Aggregation and Selection are compulsory processes that have to carry out with the data, regardless of the level of expertise of the user. The best way to get familiarized with the data is to check their attributes and look for possible relationships or patterns.

Furthermore, from the attribute tables, spatial selections, or statistical aggregation, a lot of information can be extracted without consulting additional sources, reducing time and resources spent.

In this lab, I will implement selection and aggregation processes to extract as much information as possible about the Colombian armed conflict, specifically, the forced displacement and children recruitment events. Those social problems are hard to approach, but for reconciliation between the armed conflict actors, there must be recognition (with statistics) of the damage that has been done.

For this lab, I worked with three layers provided by the " Centro Nacional de Memoria Historica ", regarding the historical forced displacement and children recruitment in Colombia, between 1980 and 2014. (Map I)

1.  Colombian_Municipalities : Contains the municipality's polygons with their corresponding Area values in km2.
2.  Children_Recruitment_1980_2014 : Layer with information regarding on how many children have been recruited between 1980 and 2014 in each municipality.
3.  Forced_Displacement_1980_2014 : Layer with the number of forced displacement events in Colombian municipalities between 1980 and 2014. Additionally, it contains Area values for each municipality in Km2.

That data is really important as Colombia faced an internal armed conflict for more than 60 years, until 2012 when a peace agreement was signed with the guerrilla "FARC". However, Colombian communities are still missing the most important: truth and recognition of them as armed conflict victims. This lab will try to share with communities around the world, how the conflict evolved through time and along the Colombian territory.

The following questions will be answered during the lab:

1. Using the “Forced Displacement 1980-2014” and “Colombian Municipalities” layers, determine the percentage of municipalities area in which there was at least 1 person displaced between 1980-1988.
2. For the 1989-1996 period, determine the total number of children recruited by the Guerrilla. Use the “Children Recruitment 1980-2014” layer.
3. For the period 2005 – 2014, identify the Municipality with the most displacement cases and identify the distance for the 2 ^nd  degree landlocked municipalities that the affected communities could have migrated to. Additionally, show the average distance.
4. What is the total of displaced people in Bogota municipalities for the complete period (1980-2014)?
5. Create a web map that enhances the visualization through aggregated data using heatmap symbology.

For this task, I filter the "Children Recruitment 1980-2014" layer. First, I filtered the records that occurred in 1989-1996, and then, just those that were carried out by the guerrilla. (Please, apply the interactive filters (F1) and (F2) in order).

Then, with field statistics for the field "Recruited Children", I determined that 170 children were recruited by the Guerrilla between 1989 and 1996. (Image I)

For this question, the "Forced Displacement 1980 - 2014" layer was filtered to display just Bogotá data. (Map II)

And then. using the field statistics for the "Number of Displaced" field, I determined that the total of displaced people in Bogotá was 9,882. (Image VII)

For this task, I applied first the heat mat symbology to the Children Recruitment (1980 - 2014). As you can see in the swipe below, when having the point with counts and amount symbology (Map III - Right), the data may be confusing, and is not easy to infer spatial relationships. However, when using the dynamic heat symbology (Map III - Left), the time spatial relationships are visualized with no effort. Specifically, in the Colombian context, we can observe that the central and western municipalities have been mostly victims of child recruitment.

Furthermore, you can extract additional relationships by combining heat maps with regular feature layers. In this case, we added the Forced Displacement layer, to overlay the heat map (Map IV). As result, we can evidence a direct relationship between the municipality's victims of forced displacement and the number of children recruited.

In this Lab, we explored several ways in which selection and aggregation can improve not only the way the data is presented to the end-user, but more importantly, the relationships that can be extracted with the received data.

Usually when we are working with any data or table, and we do not find the relationships/phenomena we are looking for, we search for more data and layers that may contain that information. However, as we demonstrated, those same relationships can be created directly from the raw data with less time and resources consumed. In the Colombian scenario, through selection and aggregation processes, a direct relationship between forced displacement and child recruitment was shown.

In conclusion, the data aggregation and selection processes must not be optional but mandatory. They have to be performed as the initial exploratory phase for unlocking all the data potential and improving data relationships.