Nowadays, not a week passes without a story in the press about the impact of covid-19 on mental health and more specifically on suicides. Claims on social media seem to appear daily and discussions on these topics are getting a huge attention. Therefore, we want to analyze the socio-economics effects on suicides, over the years, in different countries and generations, so that to better understand some of the reasons behind this illness.
For the complete documents:
We are using the Suicide Rates Overview 1985 to 2016 dataset from Kaggle [1], that was in turn taken from the World Health Organization databank [2]. It contains 27.8k valid entries with the following attributes: country, year, sex, age, # suicide_no, # population, # suicides/100k pop, country-year, HDI for year, gdp_for_year, gdp_per_capita, generation.
The service is composed by the following visualizations:
-
Map [ #suicides/100k population colorscale ] (select one or more countries)
- It allows zoom and pan
- On country selected, Age chart will show only the related deaths
- On country selected, Sex chart will show only the related deaths
- On country selected, PCA will be replaced by the Radar chart that shows all the main information about the selected cuntry
- On country selected, Scatterplot will be replaced by the Line chart that shows the Suicide radio over the years
- On country selected, the other countries inside Map will reduce their opacity. We can select at most three countries
- On mouse over one country, the service shows a tooltip contains all the related information
- On mouse over one country, the relative circle on PCA chart increases its size and gets a blue stroke
- On mouse over one country, the relative circle on Scatterplot increases its size and gets a blue stroke
-
PCA chart
- On mouse over one point, the relative country on Map becomes blue
- On mouse over one point, the relative circle on Scatterplot increases its size and gets a blue stroke
-
Sex bar chart [ #suicides/100k population - sex ] (can select male or female)
- On sex bar selection, Age chart will show only the related deaths
- On sex bar selection, Scatterplot will show colors referring to related deaths
- On sex bar selection, Map will show colors referring to related deaths
- On mouse over one bar, the other bar shows the divergence to the current suicide ratio value
- When the year selector has a value that is different than “All”, a slim bar is shown to compare the current data with the global data
-
Age bar chart [ #suicides/100k population - age group ] (can select age group)
- On age group selection, Sex chart will show only the related deaths
- On age group selection, Scatterplot will show colors referring to related deaths
- On age group selection, Map will show colors referring to related deaths
- On age group selection, the average bar for only-selected-groups is updated
- On mouse over one bar, the other bars show the divergence to the current suicide ratio value
- When the year selector has a value that is different than “All”, a slim bar is shown to compare the current data with the global data
- Every time a filter is applied by other visualizations, the mean bar is updated
-
ScatterPlot [ relationship GDP_year/GDP_capite - #suicides/100k population colorscale ] (brush on points)
- On brush, Scatterplot will zoom over the selected points and the avg bars for the axis values are updated
- On brush, Age chart will show only the related deaths
- On brush, Sex chart will show only the related deaths
- On brush, Map will show colors referring to related deaths
- On mouse over one point, the service shows a tooltip contains all the related information
- On mouse over one point, the relative country on Map becomes blue
- On mouse over one point, the relative circle on PCA chart increases its size and gets a blue stroke
- Every time a filter is applied by other visualizations, the mean bar is updated
-
Line chart
-
Radar chart used to compare the paramaters of the seleced countries. Specifically compare: population, gdp per capita, gdp for year, # suicides and suicides rates
Moreover, the interactive legend shows has the following features:
- On mouse over a legend circle, Map will decrease the opacity of points that are not inside the range
- On mouse over a legend circle, ScatterPlot will decrease the opacity of points that are not inside the range
- On mouse over a legend circle, the service shows a tooltip contains all the countries that are inside that range
- When other visualizations update the values of suicide ratio, the blue outline used to denote the mean is updated;
Finally, an year selector is used to filter data in order to show only those related to the selected year. Every time a year diffrent than 'All' is selected, the bar charts show a second small bar for each group that is referred to the value without any filter applied.
Make sure you have python3 installed, then run:
python run.py [port]
[1] https://www.kaggle.com/russellyates88/suicide-rates-overview-1985-to-2016
[2] https://www.who.int/teams/mental-health-and-substance-use/suicide-data