Association Flashcards

1
Q

Association between variables

A

correlation, regression, chi squared test

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

association aims to see if

A

x and y are related

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

correlation

A

examines direction/strength of association, both variables are continuous (pearson moment), or one is continuous one is variable is ordinal (spearmans), no outliers, linearity, related pairs

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

regression

A

examines how well the line describing association can be used to explain y variation

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

numeric variables

A

correlation, regression

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

chi squared

A

contingency tables, examines association of 2 independent variables

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

categoric values

A

chi squared

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

correlation analysis

A

plot data using scatterplot to see outliers/linearity, then work out coefficient (standardise data, observation-mean/standard deviation) then determine if can happen by chance (set null the p value)

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

negative correlation

A

as x increases y decreases

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

positive correlation

A

as x increases y increases

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

what does 1/-1, 0 mean

A

stronger correlation, 0 is weak/no

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

test equation

A

r\/(n-2)/ 1-r(2)

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

however

A

significant correlation doesn’t mean strong correlation, large samples can have significant correlation when its weak (vice versa), linear relation doesn’t mean there’s a relationship

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

regression equation

A

y=b(o) +/- b(1)X

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

what do they mean

A

b(o) is intercept, +/- is correlation, b is slope

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

b(1)

A

b(1) = sy/sx

17
Q

b(o)

A

b(o) mean of y - b(1) x mean of x

18
Q

regressions assumptions

A

related pairs, linearity, homoscedascity, data is normally distributed

19
Q

when to use correlation

A

individuals sampled at random and you want to know if 2 variables are linear

20
Q

when to use regression

A

when values of one variable respond linearly to other variable, predictions