Research Portfolio Post #6: Quantitative Data Sources

I am studying global climate initiatives because I want to find out what explains the variation in implementation of these initiatives in order to help my reader understand whether or not international action to combat climate change is effective and how solutions can be crafted to create equality in burden sharing and make significant environmental progress.

In a large-N statistical analysis of my project, my dependent variable would be how much a country is lowering their greenhouse gas emissions. According to the Environmental Protection Agency, the four main greenhouse gases are carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), and fluorinated gases.[1] In measuring my dependent variable, I would use the ‘total greenhouse gas emissions’ statistic measured in kilotons of carbon dioxide equivalent from World Bank’s DataBank, which includes the four main greenhouse gases as listed by the EPA, and is sourced from the European Commission’s Emission Database for Global Atmospheric Research.[2] This database gives emissions levels by year, which I could use as interval data. This would be my dependent variable because many global climate initiatives, such as the Kyoto Protocol and the Paris Agreement focus on slashing greenhouse gas emissions or even creating binding emissions reduction targets.[3] I could also narrow down this variable into specific gas emissions, such as Perfluorocarbon which is a byproduct of certain manufacturing processes, or narrow it down to specific sectors of the economy, such as agricultural emissions or emissions from transportation.[4] The cases I would study with this dependent variable are all 264 countries and regions that the World Bank’s World Development Indicators dataset has information on.

The limitations of this data set are that it does not have data on certain countries for every year. For example, there is no emissions data for Afghanistan from the year 2013 to present day, due to conflict in the region.[5] Another limitation is that the data excludes emissions from some biomass burning, like the incineration of agricultural waste, which is a large source of greenhouse gas emissions, as well as other gases which are not included in the main four greenhouse gases.[6]

[1] “Overview of Greenhouse Gases,” Overviews and Factsheets, US EPA, December 23, 2015,

[2] “World Development Indicators | DataBank,” accessed October 11, 2019,

[3] Hiroki Iwata and Keisuke Okada, “Greenhouse Gas Emissions and the Role of the Kyoto Protocol,” Environmental Economics & Policy Studies 16, no. 4 (October 2014): 325,; Jana Lippelt and Lea Mayer, “After the Paris Agreement – What’s Next? Worldwide Implementation,” CESifo Forum; München 18, no. 4 (Winter 2017): 43.

[4] “World Development Indicators | DataBank.”

[5] Ibid.

[6] “Air Pollution | Partnership for Policy Integrity,” accessed October 11, 2019,

2 thoughts on “Research Portfolio Post #6: Quantitative Data Sources”

  1. Hello Carly,

    I think the data you found is very strong. I definitely see how context is going to play a key role in developing these variables. I thought it was very interesting that you pointed out that there are gaps in the types of green house gas pollution that us recorded. Are you finding that certain types of greenhouse gas or pollution are being recorded more than others? If so, have you read any literature on why that might be? I guess I am wondering if some forms of pollution are easier to track or if there are political or economic reasons for why this information can be difficult to track.


  2. Overall an excellent job here, Carly! The data sources that you discuss are clearly relevant for your project, and you have given some good thought to how you would operationalize your DV for this methodology. The missing data points that you mention are worth noting, to be sure, but on the whole it sounds like the data coverage that you have for this DV is much more comprehensive than is often the case. My only suggestion here is to think carefully about what your cases would be for a project in this methodology. You note “The cases I would study with this dependent variable are all 264 countries and regions that the World Bank’s World Development Indicators dataset has information on.” — but remember, your cases all have to be the same type of unit. You could compare and analyze for variation across countries, or you could compare and analyze for variation across regions, but not both in one analysis. Which of those units of analysis would make more sense for your project?

Leave a Reply

Your email address will not be published. Required fields are marked *