Paper ID #26579

Use of Minitab Statistical Analysis Software in Engineering Technology

Dr. Ali Ahmad, Louisiana Community and Technical College System-MEPOL
Dr. Ali Ahmad is Director of Manufacturing Extension Partnership (MEP) of Louisiana, which operates
under the Louisiana Community and Technical College System. Dr. Ahmad is a professional with over 18
years of experience in industrial engineering, research and management fields. He was previously an As-
sociate Professor and Head of the Engineering Technology Department at Northwestern State University
of Louisiana. He obtained his Ph.D. in Industrial Engineering from the University of Central Florida. Dr.
Ahmad has diverse expertise in human-computer interaction, quality engineering, and simulating man-
ufacturing systems. Ali worked on projects related to transfer of training, user-centered design, process
improvement, and virtual environments. Dr. Ahmad is a Certified Simulation Analyst and a Certified Six
Sigma Black Belt.

American
c Society for Engineering Education, 2019
 Use of Minitab Statistical Analysis Software in Engineering Technology
Abstract

The Engineering Technology curriculum provides wide spread knowledge in problem solving,
management of resources, and process planning. Statistical decision-making is a key skill
required by Engineering Technologists, and is required under ETAC of ABET program criteria
for Industrial Engineering Technology and similarly named programs.

Traditional approaches to teaching statistical analysis tend to be mathematical in nature.

Anecdotally, more than 75% of students dread their first Statistics class. This paper discusses the
incorporation of Minitab Statistical Analysis Software in engineering technology courses. It
looks at applying statistical decision-making without delving deep in statistical theory. It builds
on industry specific approaches to empower non-statisticians to apply statistical tools in
everyday decision-making. This is enabled using menu-driven statistical analysis software with
powerful computational algorithms and graphics. Statistical analysis tools (such as descriptive
statistics, confidence intervals, hypothesis testing, regression analysis, and ANOVA) can be
applied using software, then, the decision-maker is able to use simple rules to interpret the
software results. Moreover, the decision-maker can also test the assumptions of applying
statistical tools, a process that is hard to teach in traditional Statistics courses. The paper
concludes by providing directions for including real-life case studies to illustrate statistical
decision-making to students.

Introduction and Background

The National Academy of Engineers forecasts that engineers and technologists will continue to
operate in a rapidly changing innovation environment1. This is compounded by globalization of
economies, diversity of social and business groups, multidisciplinary research trends, and
cultural and political forces. Engineering systems are of increasing complexity in energy,
environment, food, product development, and communications1. Hence, it is imperative to
introduce engineering and technology practices in undergraduate education, where students can
experience the iterative process of designing, analyzing, building and testing. There is a growing
importance for engineering practice, but the engineering profession seems to be held in low
regard compared to other professions and industry tends to view engineers and technologists as
disposable commodities2.

Engineering Technology prepares graduates with knowledge skills and technical problem-
solving abilities necessary to success in a wide range of engineering technology disciplines3.
The specific ABET ETAC student outcomes for Engineering Technology are3:
(1) an ability to apply knowledge, techniques, skills and modern tools of mathematics,
science, engineering, and technology to solve broadly-defined engineering problems
appropriate to the discipline;
(2) an ability to design systems, components, or processes meeting specified needs for
broadly-defined engineering problems appropriate to the discipline;
 (3) an ability to apply written, oral, and graphical communication in broadly-defined
technical and non-technical environments; and an ability to identify and use appropriate
technical literature;
(4) an ability to conduct standard tests, measurements, and experiments and to analyze
and interpret the results to improve processes; and
(5) an ability to function effectively as a member as well as a leader on technical teams.

The program criteria for Industrial Engineering Technology and similarly stated programs state
that4 “Graduates at the baccalaureate level must demonstrate the ability to apply knowledge of
probability, statistics, engineering economic analysis and cost control, and other technical
sciences and specialties necessary in the field of industrial engineering technology.”

The field of manufacturing is wide, and engineering technologists must understand the processes
and materials involved in the creation of a useful product4. The emergence of non-traditional
education providers (such as online and hybrid) poses challenges for US higher education
institutions. To remain competitive, US universities and colleges should re-adapt the way
education is delivered, and develop curricula that meets the core competencies required in the
market place5. At a time when local, state, and national resources for education are becoming
increasingly scarce, expectations for institutional accountability and student performance are
becoming more demanding. There is a need for more educational innovations that have a
significant impact on student learning and performance6.

This research takes a pragmatic approach to illustrating statistical decision-making using menu-
driven software. The paper proceeds by discussing the method used to carry out the research.
After that it provides a summary of the results. The paper concludes by a discussion of the key
findings and provide directions for future development.

Method

This paper uses a case-study approach. Minitab© Statistical Analysis Software was used to
develop the illustrations used in this paper7. Despite the existence of several types of statistical
analysis software, Minitab was selected because it is commonly used in industry, manufacturing
and healthcare, is tailored for quality control applications, provides tools to check the
assumptions associated with statistical tools, and is equipped with an Assistant tool that can aid
students and practitioners in systematically selecting the proper technique to analyze data and
interpret the results. It is important to note that the use of a software does not eliminate the need
of studying statistical analysis. Additionally, it is expected that the students have covered the
following topics: probability, random variables, probability distributions, measures of central
tendency and variation, confidence intervals, hypothesis testing, ANOVA, and regression
analysis prior to using Minitab (or similar software) to automate the required calculations. The
paper is presented as a set of scenarios with associated results in a way that emphasizes statistical
decision-making for non-technical individuals. Figure 1 provides a layout of the Minitab©
Statistical Analysis Software environment. It provides a spreadsheet view where data are input
(like Excel), variables are entered as columns while values are entered as rows, and a session
window where analysis results are displayed. Various commands can be launched from the top
menu-bar. The datasets used in this paper can be downloaded from:
https://support.minitab.com/en-us/datasets/. This paper is not intended to replace formal training
on the software or formal instruction on statistics.

Figure 1. Minitab© Statistical Analysis Software

Results and Discussion

One of the challenges typically encountered by Engineering Technology students is learning

which statistical tool to apply to a given data set. Figure 2 provides a roadmap that aids students
in making that determination. It is based on Minitab’s Assistant tool, which can be used to
provide guidance on selecting a proper hypothesis testing technique.

Figure 2. Minitab Assistant- Hypothesis Testing

As an example of how Minitab can be used to help understand statistical decision-making,
consider the following scenario, a medical administrator collected satisfaction ratings for two
hospitals, as shown in Figure 3, which includes a snippet of the data and a side-by-side boxplot.

Figure 3. Hospital Data

In this example, the hospital administrator could be interested in determining if the average
ratings of the hospitals are different. A two-sample t-test can be used to inform that conclusion,
as depicted in Figure 4.

Figure 4. 2-Sample T-Test for Score

Then, students then compare the T-Value (i.e., test statistics) to a reference value based on a
chosen confidence level (i.e., α). Alternatively, the conclusions can be made based on the
calculated p-value, which is 0.000 in this case, and if this p-value is less than the selected
confidence level, say 0.05, then this results in rejecting the null hypothesis. Non-technical
practitioners commonly use the following statement: If the p is low, the null must go. In this
example, the conclusion from the test is that the null hypothesis can be rejected, and that the
means are not equal. One other challenge is verifying the assumptions associated with the using
the t-distribution to make conclusions, that is the underlying distribution of the data is normal,
which can also be tested via the software, see Figure 5. From the figure, the Anderson-Darling
(AD) p-value associated the goodness of fit test for a normal distribution is 0.719. The AD test
has the following hypothesis:
• Null Hypothesis: H0: Data is coming from a normal distribution
• Alternative Hypothesis: H1: Data is not coming from a normal distribution
Since the p-value is high, i.e., greater than 0.05. there is not enough evidence to reject the null
hypothesis. Non-technical practitioners sometimes apply a pencil test on the plot in Figure 5, i.e.,
if you can cover all the points with a pencil, then, you can conclude that the underlying
distribution is normal.

Figure 5. Normality Test of Breakfast Data

In a second example, a manufacturer collected recycled metallic cans, see Figure 6, which
includes a snippet of the data and a side-by-side boxplot of scores.
 Figure 6. Can Weight Data

In this example, the manufacturer could be interested in determining if the average weights of the
15 groups of cans is equal or not. An ANOVA test can be used to inform that conclusion, as
depicted in Figure 7. The results show a low p-value, which can be used to conclude that at least
one can group has an average weight that is different.

Figure 7. ANOVA for Can Weight

Moreover, an engineering technology student can proceed with testing the various ANOVA
assumptions, such as the residuals are normally distributed, with a mean of zero and constant
variance, and that there are no trends or patterns related to data order. This can be done using the
four-in-one plot produced by Minitab, as shown in Figure 8.
 Figure 8. Testing ANOVA Assumptions using Four-In-One Plot

In a third example, a researcher is interested in evaluating if there is a difference in preference

for chocolate type (e.g., Dark, Milk, or White) between males and females, as shown in Figure 9,
which includes a snippet of the data along with a cross-tabulation.

Figure 9. Chocolate Preference Data

In this example, students can use a Chi-Square test of independence to test the following
hypothesis:
• Null Hypothesis: H0: The variables are Independent
• Alternative: H1: The variables are dependent
Figure 10 shows the results of the Chi-Square test. The p-value is less than 0.05, which results in
rejecting the null hypothesis; thus, concluding that there is a difference in preference for
chocolate type between males and females.

Figure 10. Chi-Square Test

Taken together, the three examples above illustrate that Minitab software can be incorporated in
Engineering Technology courses to (among other benefits):
• Automate calculation of statistics and p-values
• Generate charts that aid in data understanding
• Verify assumptions underlying statistical techniques
• Provide guidance on selection of proper statistical technique
• Facilitate statistical decision-making using simple rules of thumb

The incorporation of a software in statistics course curricula allows for using real-life case
studies, such as those included in Minitab training datasets available at:
https://support.minitab.com/en-us/datasets/. Dealing with real-life case studies allows for
incorporating contextual information on data analysis and dealing with larger datasets.
Bibliography
1. Phase, I. I. (2005). Educating the Engineer of 2020: Adapting Engineering Education to the New Century.
National Academies Press.

2. Duderstadt, J. (2008). Engineering for a Changing World: A Roadmap to the Future of Engineering
Practice, Research, and Education. The Millennium Project, University of Michigan, Ann Arbor,
MI, http://milproj.dc.umich.edu.

3. Accreditation Board for Engineering and Technology (ABET; 2019)- Criteria for Accrediting Engineering
Technology Programs, 2019 – 2020 [https://www.abet.org/accreditation/accreditation-criteria/criteria-for-
accrediting-engineering-technology-programs-2019-2020/]

4. Rhoades, L.J. (2005). The Transformation of Manufacturing in the 21st Century. Bridge, 35 (1), pp. 13-20.

5. Christensen, C. M., & Eyring, H. J. (2011). The innovative university: Changing the DNA of higher education
from the inside out. John Wiley & Sons.

6. Jamieson, L. H., & Lohmann, J. R. (2012). Innovation with impact: Creating a culture for scholarly and
systematic innovation in engineering education. American Society for Engineering Education, Washington.
[Available online: http://www.asee.org/member-resources/reports/Innovation-with-Impact].

7. Minitab Software. http://www.minitab.com/en-us/.