Scribd
Upload
5 views

Scatter plot

The document provides an overview of scatter plots, including definitions of independent and dependent variables, types of correlation, and how to draw and interpret scatter plots. It explains the correlation coefficient, the line of best fit, and methods for making predictions using regression lines. Additionally, it includes class activities and examples related to analyzing the correlation between different sets of data.

Uploaded by

Itumeleng Amos Mogale
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
5 views

Scatter plot

The document provides an overview of scatter plots, including definitions of independent and dependent variables, types of correlation, and how to draw and interpret scatter plots. It explains the correlation coefficient, the line of best fit, and methods for making predictions using regression lines. Additionally, it includes class activities and examples related to analyzing the correlation between different sets of data.

Uploaded by

Itumeleng Amos Mogale
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 20

JLW 312: Mathematics

Education
07 March 2024
Dr Mandlenkosi Sibiya
Email: [email protected]
Office no: 4 – 14
Tel: 012 420 5631
SCATTER PLOTS
.
A scatter plot is a graph that helps you to see
whether there is a correlation (relationship) between
any set of two numeric data.
o The x-coordinate is the independent variable.
An independent variable is the variable that
determines the value of another variable (the
dependent variable). This variable can often be
manipulated.
o The y-coordinate is called the dependent variable.
A dependent variable is the variable whose values
Statistics
Drawing a Scatter Plot
NOTE:
Example
Scatter plots
Types of Correlation
.

o When describing the relationship between two variables displayed


on a scatter plot, we should comment on:
a) The form – whether it is linear or non-linear (either a quadratic or
exponential curve).
b) The direction – whether it is positive or negative
c) The strength – whether it is strong, moderate or weak.
This table shows different types of correlation.

.
Frequency polygons
.
In many real-life situations, scatter plots follow
patterns that are approximately linear. However, it
might sometimes look as though there is no
correlation between the variables.
The points might look as though they are randomly
scattered over the plane. However, on closer
inspection you may be able to recognise a
quadratic or an exponential shape to the pattern
of points or any other pattern.
Consider the examples given below.
.
Class Activity
Self-assessment

Given below are heights and foot lengths (both rounded off to the nearest centimetre) of eleven
learners from Eastern Cape schools as recorded in the 2009 Census@School:

a) Draw a scatter plot to show the relationship between foot length and height.
b) Are any of these points outliers? Explain why they are outliers.
c) What does the graph tell you about the correlation between foot length and
height?
CORRELATION COEFFICIENT
.
 Where a linear association exists between two variables, we say that the two variables
correlate. A commonly used statistical measure of association is called the correlation
coefficient.
 The correlation coefficient is a measure of the strength and direction of the linear
relationship between two variables.
 The symbol ‘r’ is used to represent the sample correlation coefficient.

The range of the correlation coefficient is –1 to 1.


o If x and y have strong positive correlation, r is close to 1.
o If x and y have strong negative linear correlation, r is close to –1.
o If there is no linear correlation or there is a weak linear correlation, r is close
to 0.
Consider the examples given below:
.
Finding the Correlation Coefficient Using the
Formula
.
Example
.

As part of the 2009 Census@School learners were asked the date of their birth and the
length of their right foot without shoes on, correct to the nearest centimetre. The
following table shows the data collected from 7 learners randomly selected from the
data base.

a) Draw a scatter plot to illustrate the data


b) Use the formula to determine the value of r.
c) Use the value of r to describe the type of linear correlation that exists between the age and
the foot length.
Solution
.
LINE OF BEST FIT
.
 If there is a strong linear correlation between the two aspects of the data, it may be possible
to draw a line that best models the data. The line is called the line of best fit or the
regression line or the least squares regression line.
 This line can be used to predict/estimate the value of one variable, given the value of the
other variable.
 The slope of the line shows the trend of the points

Drawing an intuitive line of best fit


o To draw a line of best fit, try to draw the line so that there is the same number of points above
it as below it.
o Outliers should clearly be ignored when fitting a line to the points.
Finding the Equation of the Regression Line
.
 Because your line of best-fit may not be the same as someone else’s, it is helpful to
have a systematic method that always gives the same result. One procedure
commonly used is the ‘method of least squares’.
 The equation of the linear regression line is + "# where is ‘y-hat, b is the
gradient of the line and a is the cut on the y-axis (or the y-intercept).
Least squares regression line
.
For the least squares regression line, if x is the mean of the x values and y is the mean for
the y values.
Example
.
The table below shows the masses and the heights of seven learners.

a) Use your calculator to determine r, the correlation coefficient.


b) Use the method of least squares to determine the equation of the regression line.
c) Draw a scatter plot to illustrate the data.
d) Draw the regression line on the scatter plot.
e) Use your graph to determine Joan’s height if her mass is 50 kg.
Solution
.
Using a Least Squares Regression Line to Make Predictions

.
o When a value for one of the variables that was not originally in the data is found, you are
making a prediction.
o The required value can be read off from the scatter plot or by using the equation of the
regression line. Predictions made from the equation of the line can be made through the
process of interpolation and extrapolation.
 Interpolation is a method of predicting/estimating new data value(s) within the known
range of data values.
 Extrapolation on the other hand is a method of estimating new data value(s) beyond a
discrete set of known data values.
Note that data values that are the result of extrapolation from statistical data are often less
valid than those that are the result of interpolation. This is because the values are often
estimated outside the tabulated or observed range of data.

You might also like