Learning Statistics: Before and After Minitab

Minitab Blog Editor | 08 October, 2012

Topics: Minitab Statistical Software, Articles

Minitab Statistical Software revolutionized statistics education and helped students see data analysis not as a hurdle, but a relevant and vital tool. Forty years after its initial release, Minitab remains the leading package used to teach and learn statistics, even as its ease of use has brought it from the university into the professional world, too.

Let’s be honest: most of today’s students aren’t exactly in love with statistics. But in the years before computers were widely available, statistics filled many students with outright dread. And why wouldn’t it? For weeks upon weeks, students in introductory statistics courses memorized complex formulas and slogged through tedious calculations by hand. Time permitting, they might be exposed to some theoretical examples of how the calculations they’d been laboring over might be applied in a practical situation. Analyzing even small data sets under these conditions was an enormous task.

By the end of the 1960s, statistics education faced a crisis. Maurice Lee, dean of business at the University of North Carolina, lambasted the typical introductory statistics course offered in those days: lack of practical relevance, he wrote, “…leads inevitably to the conclusion that the student would be well advised to use the hours now tardily given over to the undergraduate statistics course for attendance at the sessions of almost any other subject offered by the university.”

A quote attributed to Cornell statistician Walter Federer put it more directly: “If the statistical profession doesn’t do something about the teaching of statistics, other groups will.”

Introducing Minitab

The situation began to change in 1972, when the introduction of Minitab Statistical Software transformed statistical computing into a viable classroom tool.

Minitab was not the first statistical software. But the statistical packages available for mainframe computers in the mid-1960s weren’t well suited to teaching. A package developed by the National Institute of Standards and Technology, called OMNITAB, was powerful, but not easy to use. With the goal of making statistics easier to learn, Penn State statisticians Tom Ryan, Brian Joiner, and Barbara Ryan created a more accessible version of OMNITAB, which became the first version of Minitab.

Using Minitab in introductory courses helped students see the practical value of statistics. By reducing the amount of heavy computation necessary to make statistical inferences, Minitab gave students more time to think about what their analyses mean. Statistics education has not been the same since.

Its creators laid out four immediate benefits to learning statistics with Minitab:

  • Eliminating computational drudgery helped students grasp the important concepts without getting lost in a mass of details.
  • A large number of real data sets could be studied, enhancing students’ ability to transfer textbook knowledge to practical situations.
  • Plotting the data in a variety of ways became standard operating procedure.
  • Simulation could be used as a learning tool.

Using the computer in an introductory course also made it practical for students to learn more advanced techniques. For example, the amount of hand calculation required had been a major deterrent to teaching nonparametric methods. Minitab eliminated this barrier.

Minitab Makes Its Mark in Education

In 1976, The Minitab Handbook detailed how to use the software in the classroom. In a review, the University of Rochester’s James Inglis outlined the impact it had: “With Minitab making computations less important, emphasis…can be shifted away from computations to the broader questions of data analysis that statisticians and experimenters face: Which model should be used? Are the assumptions appropriate? How does one interpret the plots, numbers, and test results? What is the practical significance of the statistical analysis? It provides a better opportunity than ever before for instructors to introduce to students how statisticians really analyze data, without spending a lot of time teaching computing.”

Statistics educators witnessed big benefits in the classroom. Ruth Hubbard of Australia’s Queensland University of Technology used Minitab in a course designed specifically for students who have weak mathematics backgrounds, but need to learn about statistics because it is such an important component of most disciplines. “The students loved using Minitab to do their exercises,” she reported. “Producing their own results on the computer seemed to give them a sense of power and mastery over the subject. If they made an error it was easy to rectify. They were intrigued by unexpected results in their printouts and were keen to have them explained.”

But the software didn’t just help students who lacked a strong math background; it made key concepts easier to teach to any student. “Numerous concepts are best illustrated using an easy-to-use statistical package such as Minitab,” said Timothy O’Brien of Loyola University. “My students easily produced histograms…illustrating the Central Limit Theorem for a Bernoulli proportion when n = 100 and p = 0.40 and when p = 0.05. Even though the latter situation barely satisfies the usual ‘largeness’ (of sample size) criterion (that is, both n*p and n*(1-p) at least 5), students easily see that the normal approximation is becoming dubious for this case.”

Minitab Today

Minitab Statistical Software remains the world’s leading statistical package for teaching statistics, used at some 4,000 universities worldwide. More people have learned statistics with Minitab than any other statistical package. Of course, easy access to statistical power makes Minitab as useful to professionals as it is to students. Learning with Minitab gives graduates an edge in the job market, because Minitab is also the leading software used in quality improvement initiatives in business and industry.

Many professionals using Minitab today remember all too well what it was like to learn statistics without the software. “I've learned statistics twice,” says Paulo Scheidegger, a senior consultant at INDG, the largest management consulting firm in Brazil. “In the early '70s as a graduate student, my teachers used old-style techniques that stressed much more theory and almost nothing about practical uses and examples. Needless to say, it was a pain to learn and I forgot everything as soon as I graduated. Many years later I had to refresh my concepts in order to help quality control at my former job, and I was introduced to Minitab. Wow! How easy it was to generate vast amounts of data to check my concepts and actually see what changes would happen under this or that assumption. Soon I was able to teach basic statistics to my colleagues from a practical point of view. I am still learning at my sixties, and now I am indeed involved in stats teaching as a consultant, but sometimes I wonder how it would have been if I'd met Minitab back in the ‘70s.”

The difference between teaching and learning statistics before and after Minitab is crystal clear for Alvaro Miró, the CEO of FORMATEX, an Argentinian consulting company. “At first we taught using heavy theory, with theorems, properties, etc., and using incredibly complex formulas, heavy tables and neverending exercises for solving very simple matters like an OC curve or a simple correlation. Pupils really hated it because the real value of statistics is not knowing the formula of sigma or the binomial distribution, but the applications of these formulas to problems. So the subject was very hard to teach, and everything was numbers. Today I teach in industries, and it is such a pleasure to be using Minitab, talking about concepts and being able to demonstrate the real object of statistics.”

If you’re not already using Minitab Statistical Software, download our free 30-day trial!