The ASA and Big Data: What’s Been Happening Lately?

Nathaniel Schenker

During the past year, the ASA has engaged in a number of activities to strengthen the role of our profession in the areas of Big Data and data science. This column summarizes some of those activities, with an emphasis on important steps we have taken in education and professional development.

Three teams, each composed of about a dozen ASA members and facilitated by ASA Science Policy Director Steve Pierson, have created a series of white papers. These papers are aimed at research funding agencies, making clear that statisticians are vital partners in advancing science with their expertise in study design, inference, and quantifying uncertainty. The papers articulate the distinctive input statisticians can provide to help tackle our nation’s critical research priorities. One of the papers discusses how statistics can contribute to solving problems involving Big Data in applications to business, government, health, science, and societal issues. This paper emphasizes that work on such problems is most effective if it is multidisciplinary, involving statisticians, computer scientists, mathematicians, and relevant domain scientists. The authors conclude the following:

“The statisticians engaged in interdisciplinary research involving Big Data will need to be computationally savvy, possessing expertise in statistical principles and an understanding of algorithmic complexity, computational cost, basic computer architecture, and the basics of both software engineering principles and handling/management of large-scale data. Just as critical—for both the training and recruitment—as the computational skills are interpersonal skills for the next generation of statisticians to be effective communicators, leaders, and team members.”

To broaden our thinking, the ASA is conducting meetings around the country with leaders in data science and users of Big Data. It is extremely important to listen to voices outside of our profession, so we have intentionally sought out people in different business sectors. In Cincinnati, Ohio, we met with leaders in large companies who hire master’s-level statisticians for business analytics. In Palo Alto, California, we heard from leaders in high-tech companies who hire doctoral-level statisticians for specialized problem solving. In New York City, ASA Executive Director Ron Wasserstein met with several dozen people, mostly young people, working in Big Data enterprises.

The voices we heard complement the messages of the ASA white paper with a business perspective on the skills needed by statisticians. Participants in the meetings noted that businesses need people who can “jump in and swim with data” at a time when both the volumes and complexity of their data are increasing rapidly. Statisticians need the skills to understand business problems and find solutions using data, but—even more importantly—the leadership ability to “make it to the middle,” serving as a bridge between highly technical employees and business management–oriented employees. In summary, generic ativan pill desirable traits include strong skills in statistical methods, data collection and management, and computing; understanding important applied problems and how to use data to solve them; communication; and leadership.

What has the ASA been doing to help statisticians enhance their skills in these important areas?

We introduced short courses in the analysis of unstructured textual data at the 2014 Conference on Statistical Practice (CSP) and Joint Statistical Meetings (JSM), in addition to the many other member-initiated short courses and sessions in Big Data, data mining, etc. We plan to offer more of these kinds of courses in the future. Such activities will help attendees strengthen their ability to “swim with” complex data.

We launched a personal skills development program. This program was designed to enhance the ASA’s offerings in the area of professional development and will include courses, workshops, and other training in effective communication, collaboration, leadership, and career planning. In particular, we developed training in statistical leadership and presented a JSM course titled Preparing Statisticians for Leadership: How to See the Big Picture and Have More Influence. Based on the success of this course at JSM 2014, future leadership training will be pursued for JSM 2015 and CSP 2016. We also revised our course on giving effective presentations and rolled it out at the 2014 Fall Technical Conference. The offerings in personal skills development will help attendees improve their problem solving and communications skills and to “get to the middle,” thus enhancing their influence.

We are updating the ASA’s Guidelines for Undergraduate Programs in Statistical Science, the current version of which was written in 2000. The draft updated guidelines, available here, list guiding principles. In particular, the first four, “the scientific method and its relation to the statistical problem-solving cycle,” “real applications,” “focus on problem solving,” and “the importance of data science,” are strongly related to skills emphasized in our meetings with business leaders and in the white paper cited above.

The ASA has taken on the responsibility of coordinating DataFest activities around the United States at the request of the organizers. DataFest is a data analysis competition for teams of undergraduates, who devote a weekend to analyzing a large and complex data set, looking for insight and learning to communicate what they learn. This exciting program encourages collaboration among students from different disciplines and helps students develop skills they’ll need in graduate school and in the work force. For student perspectives on DataFest, see the September 2014 President’s Corner.

Our profession has a great deal to offer to solving problems involving Big Data, and the ASA will continue its efforts to attract data scientists to the Big Tent for Statistics and prepare statistical scientists for success in these areas.