For my research project, I will be evaluating quality of life as my dependent variable, and international community-funded urban development projects as my independent variable. The primary dataset I will be using to operationalize my dependent variable is the UN-Habitat Urban Data database [1]. I will be comparing data across my cases, cities in Asia. I will be controlling the data based on region, so I will eventually create a dummy variable for “Asia” to be operationalized as a 0 or 1, however I will be observing data from all other continents. This database allows me to select the indicator “Slum Dwellers,” so that my data points are specific to the populations I am studying within each city. It also provides specific indicators of quality of life.
I am breaking down quality of life into three main indicators: health, shelter and economic well-being. Regarding health standards, I will be observing the size of the urban slum population, proportion of population with access to an improved water source, access to an improved toilet, proportion of population whose solid waste is collected (by public or private services). In terms of shelter, I will be observing proportion of urban population with durable housing, connection to electricity, and proportion of urban population living in a slum area. In terms of economic well-being, I will be observing income Gini coefficient.
The second dataset will be using to test development over time. The WIID – World Income Inequality Database from United Nations University [2], gives a more comprehensive overview of economic wellbeing. This dataset actually draws numbers from multiple other datasets from different sources. It also indicates Gini Index over the last few decades. I will use this to create “snapshots” that will show progress, or lack there-of over time. The first time interval will be 1978, when Habitat I—the first UN Habitat Agenda—was released. Habitat II was drafted in 1998, the second time interval. Habitat III was just recently released in 2016, the third and final time interval.
Notes
UN-Habitat Urban Data, distributed by UN Habitat, http://urbandata.unhabitat.org
Carlos Gradín, WIID – World Income Inequality Database (January 2018), distributed by United Nations University, https://www.wider.unu.edu/project/wiid-world-income-inequality-database
Hi Naila,
It looks like your large-n research is off to a good start. I’m just wondering, however, why your dummy variable for Asia is not expanded to a dummy variable for all different regions of the world, similar to what Ross did in his article, especially since you note that you will be looking at worldwide data. This seems like it might help you more and give you more information to work with even if you only want to look at the effects in Asia, just like Ross only wanted to see the results in the Middle East.
I look forward to seeing where your research takes you,
Phoebe
Naila — the datasets and data sources that you discuss here are suitable for your project, so that is a good start. However, you should keep thinking about how you would operationalize your DV since, as we’ve discussed, operationalizing the DV as an *interval-ratio* variable is really what is required for this methodology in order to maximize the potential of statistical analysis. As Phoebe notes, having more cases is better than having fewer, so global data with a control variable for region would likely be preferable to limiting the data/sample to a particular region.