correlation and regression (w7)

Question 1

when may data be a mistake

Answer 1

perfect straight line, one or more datapoints a long way away from others, no relationship at all between things you expect to be related

Question 2

what does correlation find

Answer 2

line of best fit by minimising the differences between data and line

Question 3

equation for correlation/line of best fit

Answer 3

r = Sxy (how much x and y change together) / Sx.Sy (how much x and y change separately)

Question 4

what does the r value tell you

Answer 4

the direction of and how strong the correlation is

Question 5

when is correlation positive

Answer 5

r is above 0
1>(and equal to) r > 0

Question 6

when is correlation negative

Answer 6

r is below 0
-1< (and equal to) r < 0

Question 7

when is correlation strong

Answer 7

if r is close to 1

Question 8

when is correlation weak

Answer 8

if r is close to 0

Question 9

what does the r2 (r-squared) value tell you

Answer 9

how much of the variance is explained by your correlation

Question 9

what does r2 close to 1 mean

Answer 9

correlation explains a lot of variance

Question 10

what does r2 close to 0 mean

Answer 10

correlation explains only a little variance

Question 11

what is r2 also called

Answer 11

coefficient of determination

Question 12

what is 1-r2

Answer 12

amount of variance not explained

Question 13

regression: when x increase by 1, what happens to y

Answer 13

y increases by the slope

Question 14

correlation and regression: describes how many relationships

Answer 14

C - describes single relationship
R - can describe multiple relationships

Question 15

correlation and regression: what do they describe about the relationship

Answer 15

C - direction and strength of relationship
R - directionS and strengthS of relationshipS

Question 16

correlation and regression: are x and y variables inter-changable

Answer 16

C - yes
R - no

Question 17

correlation and regression: do they allow prediction

Answer 17

C - no
R - yes

Question 18

correlation and regression: what are the co-efficients

Answer 18

C - r, r2
R - R, R2, F, t, SE, ß1-n

Question 19

how do you report correlations

Answer 19

r ([df]) = [pearson’s r] , p = [p-value]
r(64) = .881 , p < .001

Question 20

how do you report regressions (overall model fit)

Answer 20

F ([df1] , [df2]) = [F-value] , p = [p-value]
F (1 , 64) = 223 , p < .001

Question 21

how to report regressions (individual variables)

Answer 21

estimate = 121 , SE=8.10 , t(64) = 14.9 , p < .001

Question 22

what is t when reporting regression

Answer 22

t = estimate/SE (this is a t test)

Question 23

in multiple regression what is the outcome and predictor variables, how many

Answer 23

single outcome variable - y
multiple predictor variables - x1 , x2 , …

Question 24

in multiple regression, what do you find instead of line of best fit

Answer 24

find the best fitting surface (3d graph with a 2d shape (surface) acting as line of best fit)

Question 25

in multiple regression, what are residuals

Answer 25

distance from surface

Question 26

in multiple regression, what can predictors be

Answer 26

can be almost anything:
continuous, ordinal or discrete
normally - distributed or not
linear or non-linear

Question 27

what is a problem with correlation and regression, what may you have to do

Answer 27

extrapolation
non linear relationships (may have to transform data - quadratic, cubic, logarithmic)