An intensity map showing the population density for different ethnic groups in Texas
What is it?
Google Fusion Tables allows users to create data visualisations such as maps, charts, graphs and timelines.
You can see five great examples of data journalism using Google Fusion Tables here.
"Google Fusion is easy", claimed James Ball data journalist from the Guardian investigations team and former chief data analyst for Bureau of Investigative Journalism, during a recent talk. "You would say that", I thought. So decided to test it out for myself.
The verdict: he was right, it is fairly easy to get started. Google Fusion is able to handle all types of location data, which is a real advantage. You don't need to have postcodes to plot your data, it can read UK place names, regions and counties.
1. Find some data
If you have not got anything in mind then have a browse on Data.gov.uk, a collection of searchable public data.
- Spend over £25,000 in Kingston Hospital
- Pupil absence in schools in England
- Cancer survival by cancer network
2. Download your data
Google Fusion accepts various formats, including Excel, .xls and .csv files (a full list of supported files is here). Have a quick look at your spreadsheet as you may need to make a few changes. For example, if one of your location lines is 'north west', Google Fusion will think this is the north west in the US so amend it to 'north west, UK'.
3. Upload your data
Go to Google Fusion and click 'new table' > 'import table'. You will need to sign in / sign up to Google.
Select the row number your column names are in from the drop down and click 'next'.
Fill in the requested information (your source and a link) and click 'finish'.
You can visualise your data as a:
- intensity map
- line, bar pie or scatter graph
- motion (must include text, date and number)
- timeline (must include date, text and number)
- storyline (must include text, date, date, text, text, text, text)
Now is the fun part: playing with your data (more on that below).
5. Add the data visualisation to your website
Make your visualisation public by clicking 'share' at the top right and copy the embed code. You may experience problems with embedding intensity maps with marker points as there is a recognised problem with this.
Here are a few we made earlier
Your data can be added as pin points on a map. For this example I downloaded a .csv file from data.gov.uk to show each London primary care trust's bill for antidepressants. You can hover over the pin the find out the stats.
To exclude some information from the pop-out windows (in this case I was interested in the cost but not the number of items bought), go to the 'configure info window'. You can do this automatically or by amending the HTML. Click 'configure styles' to change the style of the pin.
This intensity map is based on United Nations data on population predictions for the 19 most populous countries in 2100.
Google Fusion allows you to choose to show a map of a country, continent or world map. It doesn't yet have a zoom function so is not suitable for visualisations of geographical regions, such as the data on London PCTs, only as the United Kingdom as a whole.
Here is a link to an intensity map with markers (as there is a problem with embedding such maps). It shows the average peak time speed of traffic on A roads by region.
Here is the dataset of London PCT spending on antidepressants as a bar graph.
A timeline is a good way to show trends over time. Here I downloaded data on infant mortality from Data.gov.uk. This timeline shows postneonatal deaths but I could easily change the data field via a field drop down to show the total birth rate, for example.
As a journalist, you may now spot a story in the data. What happened in the late 1980s to result in a drop in postneonatal deaths?
Share your Google Fusion data visualisations with us by leaving a comment below or contact us @journalismnews
Free daily newsletter
- Tip: Convey the scale of inequality using visualisation tools
- Tip: How to verify your data
- 19 essential newsletters every journalist should read
- Data journalism: five ideas for more effective industry–academic collaboration
- Deepfakes, disinformation and detection: How can journalists know what is real?