Open, Transparent, Reproducible

At Mazama Science we believe that publicly funded data and analysis should strive to be as open, transparent and reproducible as possible.

Open in the sense of freely accessible but also in the sense of open source where the data and analysis software are available at zero cost. The Population databrowser uses publicly available data from the US Census Bureau and relies on open source R for analysis and plotting. Users wishing to run the R scripts below should try out RStudio — an open source IDE for R.

Transparent is the word used to describe data graphics and user interfaces that don't need a lot of explaining. We have done our best to make using this databrowser as effortless as possible. With careful attention to variable naming and code structure we hope that the source code provided below is, if not always transparent, at least not opaque.

In elementary school we learned that science and engineering should be Reproducible. Sadly, this is not always the case as analyses are often reported without any way to assess their validity. The data and analysis scripts below offer you a chance to reproduce the results seen in the Population databrowser

Source Code

Raw IDB data and conversion code:

IDBext001.txt
original mid-year population file from IDB
fips2iso.csv
table mapping between FIPS and ISO country codes
convertMidYearPop.R
R script to convert raw IDB data into restructured, ISO compliant MidYearPop.csv

Multi-language plotting code:

main.R
run this script in an R session to generate the plots seen in the databrowser
populationOverTimePlot.R
R script with the plotting function
MidYearPop.csv
mid-year Population data from the IDB, restructured and using ISO country codes identifiers
estimateUpdateYear.csv
hand generated table containing the most recent year each country was updated according to the IDB release notes
language.json
python babel/Google translate/hand modified structure enabling foreign language versions

Feedback

population@mazamascience.com

We look forward to feedback on how we might improve this tool. Please understand that this is a pro bono project and that requests for additional functionality will be appended to our own wish list of functionality awaiting some free time on our part.

Offers to fund improvements are always welcome.

Thanks to Lucy Williams who did some early work on this project.

Jonathan Callahan and Will Leahy -- May, 2014.