Data Journalism

Chris Knox

Data Editor at the New Zealand Herald



  • Styling
  • Chart choice
  • Transpose
  • Maps

What is a visualisation?

Using the visual modes available; size, colour, shape, and position to represent (encode) the values in the data.

It it usual for quantities to be proportional to the area of the visual encoding - this makes circles very poor for accurate comparisons of values.


It's actually quite old

nightingale barchart

Drawing time

2017 Election Party Vote

National Party1,152,07544.4
Labour Party956,18436.9
New Zealand First Party186,7067.2
Green Party162,4436.3
The Opportunities Party (TOP)63,2612.4
Māori Party30,5801.2
ACT New Zealand13,0750.5
Aotearoa Legalise Cannabis Party8,0750.3
New Zealand People's Party1,8900.1
United Future1,7820.1
NZ Outdoors Party1,6200.1
Democrats for Social Credit8060.0
Internet Party4990.0

2017 Election Party Vote by Voting Type

Election Day1,151,12344.4
Party Only42,2541.6

2017 Election Party Vote by Voting Type and Party

Advance Voting

PartyTypeVotesPercent of typePercent of party
ACT New ZealandAdvance4,6900.535.9
Green PartyAdvance64,1726.539.5
Labour PartyAdvance375,29838.239.2
National PartyAdvance464,93947.340.4
New Zealand First PartyAdvance72,9797.439.1

Election Day Voting

PartyTypeVotesPercent of typePercent of party
ACT New ZealandElection Day6,2830.648.1
Green PartyElection Day63,6095.839.2
Labour PartyElection Day402,96236.742.1
National PartyElection Day534,25948.646.4
New Zealand First PartyElection Day91,1498.348.8

Party only votes

PartyTypeVotesPercent of typePercent of party
ACT New ZealandParty Only1530.41.2
Green PartyParty Only2,6336.81.6
Labour PartyParty Only22,14457.02.3
National PartyParty Only11,14928.71.0
New Zealand First PartyParty Only2,7897.21.5

Special votes

PartyTypeVotesPercent of typePercent of party
ACT New ZealandSpecial1,9490.614.9
Green PartySpecial32,0299.119.7
Labour PartySpecial155,78044.316.3
National PartySpecial141,72840.312.3
New Zealand First PartySpecial19,7895.610.6

Bad or deceiving charts

  • Charts and graphs can be used to deceive
    • Don't do this.

The best way to get a sense for bad charts is to peruse or /r/dataisugly. There is also a good writeup here

In the New Zealand context Stats Chat is great.

The most common bad things are:

  • Incorrect, missing, or misleading labels
  • Inconsistenct scales
  • Truncating scales
  • Comparing things that shouldn't be
  • Too many things

A few rules

  • Barcharts always start at 0
  • Line charts don't need to start at 0, but always ask yourself if the range you select is going to make an insignificant change look important
  • Only use pie charts for parts of a whole and only when there are less than 5 categories
  • Avoid maps for showing quantities


Pick a couple of charts from Figure.NZ and remake it.

  • Have you made them better or worse?
  • If you have improved them is the data being shown in a different way?


In pairs discuss these:

  • What do you think of these approaches?
    • Do you like them?
    • Are there other ways you would prefer to learn about this data?
  • Are they journalism?
  • Are they news?

NZH Improvements

I probably should have included a median income version as well as an average income version.

Digressions Why?

All summary statistics hide things

The mean and standard deviation are the same for each of these graphs


  • Select some data with two variables from Figure.NZ
  • Clean and reshape the data in Workbench
  • Export the data to Datawrapper and create a chart
  • Email the chart link to and share the Workbench workflow with me
  • Include a sentence about why you choose this data - why would it be newsworthy
  • If you changed to chart type from the Figure.NZ default include a sentence about why you choose that chart type.
© 2019–20 by Chris Knox. All rights reserved.
Last build: 2020-11-25