Makeover Monday 2019 Week 43: Suicides by Age

Makeover Monday

Content warning: death, suicide

This post is going to discuss suicide and information presentation around suicide. If you are at risk, please stop here. You are not a number or a statistic. You are a life, a beautiful one, likely feeling a number of things. Please don’t dismiss it. There is life to live and you can:
Canada: 1-833-456-4566
US: 1-800-273-8255
UK: 116 123

~ Bridget Cogley

This week’s Makeover Monday dataset is from the Office for National Statistics in the UK. It shows trends in suicides in England and Wales from 1981 to 2017.

What works well?

  • The ridgeline chart effectively shows the shifting peak in suicides.
  • The gradient fill highlights the peaks, while still showing the full trend.

What could be improved?

  • Placing the year labels on the left would be more intuitive.
  • Adding more commentary would help explain the pattern being highlighted.

What Ryan did:

  • Created 7 ranges to group the number of deaths.
  • Used the ranges to build a horizon chart to show the increase in age over time.
  • Limited the chart to show only 1987-2017 so that the increase in age would be more apparent.
View in Tableau Public

What Marc did:

  • Built a custom ridgeline chart with D3.js as a community visualization in Data Studio.
  • Used a newspaper style with commentary taken from the original Office for National Statistics report.
View in Data Studio

Makeover Monday 2019 Week 38: Positive Impact Events

Makeover Monday

This week’s Makeover Monday is a special collaboration with the United Nations Sustainable Development Goals (SDGs) Action Campaign. It’s also notable that the original viz this week is built with Google Data Studio (which I believe is a first).

What works well?

  • Dropdown filters allow for interactivity.
  • Stacked bar charts show distribution of responses.

What could be improved?

  • The order of the segments in the stacked bars is not intuitive. It would make more sense to sort them in order of duration (Don’t know, 1 month, 6 months, Forever).
  • Since there are a different number of responses for each action, it would be more comparable to use a 100% stacked bar.
  • The treemaps are a disaster; absolutely unreadable.

What Marc did:

  • Used 100% stacked bar charts in a small multiple layout to enable fair comparison across all actions.
  • Applied colours, fonts, and icons from the UN SDG branding guidelines.
View in Data Studio

Makeover Monday 2019 Week 21: Fatal Bear Attacks

Makeover Monday

This week’s Makeover Monday focuses on fatal bear attacks in North America from 1900-2018. The original chart is the following bar chart from an article on Vox.

What works well?

  • Summarizing the data by month provides useful insight into the annual pattern of incidents.
  • Using a bar chart allows easy comparison between months.
  • The “X” on January and March helps emphasize that there were no attacks in those months.
  • The bears on top of each bar are gratuitous, but don’t distract from the overall message.

What could be improved?

  • Not much. The original chart is very clean and clear.
  • Month labels could be consistently abbreviated to 3 letters.

What Marc did:

  • Filtered the data to attacks occurring in Canada only.
  • Added subtle gridlines with axis labels.
View in Data Studio

Makeover Monday 2019 Week 18: Space Station Spacewalks

Makeover Monday

The topic of this week’s Makeover Monday is ISS spacewalks. The original chart is from NASA.

What works well?

  • Stacked bar chart makes it easy to compare between U.S. and Russia.
  • Bars are also labelled which helps with comparisons.
  • Colors are distinct.

What could be improved?

  • A legend could be added to make it clearer which color is for which country.
  • Background image is distracting and can be excluded.

What Ryan did:

  • I used the detail data set provided to show the top 25 astronauts by total spacewalk time.
  • Created a bar chart showing the total spacewalk time broken down by individual spacewalks.
  • Colored the bars by space suit used.
View on Tableau Public

What Marc did:

  • Created a diverging bar chart to allow each series to have an equal baseline. Oriented the axis vertically to avoid unintended positive/negative associations with bars above/below a horizontal axis.
  • Used light colours on a dark background in keeping with the space theme.
  • Excluded 2019, as there is only partial data.
  • Added annotations to highlight key events, including the Space Shuttle Columbia disaster and the end of the Shuttle program.
View in Data Studio

Makeover Monday 2019 Week 17: Stephen Curry’s Stadium Popcorn Rankings

Makeover Monday

This week’s chart comes from a New York Times article on Stephen Curry’s love for popcorn. He ranks each NBA stadium’s popcorn based on five categories.

What works well?

  • Stadium’s are sorted from highest to lowest total score.
  • Colour palette makes it easy to see the different ranks in the heatmap.

What could be improved?

  • The scale used for the rankings is not stated on the chart (it is stated in the article that a 1 to 5 scale was used).
  • Rather than encoding the data by colour saturation, a bar chart or dot plot would make differences in rating easier to perceive.

What Ryan did:

  • Remade the original chart using a dot plot.
  • Colored the columns by category.
View in Tableau Public

What Marc did:

  • Inspired by a video game “team select” screen, my viz allows two teams to be compared head-to-head.
  • The ratings across the 5 popcorn dimensions is shown in back-to-back bar charts.
View in Data Studio

Makeover Monday 2019 Week 16: Info We Trust by R.J. Andrews

Makeover Monday

This week’s Makeover Monday features Info We Trust by R.J. Andrews and a word cloud of the most frequently used words in the book.

What works well?

  • The word “data” stands out due to its size, colour, and orientation. It is clearly the most common word.

What could be improved?

  • Beyond “data”, it’s very difficult to perceive which words are 2nd, 3rd, 4th, etc. Almost any other chart type would be better.
  • Colour is used arbitrarily, without any meaning.

What Marc did:

  • Charted the top 10 words by frequency, broken down by page, chapter, and section.
  • Used a bubble plot; larger circles indicate a higher frequency of the word on the given page.
View in Data Studio

Makeover Monday 2019 Week 15: Ranking States by Fiscal Condition

Makeover Monday

This week’s Makeover Monday chart is from a 2018 report by the Mercatus Center at George Mason University, ranking each state by its fiscal condition.

What works well?

  • Clean design without any distracting map details.
  • Colour palette is distinct and visually pleasing.

What could be improved?

  • As with any geo map, the size of each state can skew our interpretation. A equal-sized tile map would address this problem.
  • The map shows only the ranking, but not the actual measure on which the rank is computed. Hence, we can’t assess how the states are distributed, i.e. how far above or below average is each state?

What Marc did:

  • The overall ranking is based on an aggregate fiscal condition index, which is composed of 5 underlying indices. I re-calculated all the index values based on the underlying data, according to the formulas defined in the full Mercatus report.
  • Using a jitter plot, I charted the distribution of the states on the overall index and each component index.
  • I kept the original colour scheme for each ranking group.
View in Data Studio

Makeover Monday 2019 Week 14: Waste on UK Beaches

Makeover Monday

The data for this week’s Makeover Monday challenge comes from the Marine Conservation Society’s Great British Beach Clean in 2017. The original chart is by the BBC.

What works well?

  • Clear title and subtitle which provides enough context
  • Types of waste are sorted from most to least
  • Overall style is visually appealing

What could be improved?

  • Using a bubble chart makes it difficult to compare the types by size
  • The chart only depicts the top 10 types of waste which makes up about 69% of the total waste found per 100 meters of beach. This could be called out somewhere on the viz.

What Ryan did:

  • The BBC article emphasized how problematic plastic pollution is so I wanted to highlight that.
  • Turned the original into a bar chart showing items per 100m by type
  • Coloured the items made from plastic
View on Tableau Public

Makeover Monday 2019 Week 13: Consumer Spending by Generation

Makeover Monday

For week 13, we look at the the spending of different generations.

What works well?

  • Percentages are labelled clearly which makes comparisons easier.
  • Grid lines are simple.
  • Using a 100% stacked bar makes it easy to see part to whole relationships for each generation.

What could be improved?

  • There there is no indication of what the y-axis measures.
  • Some more context could be provided such as when the data was collected or what country the consumers are from.
  • Difficult to compare spending for categories across generations.

What Ryan did:

  • Converted the stacked bar into bar charts
  • Coloured each generation group
View on Tableau Public

What Marc did:

  • Created small multiple bar charts by category.
  • Colour-coded each generation.
  • Created desktop and mobile views on separate pages in Data Studio. Click the link to switch between versions.
View in Data Studio