Statistics is the art of drawing conclusions about phenomena in which chance plays a role. The randomness may arise through a variety of reasons: the intrinsic random nature of a phenomenon, unavoidable noise in an experiment, conscious randomization of experimental or measurement units, or as a best approximation to reality. The chance phenomena occur in a broad range of situations. This has rendered statistical science a highly multidisciplinary undertaking, but with a core body of concepts and methods that are common to the diverse applications.
The master track in Statistical Science provides students with a thorough introduction to the general philosophy and methodology of statistical modelling and data analysis, and offers two specializations: into the applications of statistical methods to the life sciences and to the behavioural sciences, respectively.
Statistics for the life sciences is almost synonymous with biostatistics. It incorporates quantitative modelling and methods of data analysis for clinical and epidemiological research (e.g. survival analysis), which in the past twenty years have become indispensable in medical research. It also includes statistical methods used in genetic research and genomics, which have a classical foundation (for instance in the work of Fisher, the founding father of statistics), but are rapidly developing in answer to present day opportunities given by data from new experimental platforms, such as micro-arrays or whole-genome scans. The program is targeted both at human and at plant or animal genetics. In the coming years systems biology will make similar demands for new statistical methodology, and the analysis of medical images will increase in importance, both in research and in clinical applications.
In the social and behavioural domain there is a long-standing statistical tradition in educational and psychological testing (psychometrics), and also in survey research, marketing research and quantitative demographics (sociometrics). Similar subdomains that emerged more recently are the quantitative study of the development of science and technology (scientometrics and bibliometrics), the quantitative study of stylistic forms and patterns in the use of language (stylometrics), the quantitative study of taste and smell (sensometrics), quantitative study of history (cliometrics), and the empirical approach to the law (jurimetrics). The common use of the term "metrics" here illustrates the important role of measurement problems in these fields. More generally, it is no exaggeration to say that all empirical research in the social and behavioural sciences relies predominantly on statistical analysis. Forensic statistics is another important field of application.
Whether more attracted to the medical or to the behavioural direction, the successful student will gain a thorough understanding of statistical models, their implementation and their interpretation, and develop the ability to invent new models and techniques when needed. Graduates will thus qualify for jobs in a wide range of areas, such as academic medical hospitals, many types of industry (pharmaceutical, agricultural, food, life science in general, oil, etc.), research institutes, financial institutions, government statistics bureaus, educational services (CITO), marketing bureaus. In view of the emphasis on statistics as a (mathematical) science, rather than as merely a collection of techniques, many graduates will qualify for PhD programs as well.
In view of the multidisciplinary nature of statistical science, it is an advantage that students have experience in another science before they enroll in statistics. The entrance requirements (see below) are therefore formulated in broad terms. However, students should feel at ease with and be attracted to quantitative and mathematical techniques and concepts. In other terms: they must have an appreciation for abstract thinking, involving use of mathematical tools and language, and they should have a decent degree of numeracy.
'I keep saying the sexy job in the next ten years will be statisticians. People think I'm joking, but who would've guessed that computer engineers would've been the sexy job of the 1990s? The ability to take data &ndash to be able to understand it, to process it, to extract value from it, to visualize it, to communicate it &ndash that's going to be a hugely important skill in the next decades...'
Hal R. Varian (UC Berkeley), Google's Chief Economist, The McKinsey Quarterly, January 2009
