Customer Story

Statistical thinking for a new generation

By integrating a free online learning resource into course curriculum, a faculty member at UNH offers students the opportunity to learn industry-ready statistical skills at their own pace

The University of New Hampshire

ChallengeToday, some of the best jobs for recent university graduates in science and engineering are in industry. Traditional coursework in most science and engineering programs, however, hasn’t kept up with the job market’s growing demand for statistically savvy graduates. 
SolutionTo provide science and engineering students with an opportunity to learn the fundamentals of statistics, some faculty at the University of New Hampshire are integrating a free online statistics course, Statistical Thinking for Industrial Problem Solving (STIPS), into course syllabi. STIPS not only offers students an interactive, self-paced study format, it familiarizes them with JMP®, an industry-standard software.
ResultsUNH Principal Lecturer Phil Ramsey says STIPS fills a critical gap in science and engineering instruction. And STIPS is popular with students: “It’s multimedia, it’s lively, things are well explained, and students can complete it at their own pace. Just last week, one student told me: ‘I think I finally understand statistics.’” 

University of New Hampshire Principal Lecturer Phil Ramsey, PhD, has had an interesting career. “I’m very much an interdisciplinarian,” he explains. “I love collaborations.” Even just 10 years ago, that might have put Ramsey in the minority among academics. But times have changed.

Today, some of the best jobs for recent college graduates in science and engineering are in industry. And with the rise of globalization and a growing emphasis on systems integration, industry leaders are instituting new operational structures that promote interdisciplinary approaches where statistics is the common language. It’s no surprise that statistical skills are therefore in high demand.

Ramsey, himself something of an early adopter of industrial statistics, cut his teeth at aerospace manufacturer McDonnell Douglas before obtaining a PhD in statistics from Virginia Tech. Later, he went on to roles at Alcoa, then Rohm and Haas.

Maybe it’s this career trajectory that makes Ramsey so relatable to students, many of whom will be applying for jobs in industry once they graduate. Or maybe it’s the nature of the courses he now teaches as part of UNH’s College of Engineering and Physical Sciences, which include Design of Experiments, Data Mining and Predictive Analytics, and Statistical Methods for Quality and Reliability, a semester-long course in which students become Six Sigma Green Belt certified.

Ramsey’s courses have become a go-to for engineering and science majors, and in more recent years have also attracted students from other disciplines. The common thread, he says, is the courses’ focus not just on teaching concept and theory, but on skills and applications – things students value most when thinking about their future careers.

“These students have all heard how it really helps to have a design of experiments course on your resume when you’re applying for jobs,” Ramsey explains. “And they’re coming to realize – having heard from the students who’ve graduated before them – how important it is to know statistics.” 

Serving up the subject matter with a side of statistics

In fact, Ramsey says, that very synergy between subject matter and statistical skills is what makes his teaching so relevant. Courses like Design of Experiments lean heavily on case studies from industry, framing key concepts in the context of real business challenges that must be solved with statistical methods.

Just as engineers and scientists in industry use a statistical tool to solve these challenges, so too do Ramsey’s students. In fact, they use the very same software: JMP®.

“For design of experiments, JMP is probably the cutting-edge software,” Ramsey explains. “Plus it’s easy. It’s fully integrated with the analysis.” With its dynamic interface and user-friendly visualization features, JMP encourages trial and error and gives students the opportunity to explore the sample data sets Ramsey shares in class.

“I give very few assignments that have correct answers,” Ramsey explains, noting that students are presented with a business challenge and the data set to solve it. They then use JMP to run an experiment, analyze it and report back on what they’ve found. Grades are based on students’ ability to design the experiment correctly and provide a compelling rationale for their conclusions; it’s the critical thinking that matters.

“I’ve had a lot of students come to me and say, ‘I’ve had courses in machine learning before, but this is the first one where we actually attacked the problem and analyzed data to solve it.’”

Getting students – few of whom have had any formal training in statistics – to this level of understanding, however, requires that instructors provide them with the basics. And filling those knowledge gaps is a priority, Ramsey says. That’s why he’s using Statistical Thinking for Industrial Problem Solving (STIPS), a free online course from JMP, as a teaching supplement. 

“STIPS is expanding how much students learn and how they learn it,” he explains. “It’s filled with realistic industry case studies. And it’s self-contained, self-paced.”

A self-paced crash course, direct from industry experts 

Experts at JMP created STIPS to help anyone interested build practical skills in using data to more efficiently solve problems. It’s an ideal complement to in-class learning, Ramsey says; all students need is an internet browser.

Students in Ramsey’s courses can access STIPS links through UNH’s learning management portal, Canvas (known to students as myCourses), where he has assigned STIPS content in sections spread out across the semester.

STIPS is comprised of seven modules, totaling 25-30 hours of self-paced learning through instructional videos, JMP demonstrations, questions and practice exercises. Modules can be taken in order, or students can pick and choose the content that’s most relevant for them. This format enables instructors to present the material at a slower pace and in the order that makes the most sense for students.

“The structure of STIPS is flexible, so it’s possible to order much of the content to suit the course one is teaching, which may not align perfectly with any one STIPS module,” Ramsey explains.

“It’s all fully integrated – students can just move through the course … STIPS actually streamlines this and makes it better.” Ramsey says that before STIPS, he had to spend a significant amount of time culling materials from external sources, navigating copyright issues and creating his own content. “But my videos were not professional quality!” he laughs. “Just think about the resources and expertise that went into creating STIPS. That’s just not something that I [personally] would have ever been able to make available to students on my own.”

Ramsey tracks students’ participation by asking them to upload STIPS badges as proof that they’ve completed a module. And for more advanced courses like Statistical Methods for Quality and Reliability, students can even get STIPS certification by the end of the semester.

“I’m getting feedback from my students that they really like STIPS. It’s multimedia. It’s lively, things are well explained and they can complete it at their own pace. Just last week, one student told me ‘I think I finally understand statistics.’ She said she really liked design of experiments and was excited about the fact that she’s getting a better understanding of statistics along the way.” 

Try STIPS

Learn statistics from home. Build your skills with this free online course.

A sea change in how students think about the role of statistics in science and engineering careers

The popularity of interdisciplinary courses like Design of Experiments points at a broader shift in thinking about how students should prepare for their futures. Siloed learning, Ramsey says, is falling by the wayside in favor of systems-level approaches. Familiarity with statistics is critical, and familiarity with JMP helps graduates hit the ground running in new industry roles.

“Students leaving here have a pretty good understanding of JMP. They understand how it works and know how to design modern types of experiments,” Ramsey explains. It’s an experience students say provides them with the skills to perform well in new roles once they graduate.

And they’re not alone. Companies are also looking to cultivate statistical skills among their employees. Many have deployed strategies to encourage their own scientists and engineers to spend time completing trainings in STIPS and learning how JMP can streamline the work they’re doing. Many have become STIPS certified. And companies are now finding ways to formalize employees’ incentive to do so by making STIPS certification part of career development programs and even the performance review process.

Ramsey shares a few examples of industry clients he’s worked with in recent years. “Last week at a government workshop, I’d been asked to present to at least 35 engineers, not one of them who had ever taken a course in statistics,” he says. “I showed them STIPS and they were all really intrigued. STIPS is exactly what they need. It’s exactly the level that people need – and that is, of course, by design.

“People love STIPS because it teaches them what they want – and need – to learn.”

The results illustrated in this article are specific to the particular situations, business models, data input and computing environments described herein. Each SAS customer’s experience is unique, based on business and technical variables, and all statements must be considered nontypical. Actual savings, results and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software.