Century Birthday

Century Birthday

Organisation: La Nación (Costa Rica) (Costa Rica)

Publication Date: 04/07/2016

Size of team/newsroom:large

Description

Century Birthday reviewed 115 years of the history of births in Costa Rica. The leading goal was determining, upon a historical basis, in which month the most births happen. The special made it possible to demonstrate that the assertion that September was the month when the most children were born in the country was a myth. October, actually, was the king of births between 1900 and 2014. Also, the application makes it possible to compare the number of births per canton and provides information on the historical, political, social, economic, and health environment in which births have taken place in the study period. Users can enter their date of birth and learn how many other Costa Ricans celebrate their birthday on that date. Moreover, through a statistical model, it was possible to estimate the behavior of births in the country during the next decade. The video which discloses what potentially might happen with births, month by month, in the next decade, is based upon an analysis of times series between 1900 and 2014. The model was built with the assistance of Erick Rodriguez, a professional in Business Intelligence and Data Mining. To that end, the R program was used. It is important to bear in mind that this outlook is an academic exercise and that its outcome may vary in the future, depending on the economy, new contraceptive methods and lifestyles, among other factors. The true monthly information of 115 years of births in Costa Rica became a time series, which was divided into two parts: one for learning and another for trial. With the trial series, several versions of the model were developed and the one that provided the best result in relation to error and comparison with the original (true) series was selected. The winning model was used for the first time with the data from the whole series, and the prediction was made. In addition, the total of births from the current year through 2024 was completed with those results. The estimated data for those 10 years were included in a series in which, starting in 1950, the gross birth rate (births per each 1,000 inhabitants) was estimated. This with the aim of showing that, even if the number of births increases and there is more population, the number of babies per couple will be increasingly lower. The projections of the population from the Central American Population Center for the period between 1950 and 2024 were used to estimate the rates per 1,000 inhabitants. An analysis of hierarchical clusters, using Euclidean distance, was applied also to the series of data with the years and overall births. The aim was creating groups of years whose behavior had been similar regarding the number of births. The aim was creating groups of years whose behavior had been similar regarding the number of births. The fourth among them broke the year sequence and showed that the number of newborns in the last decade and a half is similar to the one in the late 1970s and early 1980s.

What makes this project innovative? What was its impact?

“Century Birthday” is an application through which users learn at the same time that they play and interact with the data. The visualizations of data and videos are the leading vehicle to tell the story and inform in an entertaining way. In addition, users can take a look at the future and learn what the reality of births in the country will be in the next decade. Based on the analysis of what happened with births in 115 years (1900-2014) and with the help of a data scientist, a time series model was built and it enabled estimating the number of births through 2024, as well as the month in which more babies will come to the world and the one in which fewer children will be born.

Technologies used for this project:

The building of the interactive site involved using programming languages such as HTML5, CSS3, Highcharts, CanvasJS, Tableau Public, JSON, AJAX, SVG, Jquery, JS, AnimateNumber JS. The tools used included Brackets, After Effects, Adobe Illustrator, Adobe Photoshop, Sketch App, Pngyu App, Invision App, paper and pencil. Microsoft SQL Server 2008 R2 Enterprise Edition x64 bits was used as the main Database Manager System. To generate the thousands of JSON used to upload the data in the interactive site, the Microsoft SQL Server Business Intelligence Development Studio (SSIS) technology was used as a platform which allowed generating high-performance data-integration solutions, among which packs for mining, transforming and loading data were included. All told generation was: 1. Births per canton, month and year: 10.692 JSON File a. 81 cantons b. 12 months c. 11 years d. 2003 – 2013 period 2. Births per month: 12 JSON File a. 1900 – 2014 period 3. Births per month and year: 1.380 JSON File a. 12 months b. 114 years c. 1900 – 2014 period

Video

Featured Documents

Follow this project
Wait
Comment

Comments (0)

You have to be connected to contribute

You have to be connected to follow

Leave this project and no longer be informed about this project

By joining this project, you will be informed by email when an update or a new contribution is posted on the website.

Thank you for your active participation !
Best,

The GEN Community Team