Quantitative Data Sources

During my last meeting with my mentor, I mentioned to her that we had an assignment to find datasets for the large-n methodology which would eventually be used for an outline of large-n research. She informed me that she was unaware of any large-n analysis that had been done in the field of refugee studies. So, going into this stage of research, I was worried I would not be able to find any data for my hypothesis, my fears were not assayed by the research I did. I could not see any worldwide or even regional data looking at refugee populations demographics, so I was forced to change my question.

Given these limitations, I shifted my question to look at whether or not an increase in refugees leads to a rise in far-right activity. My dependent variable for this question would be far-right activity shown through the number of hate crimes in a country and election results. Since there is no international database on hate crimes, I would determine if there has been an upward trend in the number of hate crimes through looking at individual countries data on hate crimes, expanding on the research done by the Fundamental Rights Report, which only covers nation which have agreed to work with them, mainly Western countries.[1] Additionally, I would look at election results, specifically if any far-right parties have been elected to any level of government, if there has been an upward trend in the percentage of votes that far-right groups receive, and if there are any alt-right grassroots movements in the country—the current data I found for this variable focuses on the Western world.[2] However there, if the term “alt-right” is expanded to include more populist movements with racist elements, more nations can be covered in the data.[3] One limitation of looking at hate crime statistics is that different organizations and countries collecting data on this have different definitions of what a hate crime is and therefore are not exactly comparable although the big picture is comparable. A limitation of looking at alt-right groups is that “alt-right” does not have a set definition and therefore is up to researcher or reporters’ determinations.

The independent variable is the number of refugees entering a country[4], one limitation of this data is if it a refugee is counted but they are in transit or if the country is not traditionally a host nation. The control variables are GDP per capita[5], religion[6], and country/region.


[1] “Racism, xenophobia and related intolerance,” European Union Agency for Fundamental Rights, accessed 14 October 2018, http://fra.europa.eu/en/publications-and-resources/publications/annual-reports/fundamental-rights-2017#racism

[2] “Europe and nationalism: A country-by-country guide,” BBC, accessed 14 October 2018, https://www.bbc.com/news/world-europe-36130006

[3] Christi van der Westhuizen, „South Africa’s white right, the Alt-Right and the alternative,” The Conservation, accessed 14 October 2018, https://theconversation.com/south-africas-white-right-the-alt-right-and-the-alternative-103544

[4] “Population Statistics,” UNHCR, accessed 14 October 2018, http://popstats.unhcr.org/en/overview#_ga=2.260496072.433466411.1539549315-482120770.1536177183; “The number of displaced people in the world just hit a record high,” World Economic Forum, accessed 14 October 2018,  https://www.weforum.org/agenda/2017/06/there-are-now-more-refugees-than-the-entire-population-of-the-uk/

[5] “GDP per capita (current US$),” The World Bank, accessed 14 October 2018, https://data.worldbank.org/indicator/NY.GDP.PCAP.CD

[6] “Population by religion, sex and urban/rural residence,” UNData, accessed 14 October 2018, http://data.un.org/Data.aspx?d=POP&f=tableCode%3A28

4 thoughts on “Quantitative Data Sources”

  1. Hi Phoebe! It sounds like you’ve shifted your question to something really interesting.
    I think you make a really good point about hate crimes and how the classification differs between countries. One idea to control for that would be to look at EU official data, like what I found in this file: file:///Users/alexiagardner/Downloads/fra-factsheet_hatecrime_en_final_0.pdf (let me know if the link doesn’t work and I can email it to you).
    Looking forward to the small-n approach, you may be interested in the terrorist attack in Norway on July 22nd, 2011. I just watched the fictional portrayal of it on Netflix, and it seems to fit your topic well, as it was a large-scale, violent attack motivated by the increasing number of immigrants and refugees in Norway.

  2. Hey there Phoebe,
    This new direction sounds really interesting to me. I think it was probably wise to adjust your topic for this particular research design, considering the lack of data, but I think this direction could produce some fascinating conclusions. Someone in SIS you might want to talk to about the far-right dimension in Europe specifically is Lucas Dolan. He was my TA for World Politics last year, and his PhD research deals with how far-right movements spread and grow. I’m not sure if he deals with refugees in particular, but he would definitely have some good resources for learning more about far-right groups and could likely point you to some useful data sets. I would also just highlight that it doesn’t look like that BBC data in your second note is easily translatable into an Excel or SPSS format, and I’m not sure that the variables in it are standardized to all cases it includes. It might be worth looking into where the BBC pulled that data from to get a more formal look into how they compiled those maps. Best of luck and looking forward to seeing where this goes!

  3. Hi Phoebe!

    I really like how you framed this question, that is, it was born from greater contemplation and consideration on your topic, as well as mature intellectual flexibility. This would be a really interesting phenomenon to study, as inflamed political tensions these days hold this topic in high regard.

    One thing I would be careful about is defining the term hate crimes as part of your dependent variable, which you already seemed to have considered. As you go forward with seeking statistical data on hate crimes, you could try and look up crimes against particular groups more broadly, since they might not necessarily be tagged as “hate crimes” (kind of like how today in Inglehart had to define mass support for democracy with kinds of sub-indicators to achieve the effect of support for democracy as he defined it).

    Another dimension you could add in defining hate crime could include kinds of cyberhate, and the forms that online bigotry takes in refugees’ lives. Are these posts considered hate crimes, or simply harassment? What is the difference? Some scholarship on this includes “Mobilizing Against the Other: Cyberhate, Refugee crisis and Proximization” by Kopytowska, Grabowski, & Wozniak, and “Conceptualising the Other: Online discourses on the current refugee crisis in Cyprus and in Poland” by Baider & Kopytowska.

    I am looking forward to seeing how your research evolves, and the kind of sources you engage with to measure this phenomenon!


  4. Phoebe — the shift in focus for your project that you suggest here would certainly be feasible, and you discuss suitable data sources for the variant of your project that you discuss here. I would point out, though, that there is quite a bit of data available on refugee flows and related topic from organizations like the UNHCR, OECD, etc., that would allow you to collect the type of data needed to operationalize and analyze a DV more directly related to refugees. Whether or not there is existing large-n research in this vein is not as critical as whether there is available data for you to construct your own potential large-n analysis (and there is!)

Leave a Reply

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