Tag: applied statistics

How to Become a Medical Statistician

Medical statisticians analyze data to prevent and predict diseases.

Accurate measurements and data analyses are invaluable throughout healthcare systems. That is, sound statistical methods in clinical trials, university research, and hospital operations are crucial to better patient outcomes. Medical statisticians – also known as biostatisticians – are key to setting the foundations of medical research and to clinical decision-making.

Medical statisticians regularly require both applied statistics expertise and healthcare industry knowledge. Merging these knowledge areas is possible with a graduate-level education. If this career intrigues you, read on about the responsibilities in this role.

A Medical Statistician’s Job Responsibilities

Medical statisticians design clinical studies and quality improvement projects. They also perform other data-gathering methods in healthcare settings. That is, they are involved in every step of a project’s development from study design through publication of research results.

Typical employers for medical statisticians include the following:

  • Integrated healthcare systems
  • Pharmaceutical and medical equipment companies
  • Public health departments
  • University research institutes and medical schools

Statisticians in healthcare settings may simultaneously oversee studies and projects in different stages. For instance, take the example of a medical statistician employed by a pharmaceutical company. This statistics professional might work on one of more of the following tasks in a day:

  • Optimizing sample sizes for upcoming clinical trials of new medications and products
  • Updating case report forms with data from ongoing trials
  • Modeling trial results for data on positive and negative impacts
  • Writing the methods section for an application to a regulatory body like the FDA

Because health statisticians are involved with many aspects of healthcare, their careers are highly collaborative. Medical statisticians might team up with physician-scientists, medical writers, clinical managers, and regulatory affairs specialists. They may also work with statisticians at other organizations as part of large-scale research projects.

Preparing For a Medical Statistician Career

An undergraduate degree with a mathematics focus is a good starting point for a career in medical statistics. Professionals pursuing nursing or other care-focused degrees may also leave statistics career options open by taking a few college courses. Truly, successful coursework in calculus, linear algebra, and statistics is helpful no matter your college degree.

However, the most common entry point into a medical statistician career is a master’s degree in applied statistics. Students in this graduate program learn how to use statistical methods to analyze and to solve real-world challenges. Medical statisticians typically build advanced skills in the following areas during their graduate education:

  • Designing and evaluating experiments
  • Communicating research results to stakeholders
  • Employing programming languages, such as R and Python to solve problems
  • Using SAS software for data mining and analysis

Success as a medical statistician, however, extends beyond fluency in statistics and data analytics. That is, medical statisticians also need project management skills to oversee junior staff members and balance research responsibilities. They also require a rich understanding of medical ethics and professional standards.

Looking Ahead: Career Prospects and Salary for Medical Statisticians

A graphic of a bar chart and a trend line.

Increasing complexity in clinical research and health services means there is a high demand for medical statisticians. Overall, the U.S. Bureau of Labor Statistics (BLS) projects a 30% growth in jobs for all statisticians from 2022 to 2032.

This projection far exceeds the estimated 3% growth in jobs for all occupations during the same period. What this growth comes down to is a need for about 10,600 statisticians.

And this long-term demand for statistics professionals in health fields translates into above-average salaries. The BLS found a median salary of $104,860 for mathematicians and statisticians in 2023. Experience and specialized knowledge may lead to higher salaries, with the top 10% of earners reporting salaries of at least $161,300.

Aspiring medical statisticians can boost their career prospects by studying at a leading university. A graduate education focused on the latest methods and tools prepares statisticians for future opportunities. Michigan Tech University offers a Master of Science in Applied Statistics that achieves these goals.

Acquiring Medical Statistician Skills at Michigan Tech

Michigan Tech’s innovative Online MS in Applied Statistics Degree provides entryways into in-demand fields such as medical statistics and others. This program is also flexible. It offers seven-week courses for accelerated learning opportunities as well as three possible start dates. And Michigan Tech does not require GRE or GMAT scores in its free graduate application process.

Applied statistics students work with experienced faculty members on several subjects including the following:

  • Computational Statistics
  • Design and Analysis of Experiments
  • Regression Analysis
  • Time Series Analysis and Forecasting

Prospective students with undergraduate calculus, linear algebra, and statistics coursework are well-positioned for admission into the program. Also, a ten-week bridge course is available for incoming students who need to brush up on their linear algebra skills.

Learn More About Applied Statistics.

But a medical statistician is just one example of a possible data-driven career made possible by advanced education in Applied Statistics. Want to learn more about the courses and the program structure in Michigan Tech’s online MSAS program? Explore other career trajectories?

Linear Algebra Bridge Course Returns for Fall 2024

 A 3-D representation of Linear Algebra equations.

On Monday, September 16, 2024, Teresa Woods is once again teaching her ten-week, noncredit, asynchronous, online course: Linear Algebra: A Bridge Course for Prospective Applied Statistics Students.

If you’re unfamiliar with the term bridge course, it is a short, intensive, preparatory course. Bridge courses help learners acquire the necessary knowledge and skills to enter advanced study, which might mean an undergraduate program, graduate degree, or graduate certificate. Often, these courses are meant for students who have been provisionally accepted into a program.

Woods’ course is an effective, low-cost option for prospective students who need the linear algebra requirement to enroll in MTU’s Online Master of Science in Applied Statistics program. However, those interested in brushing up on their linear algebra, so that they can later apply to the MSAS program could also take it.

The course’s very practical curriculum covers the fundamentals of linear algebra as they are used in applied statistics. Some of the topics include, but are not limited to, the following:

  • systems of equations
  • vectors
  • matrices
  • orthogonality
  • subspaces
  • the eigenvalue problem

Students will benefit from an interactive learning experience that will make the concepts stick. That is, the course involves helpful instructor-led videos, extensive auto-graded exercises in Pearson’s MyLab Math, periodic review assignments, and regular instructor feedback.

What is Linear Algebra?

Algebra is a broad field encompassing the study of mathematical symbols and the rules for manipulating them. It includes various sub fields, such as elementary algebra, abstract algebra, and number theory.

Linear algebra, a specialized branch of algebra, focuses on the study of vectors, vector spaces (or linear spaces), matrices, eigenvalues and eigenvectors, linear transformations, and systems of linear equations. This foundational area of mathematics has applications in several fields, such as physics, computer science, engineering, economics, and applied statistics.

  • In physics, experts use linear algebra to describe physical systems, including quantum mechanics, classical mechanics, and relativity.
  • In engineering, those working in control theory, signal processing, and structural analysis recruit linear algebra tools.
  • Computer scientists use this branch of algebra in computer graphic creation, machine learning, data mining, and optimization problems.
  • Also, those in the field of economics apply linear algebra when modeling economic systems, analyzing input-output models, and optimizing resource allocation.

What is the Relationship Between Linear Algebra and Applied Statistics?

And, of course, linear algebra plays a key role in applied statistics.

Applied statistics is the implementation of statistical methods, techniques, and theories to real-world problems and situations in science, engineering, business, medicine, social sciences, and more.

It involves collecting, summarizing, analyzing, interpreting, and presenting data to make informed decisions, analyze scenarios, solve problems, and answer questions.

Applied statisticians often use linear algebra to analyze and then visualize data.

Applied statisticians also use advanced techniques, such as machine learning algorithms, to extract insights and patterns from large datasets. They work in a wide range of places: research institutions, the government, business and finance, universities, healthcare systems, and more.

These experts regularly apply linear algebra, primarily because of its ability to handle large datasets and complex calculations efficiently. 

What Are Some Real-World Examples of Linear Algebra and Applied Statistics?

Here are a few scenarios in which linear algebra and applied statistics work together:

  • A statistician working for Netflix might collect and then simplify data on user ratings for various movies. Next, they would represent that data as a matrix and train the model. By uncovering patterns in the ratings, they could then use the model to generate an effective recommendation system. This approach is also widely used in e-commerce sites and music streaming services.
  • Furthermore, a real estate agent might use linear regression, a common method for determining outcomes, to predict how housing prices will increase or decrease in the next year. This information would help them price houses in their portfolio, estimate their commission, and so on.
  • In the healthcare sector, professionals use linear algebra and applied statistics. Principal Component Analysis (PCA) helps reduce the complexity of a large dataset by identifying key patterns and relationships between variables. Through this approach, health officials can then predict and intervene on disease outbreaks more effectively.
  • And, of course, linear algebra and applied statistics work together in several processes involving elections. These include voter segmentation and targeting, predictive modeling, analyzing voting patterns, polling analysis, and redistricting and gerrymandering.

About Your Instructor

Teresa Woods, associate teaching professor in Mathematical Sciences, is helming this course. Woods also advises students and serves as assistant to the department chair.

Woods’ received her Master’s of Science in Mathematical Sciences from Michigan Tech in 2017. Her master’s report “ANALYSIS OF ALEKS MATHEMATICS PLACEMENT TEST DATA” combined her two areas of expertise (and passions): mathematics and educational assessment. That is, she holds both an MS in Mathematical Sciences and an MS in Education (with a focus on adult learning.)

If you take this course, you’ll benefit from an instructor who has considerable experience in teaching, a wealth of enthusiasm for elementary linear algebra, and a rich history in designing and delivering online courses. 

Reach Out if You’d Like to Learn More.

Need advice on whether this course is right for you? If so, please contact Teresa Woods at tmthomps@mtu.edu. Or if you have questions about our online MSAS program, contact Amanda at globalcampus@mtu.edu.

Three Ways Statistics Impact Elections

Statistics are involved in several stages of the election process.

125: that is the number of days until US Election Day, 2024. On November 5, the 47th president of the United States will be decided. So while campaigns are in full swing, and pollsters are making predictions, this blog focuses on the role of statistics in the election process.

At their most basic, elections allow citizens to exercise representative democracy by selecting individuals to occupy public office. Those selected then make critical decisions that impact citizens. And these ballots that officials tally are then transformed into statistical data, ultimately determining the election’s result.

However, statistics play a part in the election process long before voters cast their ballots. That is, officials use statistics to forecast election results, inform campaign strategy, and micro-target individuals.

An understanding of how statistics are used in elections, then, can enhance transparency for voters, as well as encourage all citizens to advocate for data privacy and security. Additionally, those interested in mathematics, statistical applications, and political science might be interested in learning about how statistics impact elections.

Statistics in Politics

Throughout history, statistics have played an important role in politics. Government bodies used statistics in the election process to support the formal decision-making processes that determine who will fill offices in the legislature. However, technological advancements, the accumulation of data, and the maturation of statistical models have made elections increasingly complex.

For example, in the past, politicians and their supporters would cast a wider net when campaigning for votes. But today, data analytics and digital resources allow parties to collect information about the public and then hyper-personalize campaign targeting. As a result, modern elections require statistical experts who can manage and leverage data while maintaining ethical standards related to trust, security, and privacy.

Below are the most obvious three ways that statistics impact elections.

Election Forecasts

Those creating election forecasts use legally available data and statistics to inform the public about the probable outcome of an upcoming election. Political statisticians recruit this data, along with reporting, historical patterns, and academic research to create a detailed account of the Senate and House forecasts.

In the United States, this process includes disclosing the favored party, estimating the number of seats in each House, and predicting whether the outcome will result in a majority government. In short, statisticians use a forecasting model to transform large data sets into meaningful predictions for future outcomes

How to Build an Election Forecast Model

  • Create a national database.
  • Clean and layer the data.
  • Plug the data points into a predictive model for forecasting.

Forecasting in Action

FiveThirtyEight is a website that uses statistics to predict election results.
The homepage of FiveThirtyEight on June 26, 2024

The popular website FiveThirtyEight, created by American statistician Nate Silver, is a staple of ABC News. The website’s primary objectives are advancing public knowledge and promoting transparency around voting outcomes.

To achieve these aims, it uses polling, economic, and demographic data to explore likely election outcomes. It also employs statisticians to build empirical statistical models for accurate election forecasts.

After the data is collected, experts then input it into Nate Silver’s forecast model. This model, which combines polling, economic, and demographic data, aims to provide an informed prediction rather than an unskilled guess.

And the website regularly updates its predictions too. For instance, on June 26, 2024, the site, after running 100 simulations, predicted that President Joe Biden and Donald Trump each had a 50% chance of winning the election. However, on July 2, 538 changed the prediction to 50% for Trump and 49% for Biden. And as the election nears, and uncertainty decreases, 538 claims its predictions will grow more accurate. This site exemplifies just one popular election forecast tool.

Election Campaign Strategy

The use of statistics in election campaigning has also changed dramatically. That is, historically, the only data that politicians and their supporters used to garner insights was that derived from the polls. In recent years, however, data and statistics have revolutionized election campaigns.

Today’s data-driven world offers campaign strategists a surplus of data points about past elections, voter preferences, and geopolitical influences. In addition, new communication platforms, such as social media, allow campaigns to profile their voters’ identities and needs. Statisticians can also harness publicly available data to inform campaign messaging, political priorities, and outreach.

Campaign research allows parties to investigate target audiences’ behaviors, attitudes, values, and beliefs to test campaign messaging, creativity, and delivery. According to The Commons Social Change Library, statisticians use the following quantitative and qualitative research methods to inform campaign strategy.

Quantitative Campaign Strategy Research

  • Benchmark Polls
  • Issue Polls
  • Longitudinal Surveys
  • Member Surveys
  • CATI (computer-assisted telephone interview) polls
  • Dial-testing

Qualitative Campaign Strategy Research

  • Deep dive interviews
  • Face-to-face focus groups
  • Online focus groups
  • Online communities

Once the previous research is complete, campaigners then test various messages. Alternatively, they might test the gap between their voters’ current stances and the desirable action. This job is a laborious one. Campaigners must strive for creating winning messages that make impactful arguments, define important issues, expose the opposition’s weak points, and tell compelling narratives.

Statisticians with a marketing background may excel in this area of research and persuasion. Why? They already have the foundational skills needed to create data-driven campaign strategies, from initial research to distribution.

Microtargeting in Elections

Before advanced data and statistics, campaigns often involved grass-roots approaches. These included direct mail, home visits, radio, television, and out-of-home marketing campaigns (ex., billboards, posters, etc.). Today, campaigns can leverage social media, digital marketing, and advanced data analytics to reach voters on their devices and tailor personalized messaging. This latter strategy is otherwise known as microtargeting.

Social media apps collect information and statistics on users in order to create targeted, personalized messaging.
Social media apps collect information on users to create personalized messages.

In microtargeting, the audience is segmented into specific groups, with each group receiving a message that speaks to their likes and needs. This profiling, though, is not new.

Consumers are already accustomed to online stores such as Amazon, as well as social media (TikTok, Facebook) understanding their preferences.

For instance, you purchase one book and Amazon recommends a similar one. You buy running shoes (a lot) and you’re now in a fitness/running channel.

Similarly, political parties and election campaigns use microtargeting to communicate with voters about their initiatives. The goal is influencing voting outcomes in their favor.

How Microtargeting Works

Micro-targeting uses statistics in a similar manner to that of election forecasting. First, statisticians collect and clean data points from a national database. Then, they layer on publicly available information, including email addresses, phone numbers, employment, education, purchasing patterns, IP addresses, etc.

Next, statisticians use predictive models to indicate for whom a voter is likely to vote and how likely a voter is to change their voting preference. These models also predict how lifestyle choices, such as being single or married, might affect voting behaviors. Statisticians also investigate how voters’ values align with topical issues like gun control, the climate crisis, abortion, immigration, and so on.

After the analysis comes the categorization. Each group is sorted into different channels. Each audience (channel) then receives personalized campaign messaging based on their beliefs and inclinations. The purpose is delivering the right campaign message, to the right voter, at the right time. (At its roots, microtargeting is a very deliberate form of kairos. In rhetoric, kairos is the identification of the critical moment to deliver a finely tuned persuasive message or to take an action.)

The Risks of Microtargeting

Advanced microtargeting, of course, has its downsides. Take the most famous example, which began in 2014. Cambridge Analytica, a political consulting firm, obtained the private Facebook data of tens of millions of users. It then unethically sold psychological profiles of American voters to political campaigns.

How did this microtargeting scam work? 270,000 Facebook users played with the supposedly innocuous personality profile app called “This Is Your Digital Life.” This app, created by scientist and psychologist Alexsandr Kogan, allegedly collected 5,000 data points from each participant.

What’s worse: participants didn’t read between the lines. When users gave this third-party app permission to acquire their data, they also gave the app access to their friends’ networks. The more friends = the more data exposed.

Kogan then sold this data to Cambridge Analytica. As a result, the company illegally compiled the data of about 87 million users who had not explicitly given Cambridge Analytica permission. The firm then used up to 50 million profiles for their predictive modeling. At the very least, the app developer breached Facebook’s terms of service by giving the data to Cambridge Analytica. After investigations began, the incident started a heated, nationwide conversation about the ethical principles of data, political targeting, and power. And about Facebook, data security, and cybersecurity.

Study Applied Statistics at Michigan Tech.

Election campaigning and increased microtargeting are very much still with us. Therefore, firms that generate value from personal data must consider the ways they acquire it, share it, protect it, and profit from it. Statisticians who work for these firms must also stay in line with ongoing legislative efforts that respect users’ privacy and security.

Curious about how statistics make a difference in elections? Are you fascinated by the data-driven side of political science? Do you want to ensure statistics are collected ethically? Alternatively, maybe you’re interested in developing the skills for collecting data and using applied statistics in business, government, finance, insurance companies, and more.

If you answered yes to these questions, Michigan Technological University’s Online MS in Applied Statistics offers students foundational knowledge in statistical science and methods while utilizing the latest industry-standard statistical and data analysis software. After graduation, you can set yourself apart in the competitive workforce with not only specialized skills, but also the accountability to act with integrity, honesty, and diligence.

And statistics jobs pay well, too. The U.S. Bureau of Labor Statistics (BLS) reported that, as of 2023, the median annual wage for a statistician was $104,860. Furthermore, the projected average growth rate through 2032 for jobs in these fields is 30%. That’s four times higher than the projection for all occupations in the same timeframe.

Upskill for the future with Michigan Tech’s Online MS in Applied Statistics.

Global Campus Grows

Whether it’s been covering new education fellowship partnerships, reporting on Michigan Tech’s collaboration with the MEDC, writing about innovative mass timber research initiatives, researching the gifts of adult learners, welcoming new team members, or rushing to keep up with Global Campus Vice President David Lawrence, this blog writer has had a busy year. And while all these initiatives, and more, have been underway, I’ve also had to keep track of Michigan Tech’s new online courses and programs.

Recent Online Programs at Global Campus

For example, in the last year, the College of Business added the online Tech MBA and the Master of Engineering Management. Both are accredited, 10-course programs that, in various ways, leverage your STEM expertise. Whereas the Tech MBA provides foundational business skills, the MEM allows students to customize degrees that merge engineering and business. To promote these programs, Dr. Mari Buche, David Lawrence, and his Global Campus team graciously led several online virtual interest sessions, which were all well attended.

Leadership and learning are indispensable to each other.

President John F. Kennedy

Furthermore, the College of Engineering met the learning and leadership challenge with its Master of Engineering, a professional terminal degree. This degree allows students to focus on either a HEV (hybrid electric vehicle) track or an engineering track. For the engineering track, learners can combine courses from several disciplines. In fact, the master of engineering is ideal for those collaborating with their employer to develop a program to meet specific on-the-job needs.

More recently, the Department of Applied Computing has also added two new programs to its roster: Public Health Informatics and Foundations in Health Informatics. Both certificates can be stacked to form a master’s degree. Like other HI programs, these prepare students for diverse roles in the data-driven healthcare industry. Guy Hembroff, the Health Informatics director, also ensured that MTU’s CHI students have memberships in HIMSS. HIMSS (Health Information Management Systems Society) is a global society. It enables health information professionals to access resources, enroll in seminars, develop networks, search for jobs, and much more. In other words, it gives MTU’s Health Informatics students an edge.

Global Campus Bridge Courses

Bridge courses are short, intensive, preparatory online courses that help learners acquire the necessary knowledge and skills to enter advanced study. This study might mean an undergraduate program, graduate degree, or graduate certificate. Often, bridge courses are for students who are provisionally accepted into a program.

Linear Algebra: A Bridge Course Offered Through Global Campus
Linear Algebra: A Bridge Course Offered Through Global Campus

For instance, in September of 2022, Teresa Woods, Associate Teaching Professor in Mathematical Sciences and Linear Algebra aficionado, taught our first bridge course: Linear Algebra. Her ten-week, asynchronous online course was aimed at prospective students who needed the LA requirement to enroll in MTU’s Online Master of Science in Applied Statistics program.

Woods’ course covered fundamental linear algebra concepts as used in Applied Statistics. Some of the topics included systems of equations, vectors, matrices, orthogonality, subspaces, and the eigenvalue problem.

To learn more about this course, email Teresa Woods (tmthomps@mtu.edu).

Linear Algebra is once again running for the Fall 2023 semester. And there are still a few seats left. Right now, the proposed start date is Sept. 18, 2023.

Newer Professional Development Opportunities

Fundamental Courses and Bootcamps

Global Campus also had the privilege of working with subject matter experts to promote in-demand professional development courses. Also known as continuing education and career training, these courses allow those in the workforce to hone skills, acquire specialized training, develop leadership abilities, and stay up-to-date on current trends.

Currently, Michigan Tech offers both non-credit and for-credit pd courses.

For example, during the summer of 2023, APS Labs rolled out its short, but rigorous course on Diesel Engine Fundamentals. Despite the turn to EV, this course recognized that diesel engines weren’t going anywhere soon. That is, diesel engines are still in light-duty vehicles, medium and heavy-duty trucks; in commercial vehicles (trains, trucks, buses, barges, and boats); in army vehicles; and in generators.

This course was conveniently available in both online and in-person versions. Its goal was educating those pursuing careers in the automotive industry, commercial vehicles, power generation, or related fields.

A Diesel Engine, which was studied in the APS Labs short course for Global Campus
A Diesel Engine

Also, Kevin Johnson, Assistant Teaching Professor, Manufacturing and Mechanical Engineering, lent his significant expertise to summer students. He taught an an intense 20-hour in-person hydraulics bootcamp. In his course, students learned about several topics crucial to hydraulics, such as valves, pumps, motors, circuits, and closed-loop hydrostatic systems.

Upcoming Professional Development Courses

Python for Modern GIS

A person working on GIS with Python, one of the courses taught though Global Campus
GIS Workshop

Furthermore, recognizing the need for more Python professionals in the GIS world, Parth Bhatt (Assistant Teaching Professor / Researcher from the College of Forest Resources and Environmental Sciences) is offering a 7-week, asynchronous, online course for Fall 2023.

His Python for Modern GIS and Remote Sensing course will help students learn beginning and immediate-level applications of Python for understanding and writing simple scripts, automating workflows, and solving day-to-day, real-world geoprocessing tasks in the ArcGIS ecosystem and open-source platform.

Dr. Bhatt, a dynamic teaching professor who lives and breathes GIS, is also on deck to develop online for-credit certificates for his department. Stay tuned for more developments.

And, yes, you still have time to register for Bhatt’s course.

Civil Asset Management

As well, the Department of Civil, Environmental, and Geospatial Engineering has recently added a 3-credit, synchronous online course in Civil Asset Management. This course is taught by Mark Declercq, who brings three decades of valuable, practical civil asset expertise to the classroom. In fact, as Grand Rapids Engineer, Declercq was one of the first experts with boots on the ground during that city’s massive flood event.

Civil Asset Management (CEE 5390) will help students develop long-term plans, as well as the strategic, critical thinking they need to recognize and maintain the value of our all-important civil assets. Declercq also maintains that to develop resilient and affordable solutions and to tackle upcoming sustainability challenges, engineers definitely need Civil Asset Management skills.

Keep Up With Global Campus as We Learn and Grow

In the future, Global Campus plans to offer additional non-credit and for-credit courses and programs. Our goals are advancing the personal development, career goals, and leadership opportunities that come with education. We also recognize the importance of challenging all learners to grow, to think creatively and critically, and to prepare for tomorrow.

We’ll keep you posted as we assist in developing and supporting new programs. For updates, read this blog or follow us on social media.

And remember, regardless of where you are in your educational journey, whether you want to take a course for fun or for your future, it is never too late to start learning.

Anyone who stops learning is old, whether at twenty or eighty. Anyone who keeps learning stays young.

Henry Ford