5.4 – The product moment correlation coefficient
www.doteducation.org +971569242777/ +971568161777/ +971525465652
1
2
3
FORMULA
PMCC varies between -1 and 1.
means
Perfect positive correlation.
means
No correlation
means
Perfect negative correlation.
PMCC varies between -1 and 1.
means
Perfect positive correlation.
means
No correlation
means
Perfect negative correlation.
Perfect negative
Strong negative
No correlation
Weak positive
Perfect positive
Maths score
English score
Strong correlation
Strong correlation
Interpreting the PMCC
Interpreting the PMCC
“Interpret” vs “State”
In general in Statistics exams, the word ‘interpret’ means “explain in context using non-statistical language”.
A bad answer (that may or may not be accepted):
“Strong negative correlation” (this is stating the correlation not interpreting it)
A good answer:
“As the waiting time increases, the customer satisfaction tends to decrease”.
Interpreting the PMCC
“Interpret” vs “State”
In general in Statistics exams, the word ‘interpret’ means “explain in context using non-statistical language”.
A bad answer (that may or may not be accepted):
“Strong negative correlation” (this is stating the correlation not interpreting it)
A good answer:
“As the waiting time increases, the customer satisfaction tends to decrease”.
Item age (years) t | 85 | 256 | 142 | 120 |
Value (£) P | 350 | 1200 | 425 | 300 |
Example:
Item age (years) t | 85 | 256 | 142 | 120 |
Value (£) P | 350 | 1200 | 425 | 300 |
Example:
Time
Population of Tanzania
The data points for population over time doesn’t fit a straight line very well, so there is not a strong correlation between time and population.
Comment on this claim.
Exponential model
Linear model
Important Point 1
Time
Population of South Korea
The data shows that as time increases, population increases.
Comment on this claim.
Correlation does not imply causation.
As time increases, the population tends to increase, so there is a positive correlation before time and population.
But there are examples, e.g. the highlighted data points above, where this is not true: time increased but the population decreased.
We say there is not a causal relationship between time and population, meaning that a (positive) change in one quantity does not directly cause a (positive) change in the other.
Important Point 2
Time
Population of South Korea
There is strong positive correlation.
Comment on the student’s answer.
The question asks to interpret the relationship, not to state the type of correlation.
Interpret means to give a worded description in the context of the problem.
Her answer could have been:
“As the time increases, the population tends to increase.”
Question: Interpret the relationship between time and population.
Important Point 3
15
EFFECT OF CODING ON PMCC
16
Does the equation of the regression line change?
Does the PMCC change?
Yes. The regression translates right so will have a different equation.
No. As the data points and regression line moved together, the extent to which the data fits the line doesn’t change.
Effect of Coding on PMCC
Effect of Coding on PMCC
Effect of Coding on PMCC
The Product Moment Correlation Coefficient (PMCC), which is a measure of the strength of linear relationship between two statistical variables.
Perfect negative
Strong negative
No correlation
Weak positive
Perfect positive
Maths score
English score
Time
Population
Recall that PMCC only measures the strength of linear correlation, i.e. how well the data fits to a straight line. For example, population growth tends to be exponential not linear, so the PMCC may be low even though population and time are strongly (exponentially) correlated.
Item age (years) t | 85 | 256 | 142 | 120 |
Value (£) P | 350 | 1200 | 425 | 300 |
We will use our calculator’s statistics mode to calculate this directly.
Calculating PMCC using a Calculator
Let’s do it on our calculators!
Baby | A | B | C | D | E | F |
Head Circumference (x) | 31.1 | 33.3 | 30.0 | 31.5 | 35.0 | 30.2 |
Gestation Period (y) | 36 | 37 | 38 | 38 | 40 | 40 |
Calculating PMCC using a Calculator
These are instructions for the Casio fx-CG50
Enter the values, pressing EXE after each and using the arrow keys to move around.
Choose 2 for Statistics.
1
2
3
4
Item age (years) t | 85 | 256 | 142 | 120 |
Value (£) P | 350 | 1200 | 425 | 300 |
Calculating PMCC using a Calculator
These are instructions for the Casio fx-570/991CW
Choose 2-Variable.
Use the arrows and OK to select Statistics.
1
2
Enter the values, pressing EXE after each value. Use the arrow keys to move around.
3
4
Item age (years) t | 85 | 256 | 142 | 120 |
Value (£) P | 350 | 1200 | 425 | 300 |
IQ (i) | 105 | 108 | 115 | 122 | 147 | 165 |
Chess rating (c) | 1450 | 1610 | 1820 | 1570 | 1480 | 2700 |
English score | 78 | 92 | 89 | 62 | 36 |
Maths score | 67 | 100 | 53 | 72 | 54 |
1
2
a
b
a
b
Your Turn
IQ (i) | 105 | 108 | 115 | 122 | 147 | 165 |
Chess rating (c) | 1450 | 1610 | 1820 | 1570 | 1480 | 2700 |
English score | 78 | 92 | 89 | 62 | 36 |
Maths score | 67 | 100 | 53 | 72 | 54 |
1
2
a
b
a
b
Your Turn
Recall that variance, which is a measure of the ‘variability’ of data, is defined to be the average squared distance from the mean.
Summing these and dividing by the number of data points then averages these squared distances.
Spec note: In the UK, this skill is only in Edexcel Further Maths FS2, Edexcel IAL, OCR FM Statistics and CCEA (Northern Ireland).
This can be simplified into a more easily calculatable form, in a similar way to the variance formula being simplified.
PMCC can then be calculated as follows:
?
Quickfire!
?
?
?
?
?
?
?
?
Quite often the values are given to you in an exam.
Quickfire!
Quite often the values are given to you in an exam.
www.doteducation.org +971569242777/ +971568161777/ +971525465652
34
www.doteducation.org +971569242777/ +971568161777/ +971525465652
35
www.doteducation.org +971569242777/ +971568161777/ +971525465652
36
www.doteducation.org +971569242777/ +971568161777/ +971525465652
37
www.doteducation.org +971569242777/ +971568161777/ +971525465652
38
www.doteducation.org +971569242777/ +971568161777/ +971525465652
39
www.doteducation.org +971569242777/ +971568161777/ +971525465652
40
www.doteducation.org +971569242777/ +971568161777/ +971525465652
41
www.doteducation.org +971569242777/ +971568161777/ +971525465652
42
www.doteducation.org +971569242777/ +971568161777/ +971525465652
43
www.doteducation.org +971569242777/ +971568161777/ +971525465652
44
www.doteducation.org +971569242777/ +971568161777/ +971525465652
45
www.doteducation.org +971569242777/ +971568161777/ +971525465652
46
www.doteducation.org +971569242777/ +971568161777/ +971525465652
47
www.doteducation.org +971569242777/ +971568161777/ +971525465652
48
www.doteducation.org +971569242777/ +971568161777/ +971525465652
49
50
CODING BIVARIATE DATA
51
52
53
www.doteducation.org +971569242777/ +971568161777/ +971525465652
54
www.doteducation.org +971569242777/ +971568161777/ +971525465652
55
www.doteducation.org +971569242777/ +971568161777/ +971525465652
56
www.doteducation.org +971569242777/ +971568161777/ +971525465652
57
www.doteducation.org +971569242777/ +971568161777/ +971525465652
58
www.doteducation.org +971569242777/ +971568161777/ +971525465652
59
www.doteducation.org +971569242777/ +971568161777/ +971525465652
60
www.doteducation.org +971569242777/ +971568161777/ +971525465652
61
www.doteducation.org +971569242777/ +971568161777/ +971525465652
62
63
www.doteducation.org +971569242777/ +971568161777/ +971525465652
64
www.doteducation.org +971569242777/ +971568161777/ +971525465652
65
www.doteducation.org +971569242777/ +971568161777/ +971525465652
66
www.doteducation.org +971569242777/ +971568161777/ +971525465652
67
68
EXTRA THINGS USED TO SOLVE QUESTIONS : A MUST KNOW
69
EXTRA THINGS USED TO SOLVE QUESTIONS : A MUST KNOW