As a note before I begin, I have decided to make modern day development my sole independent variable and removed stability as a variable, due to concerns about the scope of my research.
When examining various data sets, I found that the source World Development Indicators contained a number of data sets that I would be able to use in my research.[1] This then led me to focusing on the different aspects of development. I am attempting to determine how I am going to define development: for instance, whether I should focus on economic or social factors, or whether it would be possible to focus on multiple. For this reason, I examined a number of data sets which focused on each of those factors. One data set which I would like to discuss for my research sketch is the GDP per capita growth for Latin America.[2] This is dataset provides data for the economic development of Latin American countries. The level of measurement is the annual %, making it an interval ratio variable because the data gives precise measurements which can be tied back to real world values.[3][4] As discussed in class, this is the type of data that we should use, particularly for our dependent variable, which is a benefit of this dataset.
However, this dataset also has limitations. For instance, it only addresses one aspect of economic development, which does not encompass the entirety of my dependent variable. This means, I would need to include other datasets in my operationalization, such as a different dataset that I found in the World Development Indicators: Share of youth not in education, employment or training, which could address other types of development, which addresses social development.[5] Another limitation is the number of cases this dataset covers. It has data from the last 5 years for Latin America, and while it has data for all years for some countries, such as Argentina, other countries like Cuba had only partial data provided. In particular, Venezuela only had data for the years 2013 and 2014, which was specifically concerning. For this reason, while technically my data covers all 20 countries that I am examining, it only completely covers 17 of my countries. This is of note because I noticed that the countries which are not covered are frequently those who have had more development issues, which could skew my data and analysis.
[1]World Bank. “World development indicators database.” (2010).
[2] World Bank national accounts data, and Organisation for Economic Co-operation and Development National Account data files. GDP per capita growth (annual %). 2015-2017. Distributed by World Development Indicators. http://databank.worldbank.org/data/reports.aspx?source=world-development-indicators#
[3] World Bank national accounts data, and Organisation for Economic Co-operation and Development National Account data files. GDP per capita growth (annual %). 2015-2017. Distributed by World Development Indicators. http://databank.worldbank.org/data/reports.aspx?source=world-development-indicators#
[4] Pollock III, Philip H. “Measuring and Describing Variables”. In The essentials of political. 35-40. Cq Press, 2015.
[5] International Labor Organization, ILOSTAT database. Share of youth not in education, employment or training (% of youth population). Sept. 2017. Distributed by World Development Indicators. http://databank.worldbank.org/data/reports.aspx?source=world-development-indicators#
Overall these data sources and datasets seem quite suitable for your project, Katherine. Remember that you will want to hone in on one or two indicators that really capture the essential aspect of development that has the most interesting/puzzling variation as you continue your research (and remember that you are operationalizing your *dependent* variable here, not an independent variable). Trading off conceptual depth for the ability to analyze patterns across wide swaths of data is one of the core tradeoffs for this methodology, which again means thinking about one or two focused indicators for your DV rather than many.