Descriptive, Predictive, and Prescriptive Analytics
Analytics is the science of discovering and communicating meaningful patterns in data and developing actionable plans. Generally speaking, Analytics can be divided into three broad categories: Descriptive, Predictive, and Prescriptive. Descriptive Analytics is all about providing insights into what has already happened. Predictive Analytics focuses on generating forecasts about future performances and outcomes. Prescriptive Analytics builds on both Descriptive and Predictive Analytics to help identify solutions to specific problems and decision-making applications. In addition, the emerging field of Visual Analytics can now process massive, heterogeneous, and dynamic volumes of information into understandable formats. Visual analysis software allows the user to not only represent data graphically, but also to interact with a variety of visual representations. Dashboards, which represent the face of the Analytics paradigm, offer a visualization perspective in reporting results and recommendations.
The application of Analytics to sports, meanwhile, is not limited to the United States, but is seeing widespread use on a global basis (e.g., Chinese table tennis). On the domestic front, some specific Sports Analytics applications that are receiving increased attention involve sports injuries, dynamic ticket pricing, and player acquisition. In 2013, Major League Baseball spent $665 million on the salaries of injured players, while the NBA coughed up another $358 million. In the NFL, starters missed 1,600 games in 2013. To address these challenges, the sports industry is using Analytics to predict player injuries in advance. By utilizing Analytics in this Predictive manner, team trainers can design customized programs to minimize, for example, the impact of prior injuries.
We really think injuries are the largest market inefficiency in pro sports.
—Adam Hewitt (Peak Performance Project)
Sports pundits and the mainstream media are beginning to embrace context-driven performance data that drills deeper than previous, more superficial statistics, such as home runs. However, some former players and current commentators decry the use of Analytics, claiming that it detracts from the humanity of the game. Nevertheless, Analytics in sports is here to stay, but does require the capacity to convey relatively complex mathematical concepts in an easily digested form (i.e., visual analytics).
With the new data that MLB is capturing from high-speed video and Doppler radar, each game is on the cusp of generating one terabyte of data. We’re talking about a 10-million-fold increase in data capture. We’re getting tracking of everything that goes on the field during the entire baseball game.
—Vince Gennaro (SABR)
Management Education Analytics Applications
The application of the sports analytics paradigm to the management education universe will usher in new ways for graduates to better meet the needs of the business community. Management education and the sports world have a number of common traits: 1) Ever-changing populations, namely students graduating and athletes retiring or being traded; 2) Enormous amounts of performance data; and 3) Increasing pressures to improve organizational performance—for example, college tuition inflation tends to run about twice that of the general inflation rate, and is very similar to ticket price inflation in pro sports. Today, many schools of business now offer graduate-level courses and programs in Analytics. These initiatives are driven, in part, by the growing demand for managers and engineers trained in Analytics. The most recent Bureau of Labor Statistics employment forecast revealed continued significant growth in the demand for Analytics-trained graduates over the next ten years. Even the Graziadio School is getting into the act: Current plans call for the deployment of an M.S. in Applied Analytics by 2016. Beyond offering courses and programs in Analytics, there are a variety of applications that will actually help improve the overall performance and efficiency of the institution. Three specific Analytics applications involve: 1) Enhancing student recruiting and retention, 2) Expanding the student learning experience through the delivery of customized content, and 3) Improving administrative operational efficiencies. Along these lines, the Graziadio School is planning to deploy an Analytics-powered, social media-driven app that better aligns student needs and requirements with course availability, whether it is delivered in the traditional manner or online. This app is simply another example of the growing power of Crowdsourcing throughout the field of management education.
Student Recruitment and Retention
Recent evidence suggests that Analytics can assist in student recruitment and retention by leveraging the fact that some service representatives are extremely successful at dealing with certain types of students. Matching a specific student to a specific service counselor can significantly increase both enrollment and retention yield rates. Analytics can be used to look at past relationship data and then develop a rep assignment schedule that maximizes performance outcomes. Two additional benefits of this approach include the expansion of enrollment diversity and improved inquirer-to-enrollment conversion rates.
Custom Study Plans
Analytics-based models can also be used to design custom study plans based on students’ characteristics and interests. More specifically, they can be employed to: 1) Select student study groups with complementary characteristics and reactions to learning strategies; 2) Identify students who, in multiple choice tests, are hint-driven; 3) Locate students who exhibit low motivation and find alternate means of reaching them; 4) Predict probable student outcomes; 5) Provide information that allows students to be more proactive in their learning behavior; and 6) Allow advisors to observe students’ performances in sufficient time to effectively intervene.
Improve Learning Outcomes
Imagine a learning environment where up to one terabyte of performance data is captured in a classroom setting. This process would not only archive faculty presentations, but student interactions and engagements. This data could be used to provide additional student feedback and also to improve learning outcomes for the next class. The proposed approach mimics the process sports teams are using to improve performance in the next game and for the season. In a similar fashion, the techniques used to address sports injuries can be applied to students who are facing learning or motivational challenges. To thrive in the competitive global environment, businesses increasingly need graduates who possess not only excellent academic preparation, effective communication skills, and strong ethical principles, but experience in addressing the challenges of Big Data.
Today, less than 30 percent of higher education administrators view Analytics as a top priority. Clearly, there is a great deal of work to be done to change this landscape. In that regard, institutions of higher learning should consider establishing the position of Chief Analytics Officer (CAO), as is being done throughout the business world. This position can help mobilize the data, people, and systems needed to improve the decision-making process throughout the institution. Analytics can also help address the call for institutional accountability with regard to student performance and graduate placement, given the rising costs of higher education and current economic turbulence. Analytics solutions for the educational community should be both comprehensible and insightful. Analytics applications should be consistent with the institution’s mission statement and current information systems. However, Analytics-based solutions should also go beyond mere standard reporting and descriptive applications. The key to long-term success is to generate actionable plans that better optimize the institution’s resources. For example, Analytics should not be limited to identifying hidden trends regarding prospective student characteristics and inclinations, but also used to develop specific actions to enhance the probability of successful recruitment. Above all else, the actionable plans should be displayed in a simplified manner via dashboards to ensure consistency of use by the administrative team. When applied in this manner, the Analytics paradigm offers the promise of enhancing student learning and employment opportunities as well as improving an institution’s operational efficiencies. A key ingredient in applying Analytics to management education is to view students not as students but as customers. By embracing this philosophy, the management education community can also begin to experience Lighting in a Bottle!
When you innovate, you’ve got to be prepared for everyone telling you you’re nuts!