First intermediate implementation
This is a first approach at visualizing migration data using Svelte according to the goals presented in previous post entries. One of the figure of this first implementation is presented below. The data we used here is the migration cumulative data from 1990 to 2019. The names of migrating destinations are shown below the plot. In this figure, the number of boxes stands for the number of immigrants: the red boxes stand for female immigrants, and blue boxes stand for male immigrants. A short black line segments the immigration data of different years among boxes. We can notice that there are some countries without any boxes in their area. This condition is because the numbers of immigrants in these countries are too limited.
This plot is only our trial for Svelte. It may not be further improved and included in our final report. But the following issues need to be further enhanced. Firstly, since many immigrants differ significantly by year, some countries might have enough immigrants to generate boxes for some year. However, they may not have enough immigrants for a specific year. Under this condition, we can hardly observe which exact year is absent from the plot. To solve this problem, we expect the box with different colors (such as white standing for empty) could also be generated and added to the plot to make it more straightforward. Or we could add a panel that is triggered by hover: when you hover on a specific box, the year this box stands for will be displayed in the panel. Secondly, we set a threshold to create a box which means the differences which are not significant enough could not be illustrated and discussed. To improve this, we expect to set another smaller threshold and display it by different saturations. So, it can be addressed in a more detailed way.
The second visualization we will present is a dynamic one. There are two selection panels of this visualization. In the first one, you can choose the standard of ranking these countries. There are five options: Destination (by the number of people who immigrate to these countries), female (by the number of female immigrants), male (by the number of male immigrants), continent (by the continents the country belongs) and region (by the region the country belongs). In the second one, you can choose which type of data you expect to see. There are three options: by destination (the number of people who immigrate into this country), by origin (the number of people who immigrate out of this country) and by continent (the number of immigrants of this whole continent).
A gif of operating this visualization is shown here to better explain how it functions and to show more function of this visualization.
This is an example from the continent option. The countries are ranked by continent. We can observe that the circles with the same color are gathered together because the color stands for the continent these countries belong to. When you hover on the circle, you could see the name of the country. Each circle stands for one country. These circles have different colors, and they are constructed by a colorful filled circle and a circle line. Filled circles stand for the numbers of female immigrants. And circle lines stand for the numbers of male immigrants. In this way, we can easily compare differences in the number of female and male immigrants. The colors stand for different continents. The meaning of colors is illustrated in the table below.
Let’s see another gif from our exciting visualization!
This example is from the female option. We can observe that the circles of the same color are no longer gathered together. Because in this option, the numbers of female immigrants are considered as the ranking standard instead of the continent. Countries are ranked in ascending order. We can observe that the next circle is equal to or bigger than the circles before. We also can notice that the countries with bigger filled circles do not essentially have bigger circle lines. This shows that female and male immigrants may have different preferences on the countries they immigrate to.
From the general visualizations, it was evident that some countries have significant differences between male and female migration. Moving forward, we wanted to explore in more detail how the flow of migration behaves in time and for each sex. We decided to focus on one country that showed interesting behavior to start implementing visualizations and set a starting point for future designs with further interactivity that convey more information. India seemed to be one of those countries because it has a big difference in migration among men and women, and it also has big figures in the outflow and inflow of people to the country.
The first visualization was implemented using Svelte and shows the flow in time (from 1990 to 2019) for all countries that travel to or from India. Each path represents a pair of countries from which India is either the origin or destination. The path goes from a point in 1990 through the intermediate years where there is available data for migration, until 2019. In addition, it shows the female (blue) and male (red) migration path for each pair of countries and includes an interactive component by hovering over each line. This tooltip shows the pair of countries that correspond to the path in the form of “origin–destination” and highlights the path along the years.
The visualizations show that the bulk of the migration from and to India is focused below 5000 people for each year. We must keep in mind that the axis has a logarithmic scale because of the considerable differences in migration numbers within each country. Therefore, any -slight change in a path could mean an increase or decrease in several orders of magnitude. In the images above, we see that by hovering over the lines, the paths for the female migration from India to Italy and the male migration from Singapore to India are highlighted. The first one shows an increasing behavior especially from 1990 to 2005, but it has been relatively stable in the last decade. In the second case, the male migration has decreased slightly over the years, but it has also been stable in the last decade.
On the other hand, the images below showcase a particular case with a significant increase in migration: The outflow from India to Qatar seemed to have abruptly increased for both males and females over the years, especially between 2000 and 2005. Even though the migration for males is higher than for females every year, the behavior of the two is very similar.
Another design we aimed to explore was related to the behavior of the top countries that migrate to one country and the top countries that people from that country migrate to. Following the example for India, below we see an attempt at showing the top 10 countries on either side of the migration flow for India. We are still figuring things out regarding the encoding to get to the best way to convey the information without overlapping labels or shapes. In the image below, the vertical line for each triangle is proportional to the total migration in 2019, and we can see the migration figures by hovering each shape. We see that Bangladesh and Pakistan were the countries that migrated the most to India, and that the United Arab Emirates and the United States of America were the countries to which people from India migrated the most in 2019. Furthermore, the inflow of migrants to India seems to be more unevenly distributed in the top 10 countries compared to the 10 top countries in the outflow. We can also see that in general the outflow from India has bigger figures than the inflow.
The previous cases are just some examples of what we will be able to find by exploring the migration data. We are excited to go even further in the data, incorporating the information for well-being and violence indexes, and explore new designs in the next stages.
