Qualifying Exam on Applied Statistics



The qualifying exam on applied statistics is an open book exam --- you can use whatever books and notes you bring with you. The exam is based on the books Statistical Concepts and Methods, by G. K. Bhattacharyya and R. A. Johnson and Beyond ANOVA: Basics of Applied Statistics by Rupert G. Miller, Jr. Although it is not required that you bring these books to the exam with you, the exam might refer to specific pages in the books in order to clarify terminology. (In some places the use of terminology in Bhattacharyya and Johnson is perhaps a bit nonstandard. The exam uses the terminolgy from Bhattacharyya and Johnson, but also clarifies nonstandard terminology.)

You can use a calculator (but not a computer), and it is recommended that you bring a calculator with you (although it is not assumed that you will have a calculator that has special statistical capabilities).

The exam may require tables associated with the standard normal distribution, various chi-square, t, and F distributions, and also tables of the exact distributions for the sign test, the signed-rank test, and the rank sum test. These tables will not be supplied with the exam --- you are expected to bring approprate tables with you to the open-book qualifying exam. (The tables in the appendix of Statistical Concepts and Methods by G. K. Bhattacharyya and R. A. Johnson will be sufficient, but tables from other books can also be used.)


The exam covers material in the following parts of the book Statistical Concepts and Methods, by G. K. Bhattacharyya and R. A. Johnson: The exam also covers material in the following parts of the book Beyond ANOVA: Basics of Applied Statistics, by Rupert G. Miller, Jr.: While Beyond ANOVA is packed with information, the exam will focus on the main points of the book (as well as the material in Statistical Concepts and Methods), and not the minutia. You should become familiar with the uses, the strengths and weaknesses, and the robustness of the inference methods presented (noting that for each setting, a "normal theory" procedure is typically presented, as well as some nonparametric and robust alternatives).