Statistical analysis seems to logically need to involve statistics correct? Of course. However, to conduct statistical practices, there first needs to be data. A great source of data happens to be databases (groundbreaking idea, I know). For my project, there are various databases that can be applied to provide adequate data for analysis. One of these is the World Development Indicators database, hosted by the World Bank.[1]
The World Development Indicators database provides data to help define the variable of “development”.[2] This variable may not necessarily be connected with the number of operating hours for metro systems, so while I can use it as a placeholder for an independent variable, it would be more fitting as a control variable, so that I can account for changes due to the development of any given country.
The database has many indicators to help show what “development” may be. Personally, I think that using GDP or GDP per capita (measured in U.S. dollars) would be a good indicator for showing development in the context of my project.[3] These indicators would come into play as indicators for how much money there may be flowing around a country to be used in any given metropolitan government to fund their respective metro systems.
Due to its backing by the World Bank, the World Development Indicators database does cover every country as a specific case for its various indicators.[4] Even still, this database is not the most helpful for my project, as the database is generally concerned with the countries themselves, while I am more concerned with metropolitan metro systems within each of these countries, and this database does not get into the more specialized details of a region’s or a city’s GDP or GDP per capita.
Moving forward, I will be looking to create my own dataset with information that I will be gathering about metro systems around the world. Some of the independent/control variables that I am thinking about including are age of the system, when the last line was added, and the aforementioned GDP/GDP per capita indicator of development. In terms of dependent variables, I would be looking at the operating hours of each metro system (with specific data for each day if there is variance within the system’s schedule).
Anyways, until next time!
Nathan
[1] “World Development Indicators | World Bank,” accessed 25 October, 2017, http://databank.worldbank.org/data/.
[2] Ibid.
[3] Ibid.
[4] Ibid.
Bibliography:
“World Development Indicators | World Bank,” accessed 25 October, 2017, http://databank.worldbank.org/data/.
Nathan — you have a good discussion of the data in the WDI database as well as some of the limitations in coverage and precision that would matter for your research. Remember that you need to be discussing the *level of measurement* for each variable/indicator as well, though! Data at the regional or local level is notoriously hard to come by, so you will likely have to collect a lot of that information from disparate sources to compile your own dataset. If you were to do so, what would the level of measurement for your proposed DV be?