Topic Planner – Modelling Associations

Lesson 1 Lesson 2/3 Lesson 4 Lesson 5/6

Exercise 2A Exercise 2B Exercise 2C Exercise 2D

Fitting a Linear Least Association Conducting

Model squares and an
regression Causation investigation
line

✔ ✔✔ ✔ oo
 Conducting an Investigation

Solve practical problems by identifying, analysing and describing associations between

two variables.

Ex 2D
MAG112
Lesson 1
 Learning Goals Success Criteria

Identify, analyse & describe the

Bivariate Data
association between two categorical
Solve practical problems by identifying,
variables
analysing & describing associations between
Identify, analyse & describe the
two variables
association between two numerical
variables
Conduct an investigation to solve
practical problems

XMAS
Presentation
Designed
These goals and criteria are also available in QLEARN
Set Work for these lessons
Ex 2D
Example 10
Q1

Showbie Upload
q Demonstrated work on PSMT

This list is also available in Qlearn

 Statistical Investigation Process
The steps involved in undertaking this process is
similar to our PSMT process:
• Pose the question (Formulate) – decide what
data would allow you to address the problem
• Data (Formulate)– collect or obtain the data
• Analyse (Solve) – summarise and display the
data to answer the question posed
• Conclusion (Evaluate and Verify)– interpret the
results and communicate what has been
learned (Communication throughout)
What to use in an investigation…
The following table summarises the data analysis tools we can use in an investigation:
 We Do (good practice for your PSMT!)
The term ‘Predictor of height’ indicates that
‘Which is the better predictor of height will be the response variable. Why?
height, arm length or head
Because we want to see if arm length OR head
circumference?’ circumference is the better predictor of height,
Example 10 from textbook: we will need to conduct 2 analysis – one for
The following data were collected from a group of 20 students (all each variable against height.
measurements are in cm):
Use the data to investigate and answer the question.

• Pose the question – Already done above

• Data – Given
• Which is EV and RV?
• Analyse – summarise and display the data to answer the
question posed.
• What type of variables do you have?
• What calculations do you need to perform?
• Conclusion – interpret the results and communicate what
has been learned.
Steps to completing this question.
This is different from your textbook,
and more in line with what you
need to do for your PSMT.
EXCEL: REPORT:
Construct a short report, including the following for each
• Open the Excel spreadsheet from QLearn (Ex2D set of data:
Page). Data has been entered for you. Complete
• What was the study about?
calculations on SHEET 2 of the spreadsheet.
• What data did you use?
• Complete 2 scatterplots
• What calculations did you include in your analysis?
• Arm Length vs height
• Describe what you see in the scatterplot. What does
• Head Circumference vs Height this tell you?

• Calculate the following for each set of data: • Describe the correlation coefficient and coefficient of
determination. What do these values mean.
• Least Squares Regression line
• Include the LSRL equation in terms of the variables.
• Correlation Coefficient Interpret the slope and the intercept.
• Coefficient of Determination • What can you conclude from the calculations made
above?
• Construct a residual plot for both sets of data
• Evaluate your solution by showing that a linear model
in appropriate to use for these data sets.
 ‘Which is the better predictor of height, arm length or
We Do head circumference?’

What data analysis tools do we need to use for this

task?
Example
Based on these analyses the following report could be written to answer the question.
A study was conducted to investigate which measure was a better predictor of a person’s height, head circumference or
arm length. Data were collected from a sample of 20 students.

From the scatterplot of height versus arm length, we can see that there is a strong, positive, linear relationship between
height and arm length: 𝑟 = 0.8461. That is, those students with longer arms also tended to be taller. There are no
obvious outliers, and the linearity assumption is confirmed by the residual plot. The equation of the least squares
regression line is:
ℎ𝑒𝑖𝑔ℎ𝑡 = 44.09 + 1.742×𝑎𝑟𝑚 𝑙𝑒𝑛𝑔𝑡ℎ

The slope of the regression line predicts an increase of 1.742𝑐𝑚 in height for each 1𝑐𝑚 increase in arm length.
From the scatterplot of height versus head circumference, we can see that there is a moderate, positive, linear
relationship between height and arm length: 𝑟 = 0.480. That is, those students with larger head circumference also
tended to be taller. There are no obvious outliers, and the linearity assumption is confirmed by the residual plot. The
equation of the least squares regression line is:
ℎ𝑒𝑖𝑔ℎ𝑡 = 2.97 + 2.894×ℎ𝑒𝑎𝑑 𝑐𝑖𝑟𝑐𝑢𝑚𝑓𝑒𝑟𝑒𝑛𝑐𝑒

The slope of the regression line predicts an increase of 2.894𝑐𝑚 in height for each 1𝑐𝑚 increase in head circumference.

Comparing the values of the coefficient of determination for each variable we can see that for this sample 71.6% of the
variation in height is explained by the variation in arm length, while only 23% of the variation in height is explained by
the variation in head circumference.

Based on this comparison, we conclude that arm length is a much better predictor of height than head circumference.
How did you do?
Learning Goals Success Criteria
Set Work
Bivariate Data Identify, analyse & describe the
Solve practical problems by association between two
categorical variables
Ex 2D
identifying, analysing & describing
Identify, analyse & describe the Example 10
associations between two variables
association between two Q1
numerical variables
Conduct an investigation to
solve practical problems

Showbie Upload
q Demonstrated work on PSMT

These goals are also available in QLearn

By next lesson . . .

COMPLETE
Set work from
Ex 2D

UPLOAD REFLECT
Demonstrated work on Traffic Lights for
your PSMT Ex 2D

READ AHEAD
PSMT
Topic Planner – Modelling Associations

Lesson 1 Lesson 2/3 Lesson 4 Lesson 5/6

Exercise 2A Exercise 2B Exercise 2C Exercise 2D

Fitting a Linear Least Association Conducting

Model squares and an
regression Causation investigation
line

✔ ✔✔ ✔ ✔o
 Conducting an Investigation

Solve practical problems by identifying, analysing and describing associations between

two variables.

Ex 2D
MAG112
Lesson 2
 Learning Goals Success Criteria

Identify, analyse & describe the

Bivariate Data
association between two categorical
Solve practical problems by identifying,
variables
analysing & describing associations between
Identify, analyse & describe the
two variables
association between two numerical
variables
Conduct an investigation to solve
practical problems

XMAS
Presentation
Designed
These goals and criteria are also available in QLEARN
Set Work for these lessons
Ex 2D
Example 10 – completed last lesson
Q1

Showbie Upload
q Demonstrated work on PSMT

This list is also available in Qlearn

You Do – Q1 (good practice for your PSMT!)
The following table shows the results of a
study of obesity for a sample of 12 women
and eight men. The lean body mass, in
kilograms, and the resting metabolic rate for
each subject in the sample is shown.
The researchers hypothesised that lean
body mass (a person’s weight after allowing
for all fat) would have a strong association
with metabolic rate.
Use the data below to investigate this
hypothesis.

We want to see if there is a strong association

between body mass with metabolic rate.
Which is EV and which is RV?
 Your Turn
The steps involved in undertaking this process is similar to our
Set Work PSMT process:
• Pose the question (Formulate) – decide what data would
Ex 2D allow you to address the problem
Q1 • Data (Formulate)– collect or obtain the data
• Analyse (Solve) – summarise and display the data to answer
You will have all lesson to the question posed
complete this question.
• Conclusion (Evaluate and Verify)– interpret the results and
You need to complete the
communicate what has been learned (Communication
EXCEL spreadsheet yourself throughout)
from scratch (enter the data
yourself) and complete the
appropriate calculations.
Then write a report like last
lesson (use the outline from
last lesson to guide you.
How did you do?
Learning Goals Success Criteria
Set Work
Bivariate Data Identify, analyse & describe the
association between two
Solve practical problems by
identifying, analysing & describing categorical variables Ex 2D
associations between two variables Identify, analyse & describe the Example 10
association between two Q1
numerical variables
Conduct an investigation to
solve practical problems

Showbie Upload
q Demonstrated work on PSMT

These goals are also available in QLearn

By next lesson . . .

COMPLETE
Set work from
Ex 2D

UPLOAD REFLECT
Demonstration of work Traffic Lights for
on your PSMT Ex 2D

READ AHEAD
PSMT