Statistics Lecture notes
Lecture 9: Association interval and ordinal variables
The covariance is an indication (the ‘combined’ variance) of the direction of the correlation, but does
not indicate the strength of the correlation
Pearson’s r: from -1 to +1 (neg or pos)
Eta: from 0 to +1 (no direction)
r = 0 → no correlation at all
The covariance is scale sensitive, r is not scale sensitive
Kendalls tau:
- A pair is concordant if the subject ranked higher on X also ranks higher on Y.
- The pair is discordant if the subject ranking higher on X ranks lower on Y.
- The pair is neutral if the subjects have the same classification on X and/or Y.
Df = n-2 (when there are 2 variables)
1
,2
,Is the influence of length on jumping height this
small because of suppressing effect of BMI?
3
, Lecture 10: Linear regression part 1
Linear regression… what is it?
• A way of predicting the value of one variable
from another. (→ see lecture 3)
• It is a hypothetical model of the relationship
between two variables.
• The model that we use in a linear regression
is linear and therefore we use the equation
of a straight line.
The dependent variable is number of mistakes
The independent variable is level of effort.
We see a negative relationship.
we can interpret this as follows: a one increase in the
level of effort results in a decrease of 1.5 mistakes.
4
Lecture 9: Association interval and ordinal variables
The covariance is an indication (the ‘combined’ variance) of the direction of the correlation, but does
not indicate the strength of the correlation
Pearson’s r: from -1 to +1 (neg or pos)
Eta: from 0 to +1 (no direction)
r = 0 → no correlation at all
The covariance is scale sensitive, r is not scale sensitive
Kendalls tau:
- A pair is concordant if the subject ranked higher on X also ranks higher on Y.
- The pair is discordant if the subject ranking higher on X ranks lower on Y.
- The pair is neutral if the subjects have the same classification on X and/or Y.
Df = n-2 (when there are 2 variables)
1
,2
,Is the influence of length on jumping height this
small because of suppressing effect of BMI?
3
, Lecture 10: Linear regression part 1
Linear regression… what is it?
• A way of predicting the value of one variable
from another. (→ see lecture 3)
• It is a hypothetical model of the relationship
between two variables.
• The model that we use in a linear regression
is linear and therefore we use the equation
of a straight line.
The dependent variable is number of mistakes
The independent variable is level of effort.
We see a negative relationship.
we can interpret this as follows: a one increase in the
level of effort results in a decrease of 1.5 mistakes.
4
