Introduction to Statistics - A Graduate Programme at Three Levels
- Course Code:
- Unit value:
- Year of study:
- Year 1 or Year 2
The course will take place in the Spring term and last 10 weeks, including an introductory session to be held at the end of the Michaelmas term and an optional test/exam in the final week for anyone who is not taking the MRes in Anthropology (compulsory in that case).
Teaching will take place over four hours in the Computer Cluster Room at Christopher Ingold Building, UCL, see: http://www.ucl.ac.uk/maps.
Each week there lecturing will interleave with practical work over the course of the class. The final hour of the class will provide an opportunity for students to stay behind for a ‘surgery’ session with the TAs. This was we hope to convey most of the new ideas, in a hands on fashion.
Homework will be set each week – and marked by the TAs (no record of marks is kept for those not taking the exam as part of a MRes). This enables students to carry on practicing procedures after they have formally acquired them. The Moodle site also provides plenty of exercises (with ‘turn the page over answers’ available fro practice work).
Objectives and learning outcomes of the course
Piloted at UCL in 2006-2007, this course aims to provide anthropology students with proficiency in using, interpreting and presenting techniques for the analysis of quantitative data. It provides an innovative series of classes for those new to quantitative analysis, designed to complement ethnographic methods of analysis. The following areas are addressed: descriptive statistics; exploratory statistical data analysis; statistical inference and measures of association.
The objectives are to equip students
- to understand some basic principles of statistical analysis and sampling methods,
- to apply them appropriately in their own research,
- to read critically anthropological texts in which these methods are applied or referred to.
The overall goal is to begin to uncover the connections between statistical analysis and causal explanation, and the complementary use of quantitative and qualitative methods. Specific statistical tests will be taught, but most emphasis will be placed on understanding the reasoning behind them. Practical work will be based on a range of anthropological data sets with different substantive and technical features, and will include discussion of possible applications to students’ own research plans, including the practicalities of data gathering, and hands-on training in the use of SPSS. In line with student feedback we will begin the course with sessions on data collections, survey methods and data summary in tables.
Scope and syllabus
Week 0: introduction (end of term 1)
Overview of course themes: Quantitative and qualitative methods. Do these relate to specific subject matters or theoretical approaches? Reasons for using each. Illustrative example that highlights some key features of statistical reasoning. The history of hypothesis testing in social sciences.
Practicalities: Examination of the reintroduction of experimental methods into disciplines that have mostly relied on qualitative methods and interpretive procedures.
Themes for weeks 1 to 9
- How to draw up a questionnaire
- How to input data into SPSS/ Excel
- How to summarise and report on data
- Significance testing: the chi-square, t-distributions, Anova tests
- Correlation and regression
- Multiple regression – an elementary introduction
- Controversies about the use of statistical comparisons across societies – the Tylor-Galton debate revisited
- Sampling in an ethnographic or small scale context, randomisation for beginners
- The analysis of short causal chains
In week 10 there will be an optional exam.
Method of assessment
Students will be assessed formally or informally (according to need) on exercises distributed in the weekly practicals and finished in their own time.
For those – and only those – taking a MRes in Anthropology and any others who see benefit in using a final test as a means to assess progress we will set a two-hour examination at the end of the term (50%), with 50% also attached to the assessment of a portfolio of practical and home work produced during the term.
Methodological and statistical texts
Bernard H R 2006 Research methods in anthropology, qualitative and quantitative approaches. Fourth Edition. Oxford: Altamira Places statistical methods in the context of anthropological methodology as a whole. Highly opinionated and fun to read. Theoretically a bit weak but generally sensible judgments.
Gillies D 2000 Philosophical theories of probability. London: Routledge. A discussion of whether and why the probability arguments underlying statistics actually work. Rather technical , but nearest in spirit to the ideas underlying this course.
Heady P 2007 What can anthropological methods contribute to demography – and how? Demographic Research Vol 16, article 18, pp555-558
http://www.demographic-research.org/volumes/vol16/18/ A brief non-technical discussion of the relation between ethnography, theory and quantitative measurement. Despite the title it is not really about demography.
Madrigal L 1998 Statistics for anthropology. Cambridge University Press. Sets out the basic capabilities that the course aims to teach. Reliable but a bit flat. If you buy one book, this should be it.
Marsh C 1988 Exploring Data. An introduction to data analysis for social scientists. Cambridge: Polity Press Gives an excellent introduction to exploratory descriptive analysis.
Nelson D 2004 The Penguin dictionary of statistics. London: Penguin. Clearly written reference book. Uses enough algebra but not too much.
Pearl J 2000 Models, reasoning and inference. Cambridge University Press. Relevant to the ideas in lecture 9, but extremely technical.
We will certainly refer to
Goody J 1976 Production and reproduction: a comparative study of the domestic domain. Cambridge University Press.
Other texts will be introduced during the course.