Introduction

Suicide is one of the significant causes of death globally, and it is also one of the most severe effects caused by mental health issues. About average 700,000 people die from suicide per year worldwide, and about 14.3% of deaths are attributable to mental disorders. I always hear a lot of similar voices online saying, “Everyone has some kind of mental health issues living in this high-pressure society.” And I couldn’t agree more. At the same time, knowing that suicidal death can be attributed to mental health issues, I was also curious about what else might be the cause of suicidal death from the world perspective. Therefore, I selected suicide as my topic for the data visualization project, aiming to answer the following questions and find the answers from the visualization.

What are my research questions?

1. What is the overall trend of suicidal deaths from 2005-2014 (10 years)?
2. Is there a correlation between economic development measured by annual GDP and suicide rate?
3. Is there a correlation between geographic location and suicide rate?

Inspiration

My visualizations are inspired by many mental health statistics visualizations that are very similar from the narrative perspective but different from the content. The population of the mental health graph contains many interactive elements, such as changing by year and filtering by region. I wanted to try to make my visualizations interactive as well. And I’m also impressed by the color on the map, which uses the red intensity to the severity. The line chart and map use the same data and dimension but show a totally different message. My visualizations also have this type of feature.

• 

• 

Process

Select Dataset

Kaggle

The first and most important thing for this project is to find a proper dataset that reflects the subject matter and has rich data. My dataset was retrieved from Kaggle, which is a massive repository of community-published data. The platform contains a large number of open datasets and a great variety of topics. With the user-friendly website design, I searched my topic through the simple word “Suicide” and found many similar exciting datasets. I then examined the top 10 most popular ones with the usability rating provided by Kaggle and the details of each dimension. Suicide Rates Overview 1985 to 2016 is my final decision of dataset for two reasons. Firstly, this dataset not only includes dimensions such as suicide number, country, gender, age and year, but also includes dimensions such as suicide rate(suicides/100k pop), population, and GDP for year. These give a lot of messages about how I could possibly use them to tell a story. Second, the dataset provides data from 1985 to 2016 instead of annual data, which gives great flexibility of the utilization.  

Select Tools

OpenRefine

After selecting the dataset, I imported the csv file into OpenRefine to double-check if the data is clean enough. As you can see, Open Refine is an open-source desktop application for data cleaning and format transformation.  I first checked with all columns to see missing values, duplications, and consistent values and corrected them using Open Refine. I then got rid of the extra useless column of data to keep my eyes focused. 

Tableau

After cleaning up the data, I transferred the dataset into Tableau, a well-known data visualization tool that helps the audience easily understand and interact with data. It empowers the data with solid information and insights through visualization. I played around my dataset on Tableau with different types of visualizations and came back to my research questions to find the best visualizations for the best stories to tell. I also filtered my dataset into dates only from 2005 to 2016 to keep the most relevant relationship between countries to show on the line chart. 

• 

• 

• 

Analyze & Visualize Data

I used a wide range of visuals to show the suicide data from various perspectives, including line chart, bar chart, bubble chart, area chart and map. To keep the data clearly informed and avoid the “noise” on the graphs, I decided to filter out countries not on the top 10 of a specific dimension for line charts and bar charts.

• 

• 

Color plays a crucial role in data visualization. To ensure the color delivers the right message across and is accessible to most people, I first checked ColorBrewer to get color inspirations and selected one of the sequential data color schemes where the primary color is purple to give a sense of depression about suicide. Based on the first color scheme I chose, I then selected a set of color schemes for different visualizations based on the nature of the data and graphs(see the image above).

Result

Figure 1: Top 10 Suicide Rate Countries & Trends (2005-2014)

This graph shows continuous data about suicide rates, trends, and comparisons with each other. I limited the number of countries due to the space and number of focuses on the graph for readability. I also added 60% of avg and 80% of avg area to provide reference information. This graph shows that besides the Republic of Korea and Lithuania, other countries’ suicide rates gradually decrease. For the Republic of Korea, the peak is from 2009 to 2011. This fact raises the question of why and we can find out more from additional research.

Figure 2: Worldwide Suicide Rate (2005-2014)

This graph shows the geographical information about the suicide rate, and it is helpful if the audience knows about countries’ locations on the map to see the message immediately. For the graph, you can see the purple area is the area that has a higher suicide rate which repeats a lot around Russia and surrounding countries. You might start to think about why again and find out more information with other research. 

Figure 3: Top 10 Suicide Rate Countries (2005-2014) & Figure 4: Top 10 Suicide Rate Countries Comparison (1995-2004 vs. 2005-2014)

These two graphs show the ranking of suicide rate directly using the same color. I decided to filter down countries into the top 10 again because of the readability and number of focuses on the graph. Figure 3 shows that the Republic of Korea has the highest suicide rate. However, surprisingly, Finland being the highest happiness score in the other study, also has a higher suicide rate. I’m curious why Finland has both extremes, and this question will be answered along with further research. 

Figure 5: GDP vs. Suicide Rate (2005-2014)

The bubble graph represents the relation between GDP per capita and suicide rate. The bigger the bubble, the larger the GDP per capita is, while the darker in purple for the bubble, the higher the suicide rate is. From this graph, you can see clearly that GDP per capita is about medium to low actually getting a higher suicide rate. And this is another “why” question here for future research. 

Figure 6: Age Group vs. Suicide Rate (2005-2014)

The area graph covers age group, gender, and suicide rate with continuous data for many messages. Surprisingly, the suicide rate for males is so much higher than for females. And it seems like people older than 75 years old have more tendency to commit suicide. Why is that? We could go further on research for these findings.

Reflection

Data visualization is compelling and meaningful to any aspect of life, culture, and development. I learned a lot through this project while having a hard time trying to do something but failed at the same time. For the visualizations shown here, I have to make them more general because that’s the only way I figured to do with the dataset and the tool. However, in the future, by giving more time and practice, I believe I can do a much better job of expressing my own narrative for the data I would choose.

References

Int J Environ Res Public Health. (2018, Sep 17). Suicide Risk and Mental Disorders. US National Library of Medicine National Institutes of Health. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6165520/

Ritchie, Hannah. (2018, May 16). Global mental health: five key insights which emerge from the data. Our World in Data. https://ourworldindata.org/global-mental-health 

Dattani, Saloni.,Ritchie, Hannah. & Roser, Max. (2018, April). Mental Health. Our World in Data.https://ourworldindata.org/mental-health 

World Health Organization. (2021, June 17). Suicide. World Health Organization. https://www.who.int/news-room/fact-sheets/detail/suicide