If you are viewing this website through The National Archives it may be necessary to copy and paste external links directly into your browser.
Skip to main content
You are here: Skip Navigation Links

Quick guide: correlation charts

1. When should I use correlation charts?

Correlation charts can indicate a potential relationship between two indicators, whether it is a positive or negative relationship and how strong that relationship may be. For example, you may wish to find out whether higher levels of deprivation in an area are correlated with poorer school attainment. A correlation chart can tell you whether there is any relationship between the two issues. However, it cannot tell you whether one causes the other, or whether both are influenced by other factors.

Points to consider

  • Where possible, ensure the indicators you are plotting are close together in date.
  • Consider the logic of the indicators you have selected and whether there is evidence to suggest that they are associated.
  • Once you have chosen your indicators, think which one should be the explanatory indicator (which tends to be the one that can be controlled or affected) and which one should be the response indicator (usually a desired outcome, such as improved health or wellbeing).
  • Remember that even if indicators are associated (ie 'vary together'), it does not imply that one causes or influences the other.

top

2. How do I access the correlation charts?

From the main Data Selector page, expand the 'Correlation charts' theme using the small + sign on the left hand side or go directly to the Correlation charts theme. You will then see a short list of options, arranged by geographic level. Select the geographic level at which the indicators you are interested in are available. Please note that not all indicators are available at each geographic level. On the next page, select 'Correlation chart (html)' to open up the initial chart. By default, the indicator at the top of the list of available indicators is selected to appear on both the x-axis and the y-axis. You will need to change these by clicking on the buttons above the chart (shown below) and choosing different indicators.

buttons

top

3. How should I interpret the correlation charts?

A chart will be produced plotting the two indicators as well as a regression line. A regression line sloping down from left to right indicates a potential negative relationship, whilst a regression line sloping upwards indicates a potential positive relationship. The example shown below compares IMD average score (deprivation) against GCSE achievement. The regression line sloping downwards shows that as deprivation (the explanatory indicator) increases, the percentage of children achieving five good GCSEs (the response indicator) decreases.

chart

You can use your mouse to hover over any dot on the chart to show which area is represented. Hovering over an area in the list on the right hand side locates the area on the chart for you.

The correlation coefficient (r) is shown below the chart. This measures the strength and direction of any potential linear relationship between the two indicators. Its value is always between +1 and –1. As a general guide, the following interpretations apply:

  • Exactly –1. A perfect negative linear relationship
  • –0.70. A strong negative linear relationship
  • –0.50. A moderate negative linear relationship
  • –0.30. A weak negative linear relationship
  • 0. No linear relationship
  • +0.30. A weak positive linear relationship
  • +0.50. A moderate positive linear relationship
  • +0.70. A strong positive linear relationship
  • Exactly +1. A perfect positive linear relationship

A positive correlation indicates that as the explanatory indicator increases, the response indicator will also increase. A negative correlation indicates that as the explanatory indicator increases, the response indicator will decrease. The strength of the relationship indicates the magnitude of this corresponding increase or decrease. In the example above the correlation coefficient is -0.39 which suggests a weak to moderate negative relationship.

top

4. Why are some indicators not available?

We have to add indicators manually, so we have selected only a limited number of indicators for inclusion at this stage. If you have used an indicator elsewhere within Data Atlas and would like to be able to use it within a correlation chart, then we will do our best to add it. Contact us using the link above the chart, or through info@chimat.org.uk to request this. Please tell us the exact name of the indicator you require, with a link to where you have seen it on Data Atlas, if possible, as well as your contact details in case we need any further information from you.


Validates against XHTML 1.0  Cascading Style Sheet compliance  Single A Accessibility compliance

******