You are on page 1of 30

MA2 Applied Linguistics 2014

Quantitative methods

Mohamed Jabeur
1
Lecture Seven

Correlation

2
Outline
Introduction
1. Association
2. Correlation
2.1 Positive correlation
2.2 Negative correlation
3. Correlation coefficient
4. Correlation tests
4.1 Pearson’s correlation test
4.2 Spearman’s correlation test
Conclusion
3
Introduction
Association

Correlation

Simple regression
Multiple regression
4
1. Association
Any relationship between two
measured quantities that
renders them statistically
dependent.

5
Two VARIABLES are associated
if some of the variability of one
can be accounted for by the
other.

6
2. Correlation
Measure of linear association
A relationship between two variables in which
both variables move in tandem.

7
2.1 Positive correlation

One variable decreases, the


other variable also
decreases and vice versa.

8
A positive correlation is a
relationship between
two variables such that
their values increase or
decrease together. 
9
2.2 Negative correlation
A negative correlation is a
relationship between
two variables such that as the
value of one variable
increases, the other
decreases. 
10
3. Correlation coefficient
A quantitative value of the
relationship between two
or more variables.

11
Range from .00 to 1.00 in either a
positive or negative direction.

Perfect correlation is 1.00 (either


+1.00 or –1.00)

No relationship at all is .00


12
Perfect positive correlation

13
Perfect negative correlation

14
Examples of negative correlation
Commodity supply & demand.
Pages printed & printer ink
supply.
Education & religiosity
Conservativism & cognitive
ability.
15
16
17
18
Neither association
nor correlation
establish causality.
19
4. Correlation tests
Pearson’s correlation
test
Spearman’s rank
correlation test 20
4.1 Pearson’s correlation test
A statistic measuring the linear relationship
between two variables in a sample & used as an
estimate of the correlation in the whole population,
given by r.

Pearson’s product moment correlation -coefficient.

Named after Karl Pearson (1857-1936), British


mathematician.
21
Perfect positive linear c.

22
Perfect negative linear c.

23
Correlation r = 0

24
Correlation r = 0.63

25
Correlation r = 0.97

26
4.2 Spearman’s rank correlation
To test the association between
two ranked variables, or one
ranked variable and one
measurement variable.

27
Spearman’s rank correlation
Critical values Pearson.docx

28
Correlation test on Excel
Classeur Excel Lecture 7.xlsx

29
30

You might also like