Our culture and economy float on a sea of data. As we go about our daily lives -- paying bills, buying groceries, sending emails, reading tweets, posting to Facebook, (periodically) undergoing medical tests -- we generate streams of data. Businesses, health care organizations and governments need people who can make sense of these streams -- these torrents -- of data. And we, as citizens, need to understand how to benefit from our personal data. Statistics, the science of learning from data and making sense of variability, is well equipped to help us understand the impact of data on our lives and to prepare a future work force that can make meaningful discoveries from this data deluge. But achieving these goals will require enormous efforts in teacher preparation and changes in the statistics curriculum, itself.
Until fairly recently, statistics arguably has been one of the most-feared classes in the college curriculum. But today, because students recognize the potential value of data analysis and due to an educational reform movement started in the 1980s, statistics has become one of the most important and popular college courses. This year, more than 160,000 high-school students will take Advanced Placement (AP) Statistics. Statistics is quickly replacing calculus as the gatekeeper to higher-level study at college and is one of the fastest-growing majors among the mathematical sciences. Also, the national Common Core State Standards (CCSS) in mathematics curriculum will bring statistics to almost all levels of the kindergarten-through-high-school (K-12) curriculum.
The infusion of statistics at the K-12 and college undergraduate levels is a consequence of the reform movement in college statistics education begun in the late 1980s. This movement shifted the focus of statistics education from plug-and-chug calculations to the development of critical-thinking skills necessary for analyzing data and communicating conclusions. The new curriculum is reflected in the National Council of Teachers of Mathematics (NCTM) standards, the AP Statistics program -- which administered its first exam in 1997 -- and, most explicitly, in the American Statistical Association's (ASA) Pre-K-12 and College Guidelines for Assessment and Instruction in Statistics Education (GAISE) report released in 2007.
The data-centered reform was possible because of radical innovations in technology that put powerful analysis software into the hands of students and gave teachers new and effective ways of teaching. R, the lingua franca of statisticians, is available for free; its only cost is a steep learning curve that requires thoughtful classroom implementation. Other software programs, such as Fathom and StatCrunch, are inexpensive and allow students to delve into the mysteries of data within minutes. Technology provides powerful teaching tools in the form of applets and simulation environments that allow students to understand the implications of complex statistical theorems without years of mathematical study. It also creates an accessible environment for exploration and visualization of data, allowing instructional time to be spent understanding concepts and solving real problems.
Successfully implementing the teaching of statistics so students develop a sound foundation toward statistical reasoning and thinking will require enormous efforts in professional development. This is especially true at the K-12 level within the CCSS math curriculum, where the prevalence of technology means many teachers must learn skills and practices that did not exist when they were students.
The ASA and NCTM recently addressed the need for improved professional development in statistics in a joint position statement, Preparing Pre-K-12 Teachers of Statistics (April 2013).
This position paper is a must-read for those concerned with statistics education, as it provides five necessary recommendations for preparing and supporting statistics teachers in the pre-K-12 curriculum. These five recommendations can be summarized as promoting the establishment of well-designed professional development courses and resources and urging cooperation between local school districts, state departments of education, national organizations (such as the ASA and NCTM) and teacher preparation programs. In response to these recommendations, the ASA is supporting a team of statistics and mathematics educators, chaired by one of us (Franklin) and Tim Jacobbe of the University of Florida, to write a document that will provide guidance for implementing each.
Statistics educators also must acknowledge and adapt to the reality that our traditional concept of data has changed. Data are no longer simply "numbers in context" (as statistician David Moore famously defined them), stored in static spreadsheets and collected to answer specific research-driven questions. Today, data are dynamic, complex, highly structured collections of pictures, sounds, relationships and -- yes -- numbers. Data sets are vast and, thanks to open-data initiatives such as data.gov, readily available. A new curriculum, sometimes referred to as data science, can prepare students to work with these rich data sources.
One of us (Gould) is working with a team of researchers and high-school educators at the University of California at Los Angeles to bring data science into the high-school curriculum. This National Science Foundation-funded project (mobilizingcs.org) enables students to use their own phones and other devices to collect data about their lives and communities and teaches the computational and statistical practices necessary for them to capture data, discover trends and share findings. When fully implemented, Los Angeles Unified School District students will partake in a curriculum that currently exists in only a handful of elite colleges and universities. These students will be better prepared than their peers for working in the modern economy. Professional development is an important challenge in implementing this curriculum; there are not any in-service or professional-development programs that provide teachers the background needed.
Now is an exciting time for our fellow statistics educators. For the first time, perhaps in history, we find our subject to be in demand across all disciplines and at all age levels. The price of this success is that we must work hard to develop a teaching work force capable of preparing future generations for the challenges of modern data. Teachers are the key to success, since only they can show students that statistics is necessary for participating in our data-centric world.
Franklin is the Lothar Tresp Honoratus Honors Professor in Statistics at the University of Georgia.
Gould is director of the Center for Teaching Statistics in the UCLA Department of Statistics.