moderation

A

when relationship of 2 variables depends on another in order to have an effect

50
Q

moderator causes what

A

significant increase or decrease effect of x on y

51
Q

moderator example

A

pscyhotherapy decreases depresion fro men more than women. gender is moderator

52
Q

suppressor variables

A

uncorrelated w/ y but highly correlated with x causing an artificial improvement in x and y when x2 has no bearing on the relationship

53
Q

suppressor reg coeff=

A

non zero

54
Q

standard regression coeff for x1 is ___ correlated with criterorio

A

greater

55
Q

x1 < x2(y) who is suppressor?

A

x1

56
Q

spss suprressor. what will be greater than zero order?

A

Part

57
Q

Correlation assumptions

A

x and y linear relationship, data pair independence (x 1 does not relate to x2), bivariate normality

58
Q

correlation assumptions modifications

A

check scatterplot. should look evenly dispersed, random, no patterns

59
Q

Regression assumptions

A

normality of residuals, homoskedasticity of residuals, model properly specified

60
Q

sample size for regression

A

5(p) bare minimum, 20(p) good

61
Q

Regression modifications

A

homo-plots, normality-histogram

62
Q

model properly specified

A

no unnessary x’s, all important x’s in model, lineary between x and y

63
Q

multicollinearity

A

high correlation between 2 ore more IV’s (usually measures of similar constructs)

64
Q

multicollinearity can cause

A

unstable coefficients, large changes in splope and sign changes, increased stand error. sig F test, but not sig t test

65
Q

Detect multicollinarity

A

VIF & tol. VIF >3, tol

66
Q

fixing multicollinarity

A

combine x’s into one, drope 1 or more IV’s, collect more data

67
Q

Categorial values

A

need coding

68
Q

Different types of coding

A

dummy, non weighted, weighted, contrast

69
Q

What will change with coding

A

regression coeff. ANOVA output is the same

70
Q

Curvlinear

A

1st order linear, 2nd quadratic, 3rd cubic

71
Q

Determining curvlinear

A

a priori, scatterplot, problems in residual plot

72
Q

centering

A

creates deviation around x & makes it meaninful an dmore interpreatable, decreases non essential multicollinearity

73
Q

scaling

A

leads to non essential multicollinearity

74
Q

enhancing interactions

A

strengthens relationship

75
Q

buffering,

A

weakens or cancels out relationsihp

76
Q

Buffering canceling out

A

life statisfaction= increased job stress, decreased marital problems

77
Q

Logistic regression

A

predicting dichotomous outcomes

78
Q

probability

A

uses percentages of something happening

79
Q

odds

A

ratio of something NOT happening 1:2. (2 x’s greater it won’t happen)

80
Q

logits

A

transfromed probabilty, linear

81
Q

prob

A

negative logit

82
Q

Prob > .5, odds >1

A

positive logit

83
Q

Simultaneous regression

A

IV’s treated simultaneous on equal footing. leaves every variable with squared partial correlation. restricts amount of variaces accounted for in y

84
Q

hierarchial model

A

iv’s entered cumlatively according to research design. needs r^2 and partial coeff at each point of additon to equation. variance get portioned by entry of variables. assigns causal priority to most appropriate. 1st entry gets greater variance in y than others

85
Q

stepwise

A

selects predictors based on relative contribution to model. Uses algorithim, little control. a posterior; computer selects greates part2 and greatest contribution to r2. bwd-smalled statististically insignificant factors are dropped. doesn’t rely on research question

86
Q

sequential

A

multivariate technique of using missing data by using sequential regression models