Program Updates

Undergraduate Program

Christine Franklin, Senior Lecturer & Undergraduate Coordinator

Since the last newsletter where we listed our graduates through May 2011, we proudly graduated the following students with a B.S. degree in Statistics: summer 2011 – William Grimes and Rina Patel; fall 2011 – Jeff Anderson, Daniel Kreisler, and Alexander Lyford. We anticipate up to 12 graduates in spring 2012 and summer 2012.

Several of our students continue to be inducted into honor societies such as the National Statistical Honor Society Mu Sigma Rho, Phi Kappa Phi, and Phi Beta Kappa. Some of the students will continue with graduate school and others are planning on joining the workforce using the skill set they gained as a statistics major. Our students are sought out future employees in this data centric world. Congratulations to all our graduates!

Our undergraduate program continues to thrive with UGA students recognizing the importance of a background in statistics – this is seen in the increasing number of minors. Our capstone course also continues to thrive. This past fall, an article appeared in the The American Statistician entitled, “A Capstone Course for Undergraduate Statistics Majors” detailing the success of the capstone course at the undergraduate level. We also have several students taking advantage of internship opportunities. We welcome any leads on internship possibilities for our students.

Pictured above are the students from the Spring 2012 Capstone course taught by Nicole Lazar and Cheolwoo Park, with TA, Adam Jaeger.

Back to Top

Graduate Program

Lynne Seymour, Associate Professor & Graduate Coordinator

After incorporating some hefty changes in the structure of our Qualifying and Comprehensive exams, and upon the introduction of our new sequence of professional development courses (STAT 8910-20-30), our current students are getting into their research earlier than previous students have – M.S. students who opt to write a thesis are starting at least one semester earlier, and Ph.D. students are starting a year earlier.

Several of our alumni this year have passed along job openings for statisticians within the companies they work for. This has been an excellent source of opportunities for our graduating students, as well as a good opportunity for the department to explore in more detail what companies are looking for when they hire new graduates. If you happen to have an opening that one of our M.S. or Ph.D. graduates might be interested in, please let me know! And if you have any suggestions for how we might better prepare our students for the workplace, I will welcome those enthusiastically (706-542-3307/seymour@stat.uga.edu)!

Back to Top

Student Awards

The Department’s annual student awards went to:

Best Senior Ph.D. Student
Cong Feng

Best Beginning Ph.D. Student
Yuhang Xu

Best M.S. Student
Jacob Martin

The Department also wishes to congratulate the following award winners and honorees:

Outstanding Teaching Award
Andrew Anderson
Haileab Hilafu
Minsoo Kim

Phi Beta Kappa inductee
Julia Orr

Innovative and Interdisciplinary Research Grant Award
Jung Ae Lee

Back to Top

Exciting Developments in our Course Offerings

Daniel Hall, Associate Head and Associate Professor

In the Department of Statistics we are always asking how we can improve statistics instruction at UGA and in particular, our undergraduate and graduate courses for statistics majors, Master’s and Ph.D. students. Lately we have taken several steps in this direction that we are excited about. In collaboration with the Department of Epidemiology and Biostatistics, we are introducing two new courses: BIOS (STAT) 8140, Multilevel and Hierarchical Models, and BIOS (STAT) 8050, Intermediate Mathematical Statistics. Both of these will be cross-listed courses taught by faculty from both Statistics and Biostatistics, and are intended primarily as advanced service courses for quantitatively oriented students in other disciplines. The former course will give a broad introduction to multilevel models (a.k.a. hierarchical models and mixed-effect models), focusing on their conception, interpretation, and application in the social and biological sciences, where they are rapidly gaining in popularity and importance. The latter course, BIOS (STAT) 8050, will provide a compact (i.e., one semester) introduction to mathematical statistics focusing on those aspects of the theory of estimation and inference that are most crucial to students in disciplines such as epidemiology, educational research, and economics. We hope to offer both of these courses for the first time in academic year 2012-13.

In addition to these new service courses, the Department of Statistics regularly offers classes on new topics of interest to our own graduate students through STAT 8900, Special Topics in Statistics. Over the last several years this course has covered such cutting edge topics as Symbolic Data Analysis (taught by Dr. Billard), Statistical Learning and Data Mining (co-taught by Drs. Ahn and Park), Sufficient Dimension Reduction (taught by Dr. Yin), and Statistical Analysis of Neuroimaging Data (taught by Dr. Lazar). These classes cover some of the most interesting and novel areas in statistics and provide our students with a sense of the dynamic nature of our field as well as exciting topics for research and further study.

At a less advanced level, another recent change to our courses that we think will benefit undergraduate majors and Master’s level students involves the split-level (that is, undergraduate/graduate) course, STAT 4/6360. For many years this course provided extensive training in how to use SAS, a dominant and powerful statistical package. Although SAS remains a very important piece of software, no one statistical package can provide the entire toolbox for the modern statistician. Therefore we have recently broadened the scope of STAT 4/6360, which is now entitled, “Statistical Software Programming”. This course now covers statistical software in general, with particular emphasis on SAS as well as R, which has enjoyed an explosion of popularity in recent years and has become nearly essential to the practice of statistics.

We think the developments in our course offerings that have occurred recently are great for our students and we are very pleased about them. We will continue to work hard to improve statistics instruction at UGA. For instance, major improvements to our graduate level service courses are currently in the works, although details will have to wait until next year’s newsletter. In the meantime, if you have any thoughts about how we can improve the training we provide our students, be sure to let us know.

Dr. Lily Wang helps students in the Statistical Software Programming class (4/6360) located in the Statistics Computer Lab.

Back to Top

Alumni Profile

Rebecca (Barrett) Phillips

The Department’s featured alumna is Rebecca (Barrett) Phillips. She earned both a Bachelor’s (2005) and Master’s (2007) from the UGA Statistics Department. She began her career with Synovus Financial Corporation in 2007 as a research analyst in the Card Services Division. Some examples of projects she worked on include a logistic regression model to identify which customers should receive a promotion mailing, time series analysis for budgeting purposes, and scorecard validation to ensure cutoff scores are properly set when deciding what terms to offer to credit card customers.

After spending several years in the banking industry, she earned an M.B.A. from Kennesaw State, graduating in May 2011. In late 2011, Synovus promoted Rebecca to the position of model validation analyst on the model risk management team. This team is a part of the Audit Division, which works closely with other areas of the bank to validate their models. The goal is to ensure the models both effectively model the financial information needed to make business decisions, and also meet requirements when audited by federal regulators. “I thoroughly enjoy taking the knowledge I gained in the UGA Stat Department and applying it to a real-world situation. I keep all my books and notebooks in my desk and reference them often. The financial industry has so many opportunities to apply statistical models to enhance the company’s business plan.”

On a personal note, Rebecca and her husband Matt were married in January of 2009. They purchased their first home in the summer of 2011, and home renovation projects consume most of their spare time. They are UGA football season ticket holders, so they are sure to find their way back to Athens at least 6 or 7 times a year!

Back to Top

Statistics for Teachers Class

Christine Franklin, Senior Lecturer and Undergraduate Coordinator
Jennifer Kaplan, Assistant Professor

The UGA Statistics Department has been teaching a second course in statistics for mathematics education students since 1998. The course is designed to prepare pre-service and in-service high school mathematics teachers to teach the statistical concepts that appear in the Common Core Mathematics Standards and the Advanced Placement (AP) Statistics curriculum.

Two recent trends have converged to render knowledge of data analysis, statistics, and probability more essential than ever for teachers at the elementary, middle, and secondary levels. First, state standards, such as the Georgia Performance Standards (GPS), and national standard recommendations, such as the Common Core State Standards (CCSS), now include an unprecedented amount of content in data analysis and probability in the curriculum, from grades K-12. Second, the professional practice of teaching has become more data-driven, with standardized tests, both state and national, becoming a much larger part of the professional life of the teacher. STAT 4/6070 (both for senior undergraduates and graduate students) is designed to prepare high school teachers to meet the demands associated with these two trends.

It is important to recognize that the nature of teaching and learning in data, statistics, and probability is quite different from that of subjects like algebra and geometry. Mathematics is a foundation, but not a proxy, for statistics and probability. Statistical reasoning differs from mathematical reasoning in that it begins not with premises or definitions, but from evidence or data. Underlying statistical reasoning is a conceptual understanding of important ideas, such as distribution, center, spread, association, uncertainty, randomness, and sampling. One of the goals of STAT 4/6070 is to equip high school mathematics teachers to teach probability and statistics with an understanding of the difference between mathematical and statistical reasoning.

Pedagogically, the course has been taught using an activity-based model in which the students complete activities to master certain learning goals. For example, in one activity the pre-service teachers explore the learning progressions associated with measures of center and variability. The pre-service teachers use stacks of snap cubes to visualize the mean as the fair share value and then to quantify the variability as a measure of fairness. The next level of development for children is to consider the mean as a balance point. The pre-service teachers use this model to create dotplots of data sets that meet certain criteria and then quantify the variability in the distributions they create. This leads to a discussion of the Mean Absolute Deviation (MAD) which is extended back to the measures of fairness and forward to the concept of standard deviation. Not only does participation in this activity deepen the pre-service teachers’ understanding of center and variability, it also provides them with activities and a pedagogical model they can incorporate in their own classes.

The assessments used in the course are designed to help pre-service teachers connect their statistical learning to teaching practice and include opportunities for the pre-service teachers to read the research literature on the teaching and learning of statistics. The pre-service teachers complete a portfolio that includes summaries of journal articles chosen to reflect best practices in teaching statistics as specified by the Guidelines for Assessment and Instruction in Statistics Education (GAISE). In addition, the pre-service teachers compare and contrast the student learning goals for statistics and probability specified by state (GPS) and national standards (CCSS and GAISE). Finally, the pre-service teachers find tools appropriate for use in their classroom, such as data sets, web sites and lesson plans and explain how they might implement their finds in their future teaching.

This course has been nationally recognized as a model statistics course for teacher preparation programs. The course is unique by integrating both statistical content and pedagogy. This course is also the foundational course for the developing Masters of Science in Statistics Education degree our department will soon offer. This degree program has also received national recognition (for example, the American Statistical Association) for the vision our department has shown in becoming a leader in the growing field of statistics education.

Students sampling from a population to learn about confidence intervals.

Students using online statistics applets to develop conceptual understanding of the material.

Back to Top

Statistics News Spring 2012, Volume 5

Contributors: Timothy Cheek, Julie Davis, Christine Franklin, Daniel Hall, Jennifer Kaplan, Yehua Li, Kim Love-Myers, Chris O’Neal, Cheolwoo Park, Lynne Seymour, John Stufken, Lily Wang