Scatter Plot Matrix

Introduction

A Scatter plot matrix is a type of diagram which displays the pairwise relationships between a number of variables within a dataset. The data is displayed as a collection of points, each having the value of one variable determining the position on the horizontal axis and the value of the other variable determining the position on the vertical axis. These plots are organised into a matrix to look at all pairwise correlations in one place It is usually used to compare several sets of observations or data. It is similar in concept to the Correlation Matrix tool, with the difference being that it shows the actual scatterplots between the variables, rather than a representation of the strengths of the relationships.

Inputs

To show the Scatter plot matrix tool in use, we will run it on some income and inequality datasets from Melbourne to see if there is a relationship is between the rate of them across the city

  • Select Melbourne GCCSA as your area
  • Select SA2 OECD Indicators: Income, Inequality and Financial Stress 2011 as your dataset, and select the following variables:
    •  SA2 Name
    •  Standardised Median Disposable Household Income (Synthetic Data)
    •  Standardised Gini Coefficient (Synthetic Data)
    •  Standardised Poverty Rate (Synthetic Data)

Once you have selected these, open the Scatter Plot Matrix parameter input window (Tools → Charts → Scatter Plot Matrix) and enter the parameters as in the image shown below (these are explained in more detail below the image)

[Click to Enlarge]

[Click to Enlarge]

  • Dataset Input: Select the dataset that you would like to run the Scatter Plot Matrix tool for. In this instance, we choose SA2 OECD Indicators: Income, Inequality and Financial Stress 2011
  • Variable Name: Select the variables that you would to include in the Scatter Plot Matrix tool. Select them in the order that you would like them to appear from left to right along the matrix. In this instance, we have selected Standardised Median Income, Standardised Gini Co-efficient and Standardised Poverty Rate
  • Chart Title: Enter a title for your graph. In this instance we have entered Income versus Gini co-efficient versus Poverty Rate
  • Greyscale: Check this box if you would like your graph in greyscale, keep unchecked (default) if you would like the graph to be in colour

Once you have entered your parameters, click Add and Run

Outputs

Once your graph tool has run, click the Display button that appears on the pop up dialogue box. Your graph should look something like the image shown below

[Click to Enlarge]

[Click to Enlarge]

The bottom left and the top right corners are mirror images of each other  – you only to look at the results in one of the corners. We can see in this instance that there doesn’t appear to be a relationship between median income and the Gini co-efficent, a negative relationship between median income and poverty rate, and a positive relationship between poverty rate and the Gini co-efficient