What is this Module about?
This module is a core component of the MA Politics research training programme. It offers an introduction to data analysis and interpretation. It has two aims. Its primary purpose is to provide students with the skills needed to understand, interpret, and critically assess statistics and other empirical data presented by the news media and in academic journals. A second purpose of this course is to provide students with a solid foundation to conduct own empirical analyses and to familiarise students with the complexities inherent in analysing large datasets (such as the British Election Studies, World Values Surveys, Afrobarometer), using SPSS/Win (Statistical Package for the Social Sciences in the Windows environment). By the end of the module students will possess the ability to undertake quantitative research and use computing programmes to analyse and interpret data in Politics.
The sequence of topics in brief
- Descriptive Statistics, Measures of Central Tendency and Dispersion
- The Normal Curve
- Inferential Statistics
- Hypothesis Testing
- Bivariate Analyses
- Measures of Association
- Multivariate Techniques, Multiple Regression and Correlation
- Direct, Intervening, Spurious Relationships and Interaction
- Einspruch, E.L., An Introductory Guide to SPSS for Windows, Sage, 2005.
- Healey, J., Statistics: a Tool for Social Research, Wadsworth, 2005.
Required Journal Articles:
- Blais, A. and R.K. Carty (1996), 'Does Proportional Representation Foster Voter Turnout?' in European Journal of Political Research, vol. 18, pp. 167-181.
- Lipset, S.M. (1959), ‘Some Requisites of Democracy: Economic Development and Political Legitimacy’ in American Political Science Review, vol. 59, pp. 69-105.
- Field, A., Discovering Statistics Using SPSS, Sage, 2005.
A great deal of useful reading material is to be found in a wide range of science journals or periodicals. As a matter of routine you should consult the most recent issues of a number of journals as they come into the library and establish for yourselves whether they contain pertinent articles. Examples of journals with statistical analyses: American Political Science Review, European Journal of Political Research, British Journal of Political Science, Comparative Political Studies.
Method of assessment
Assessment is by the standard pattern for MA/Diploma option modules, namely one assessment essay of 5,000 words. For the assessed essay, you can either choose a title from the Assessed Essay title list below, or alternatively you can negotiate your own title. You are strongly encouraged to make your own choice of title. Consultation well in advance of the department’s title submission date allows preliminary proposals to be given proper consideration and refined as appropriate. It is essential that you carry out a library search in advance of choosing a title, even when a title is selected from the list included in this document. It is also essential to show in your essay that you fully understand the different quantitative techniques (as discussed during the module) and can apply them in an appropriate way. In other words, when answering the research question, the student should test hypotheses, and use cross-tables, charts, measures of association, correlation and regression analyses. Important: please check whether the needed data sets are available before you choose your title and topic. Large datasets (such as the British Election Studies, World Values Surveys, Afrobarometer), and SPSS can be used to answer a particular question in the essay.
Outline of the final research essay
The final research essay should contain at least the elements as described below.
What is your research question, and why is it interesting to scholars? Review some basic literature to demonstrate the importance of your question. Explain the logic behind the suspected relationships, e.g. why should we expect an impact of education on voting behaviour? What are the theoretical ideas? What are your null hypotheses? Your study should include a single dependent variable and at least four independent variables. What are the independent variables, dependent variable, control variable(s), cases, data and used data sets?
(II) Quantitative Data Analysis and Interpretation
(a) Univariate analyses and interpretation
Take each variable one at a time and describe the concepts and measurements of the variables in the research. Concepts: How do you define the variables? E.g. how do you define ‘education’ and ‘voting behaviour’? Measurements: How do you measure those concepts? In other words, what operational variables are used to represent these concepts and how well do they do so? What are the levels of measurements of the variables? Describe each variable in the appropriate way, that is, with frequency distribution tables, pie charts, bar charts or line charts. Which measures of central tendency are appropriate to describe these variables?
(b) Bivariate analyses and interpretation
Take each bivariate relationship one at a time and describe the three characteristics of each bivariate relationship: the existence, strength, and pattern/ direction. Use any of the appropriate measurements of association to analyze and describe each bivariate relationship. Describe the results in non-technical language. If your null hypotheses can be rejected, what statistical results would you expect to find? Run the statistical tests and decide whether each alternative hypothesis can or cannot be supported. Describe the results in non-technical language.
(c) Multivariate analyses and interpretation
Make the assumption that all your variables are measured at the interval-ratio level. Thus assume that your single dependent variable and at least four independent variables are all interval-ratio variables, although this may not be true in reality. To what extent can your independent variables explain your dependent variable in multivariate analyses? Calculate and interpret slope (b), Y intercept (a), and the multiple correlation coefficient (R2). Use regression and correlation techniques to analyze and describe the relationship. Describe and explain the three characteristics of the multivariate relationship in the research for this essay: the existence, strength, and pattern/direction. Describe the results in non-technical language.
Were any results unexpected – not supporting a hypothesis, or not supporting it as strongly as you expected? If so, why might this be? Were their possible problems in concept measurements, your logical rationale for expecting a relationship, or confounding influences? If you had ample time and resources, what other things would you do to explore your research topic question? That is, what extra variables might be included, which concepts measured more precisely, etc.?
Some examples of assessment essay titles
The list below is indicative of topics that some students have chosen to write about in the past. It is not a list from which you are required to choose a title and not a list from which you are recommended to choose a title. Please notice that the topic of your essay does not only depend on your personal interests and theoretical notions, but also on the availability of the data.
- Economic dependency and poverty: a quantitative analysis of developing countries
- Economic development on support for democracy in South Africa in the late 1990s
- Do consensus systems perform better than majoritarian systems?
- Voting behaviour in Japan: explaining the decline of leftwing parties
- Support for the European Union in France, the UK and Germany
- Ethnic heterogeneity and war in Africa
- The rise of extreme right in Western Europe
- Unemployment and inflation in the USA
- The relationship between class, religion and voting behaviour in the Netherlands
- The impact of ethnicity on democratic sustainability in India