Home » Departments, Education, Featured, Statistical Literacy

Teaching Statistical Literacy to Non-STEM Majors: Challenges and Opportunities

1 September 2017 173 views No Comment

Michelle Everson is a professor and program specialist of statistics at The Ohio State University and former editor of the Journal of Statistics Education. Since earning her PhD in 2002, she has been teaching introductory and intermediate statistics courses. She is particularly interested in distance education and ways to actively engage students in online learning environments.

 

Ellen Gundlach has been teaching and coordinating several introductory-level courses since 2002. She co-authored a probability book with Mark Daniel Ward and has also won several teaching awards. She is an associate editor of the Journal of Statistics Education.

 

We teach statistical literacy to non-STEM majors in public mid-western universities using large-lecture traditional, fully online, and flipped teaching methods. Between 500 and 1,100 students enroll in our courses each semester, and we both not only teach sections of these courses, but we also serve as the course coordinators (with valuable assistance from many talented graduate teaching assistants (GTAs) who lead recitation sessions and sometimes serve as lecturers).

Our students are more likely to be consumers of statistical information than producers of statistics. As such, we want them to be able to develop the necessary skills to reason carefully through statistical information to better understand important issues in society today, such as whether to vaccinate their children, whether the American Community Survey is a good use of our tax dollars, how Big Data is something to be both excited and cautious about, how to interpret medical test results, and whether to panic when a local news station suggests there may be a cancer cluster in the community./

GAISE Guidelines

The 2005 Guidelines for Assessment and Instruction in Statistics Education (GAISE) College Report and the recently revised GAISE College Report have had a profound effect on our thoughts about how to best structure and teach our statistical literacy courses. The GAISE College Report recommends teachers of first courses in statistics (1) teach statistical thinking; (2) focus on conceptual understanding; (3) integrate real data with a context and purpose; (4) foster active learning; (5) use technology to explore concepts and analyze data; and (6) use assessments to improve and evaluate student learning.

Given that our courses are designed to help students become critical and thoughtful consumers of statistical information, there is a strong emphasis in our classrooms on learning the language of statistics and understanding the big ideas.

Technology—often in the form of applets and statistical software such as JMP—is used to illustrate different ideas and concepts (such as distributions of sample statistics or correlation), to graph data, and to perform calculations. Whenever possible, we attempt to use real data that has been gathered from our students or that comes from published research. We use this data to better illustrate the investigative nature of statistics. We want our students to know that data, if gathered under the right conditions, can help us answer important questions. We also want our students to appreciate the need to ask critical questions when they are trying to make sense of data.

To determine if students are meeting the learning objectives in our courses, we use a variety of formative and summative assessments, including clicker questions, weekly homework assignments and lab activities, projects, and exams.

What to Teach

Topics That Work

Our students appreciate daily connections to current events so they can see immediate relevance of statistical literacy to their daily lives. This can be accomplished through a “statistics in the news” segment incorporated into each lecture or with whole lectures written to explain a particular hot topic.

For example, during election season, understanding how pre-election polls could vary from each other and the final election outcome allowed us to discuss sampling variability, sample size, margin of error, sampling design, response bias, undercoverage, and nonresponse. When the Affordable Care Act was being debated by Congress, we talked about how the Law of Large Numbers is important to the business plans of insurance companies. A few years ago, Facebook announced results from an experiment it performed on hundreds of thousands of users without explicit permission, and this led to discussions about exactly what informed consent means and whether Facebook is considered a “public space” that would allow some exemptions from usual informed consent rules.

We’ve enjoyed sharing and discussing news articles about topics we think our students will relate to and find interesting. For example, we’ve talked about whether Facebook use has a negative effect on GPA, whether dogs like to be hugged, if college students will literally “text through anything,” if using electronic devices during class affects learning, the typical cost of a wedding, the best day of the week to weigh yourself, the best excuse to give if you ever need to call in sick to work, whether the outcome of a coin toss can predict the winner of a football game, and the calorie content of Chipotle burritos.

We also frequent YouTube and share many video clips in our classrooms. Videos from the “Colbert Report,” “Saturday Night Live,” “Bill Nye the Science Guy,” Neil deGrasse Tyson’s “Cosmos,” TED talks, and, of course, Hans Rosling’s “The Joy of Stats” all enliven explanations and discussions while showing the wide-ranging relevance of statistics.

Topics That Don’t Work

Many of our students have math anxiety, or at least a strong dislike for anything math-related, so we carefully plan our courses to build confidence and show students many examples with step-by-step guidance and ample opportunities to practice alone and in groups. We make it clear how statistics and math are different, and we attempt to emphasize that we use math as one of our tools to understand what is happening in the world. About one-third of our topics are taught primarily with words instead of numbers. Proofs are not as important for students as understanding what the results of their data explorations mean and how to communicate these results to others.

How to Teach

Active Learning

We attempt to incorporate many opportunities into our courses for students to play an active role in the learning process. Although we both teach in large lecture halls, we use clickers to engage our audience and give as many students as possible the opportunity to be part of class discussion. This has allowed us to sometimes gather data from our students that we can incorporate into future class discussions to illustrate particular ideas or concepts. Clickers also allow us to gauge whether our students have read assigned course readings or whether they have misconceptions or misunderstandings. At The Ohio State University, students have free access to a system called Top Hat that allows them to use their cell phones as clickers, therefore turning a device that sometimes distracts students into a valuable educational tool.

Whenever possible, we try to bring as much activity and discussion as we can into the lecture hall. One favorite activity involves trying to conduct a horrible experiment with students by simply dividing the lecture hall in half and asking each half of the room to try to remember a different list of words read aloud by the instructor. Many lurking variables are built into the experiment, and, because students are active participants in the experiment, they often have much to say about the shortcomings of the experiment and how a better experiment could be designed to assess memory for words. Our hope is that this exercise will be more powerful than simply describing what an experiment is and telling students about the terms and ideas related to experimental design.

Helpful Resources for Planning Courses

While some of our best ideas come from talking to colleagues or paying attention to what is in the news, here are some other resources we have found especially helpful:

Consortium for the Advancement of Undergraduate Statistics Education (CAUSE)

This site has been compiled by statistics educators for more than a decade, beginning with a National Science Foundation grant. We especially like the webinars (Teaching & Learning, Activities, and Journal of Statistics Education author talks), but there are labs, projects, jokes, songs, data, lesson plans, quotes, and many other resources.

United States Conference on Teaching Statistics (USCOTS) and Electronic Conference on Teaching Statistics (eCOTS),

These conferences (in person for USCOTS, online for eCOTS) are wonderful ways to connect with the statistics education community. Having conversations over meals at USCOTS is ideal, but even looking through the resources posted on the CAUSE website once the conferences have ended can be inspiring. Many big changes in the statistics education field such as teaching hypothesis testing through simulation were born and nurtured at these conferences.

Journal of Statistics Education (JSE)

This free online journal contains excellent scholarly articles about research in statistics education, interesting data sets with suggestions for how they can be used in the classroom, interviews with leaders in the field, and summaries of interesting articles in other journals that could be relevant to teaching statistics.

Multimedia Educational Resource for Learning and Online Teaching (MERLOT),

This is a collection of online resources such as applets and modules that can be used as part of lab activities, lecture demonstrations, or at-home practice for students.

Science Daily

We both signed up for the daily email newsletter, which updates us on many new scientific articles published in a wide variety of fields. Almost every day, we can find a good experiment or observational study to bring to our students from this newsletter.

Assessment Resource Tools for Improving Statistical Thinking (ARTIST)

This is a collection of assessment resources for first courses in statistics. Over the years, we’ve used many of the assessment items from this site and been inspired to create course projects based on sample projects included here.

When students are not in our lecture halls, they are in smaller recitation sections in which they work through lab activities with their peers under the guidance of our GTAs. Many of these activities involve wrestling with problems in which students need to apply what they have learned through lecture and course readings. Students are encouraged to work through these problems with peers and use technology to arrive at solutions. We attempt to structure each lab activity around a question that students need to answer, often with real data.

As an example, in our first recitation activity (inspired by a webinar from the Consortium for the Advancement of Undergraduate Statistics Education [CAUSE], we break students into groups of three and give each group two small rubber pigs from the game Pass the Pigs. We tell students to imagine they are trying to gather information to be used to design a game that will incorporate the pigs. We also tell students that each pig can land in one of six possible ways, and we ask each group to come up with and execute a plan to gather enough data to determine how often a pig would be expected to land in each of the six positions. Each group is then asked to summarize their data and share it with their peers.

We start our semester with this activity because we’ve found it to be a good way to introduce the idea of statistics as an investigative problem-solving process, and we also find that we can use it to illustrate and foreshadow many of the topics that will be covered in the course.

For instance, students have to think carefully about what it means to “summarize” a data set, and it’s always interesting for us to see what kinds of summaries they choose to share. We can question students about the sample size they choose and what they think might happen if they were to increase the sample size. We can ask them if they would feel comfortable generalizing their results to a larger population, and we can ask them if they think factors such as how they tossed the pigs and the surfaces they tossed the pigs onto might have had an effect on their results. We can even present a claim to them about how the pigs should land, and we can ask them if they think they have found evidence in support of or against this claim.

For all these reasons, we find this to be a rich activity to which we can refer many times to better help students see the connections among different topics in the course. It also tends to be a fun way to start a new semester and break the ice with our students.

Other recitation activities have involved exploring data on movies to determine if a movie’s budget affects the revenue it generates, using Zener cards to determine if students have ESP (to illustrate important ideas of hypothesis testing), playing games (such as those developed by Shonda Kuiper ) to better understand the principles of experimental design, and examining the distributions of colors in M&M candies to illustrate the idea of the sampling distribution.

We have also tried to use different projects in our course. One group project (inspired by work presented at the Electronic Conference on Teaching Statistics (eCOTS) involved finding an advertisement that uses data to make a claim about a product and then doing research to determine if that claim is credible. Another project involved asking students to keep track of data from their daily lives (e.g., number of steps taken per day, minutes spent texting per day, calories consumed per day, etc.) for several days and then use technology to explore the data. Yet another project involved asking students to keep a journal of news reports they were seeing that involved statistics in some way (e.g., poll results, graphical displays of data, descriptions of research studies, etc.).

Big Data is a hot topic in the news, and while our students might not have the skills and technology to analyze Big Data, they are contributing to Big Data with daily activities online, so another project asks them to keep a 24-hour data diary in which they record everything they do within a 24-hour period where data could be collected from them, investigate two privacy policies, write a “Big Data privacy bill of rights,” and share an exciting use of Big Data in a field that interests them.

One other favorite project we have used is an online discussion and investigation into the research on one of a student’s own good or bad habits. This gives our students the chance to talk with each other about a topic of personal interest to them, and students seem to enjoy the chance to get to know each other and critique research that has immediate implications for how they behave each day. Details about several of these projects were shared through a CAUSE webinar.

Finding Ways to ‘Hook’ Students During the First Lecture

First impressions are important, and we constantly strive to put our best foot forward on Day 1 of every new semester. We know our students are often not happy to be taking a statistics course, and we sometimes question if they see the relevance of it. We want our students to walk away from our courses with a new appreciation for the fact that data are all around them all the time and there are many decisions they will make in their lives that will require a careful and mindful synthesis of statistical information. We want to provide students with the tools to critically evaluate the statistical information we know they will be bombarded with on a regular basis outside our classrooms. We also want them to appreciate the active nature of our courses.

A typical first day might involve asking students to meet and greet some of their peers. Students are given a few minutes to meet as many of their classmates as they can and gather information from each classmate. This activity, and the information gathered, can then be used to illustrate several of the new terms students will read about in the first two chapters of their textbook. We’ve found the activity works even in large lectures halls.

We also like to spend the first day sharing several snippets from the news. For instance, we will compile several recent stories that include the results of a study or survey aimed at answering a particular research question. We’ll share the research questions with students and ask them to tell us what they think the answers are. We will then tell students what the news stories actually claim the answers to be. Many of these stories are ones we’ll come back to at different times during the semester to introduce different concepts, and some of the stories have flaws we will eventually critique as a class. Our hope is that sharing these stories right away will better illustrate the relevance of the course content and the usefulness of the content in navigating daily life.

We also try, as much as possible, to not spend a great deal of time talking about the syllabus on Day 1; instead, we might have a syllabus quiz or a syllabus scavenger hunt activity for students to work on outside of class.

Learn About Student Interests and Incorporate Them into Their Work

It’s critical, particularly because we teach large courses, for us to get to know our students and learn about their backgrounds, interests, and concerns. Students from many departments take our courses, and this makes for a great deal of diversity in our classrooms. We might have students who are brand new to college or students who have put off our courses until the end of college and are just about to graduate.

The majority of our students are taking our course because they must to fulfill a general education requirement, and this is not necessarily a course they would have chosen on their own. Some students might have negative attitudes or misconceptions about the course content. As soon as we can, we attempt to gather information from our students to help ease their concerns and find out more about what they are passionate about.

During our first lectures, we often accomplish this by handing out index cards and asking the students to write down answers to questions about themselves. We then collect the cards and spend time reading through and responding to them. We might notice, for example, several students from similar majors, or several students who enjoy athletics or are involved in athletic programs on campus, and this might give us ideas for examples we can share during future class discussions. We can also reassure the students who have anxiety about math that this is a course designed for people exactly like them, and we are happy to help them along the way.

Traditional, Online, Flipped Delivery Formats

We both teach our courses in multiple formats, and this gives our students more opportunities to choose a format they think will work best for them or fit best into their schedule. In addition to teaching in a traditional, face-to-face setting, we also teach sections almost completely online (with the exception of exams that students must take in proctored settings). At Purdue, there is also a flipped section that presents the lectures online but has the students meet weekly to solve problems or do hands-on activities to reinforce the material. We have found the attitudes, learning, and performance of students in all three types of section to be fairly similar, but the students appreciate having an opportunity to choose.

What We Want

We both enjoy the challenge of explaining the importance of statistics to an audience that might not have an appreciation for this discipline before taking our class. We want to send our students into the world as informed citizens who are curious and willing to challenge speakers, reporters, politicians, medical care providers, advertisers, and others about where their data and conclusions come from. We are not trying to create cynics, but we do want to empower our students to ask questions and think critically, rather than simply accept what they are hearing and reading at face value. We want our students to appreciate logic and good investigative procedures. We want them to understand the quality of their sources of information, and we want them to appreciate how much information the right graph can convey, what sampling variability is, and how a properly selected sample can represent a much larger population. We know our students will leave our classrooms and be expected to make sense of a world filled with data, and, most of all, we want to make sure they have the tools to get them started on their journeys.

1 Star2 Stars3 Stars4 Stars5 Stars (1 votes, average: 5.00 out of 5)
Loading...

Leave your response!

Add your comment below, or trackback from your own site. You can also subscribe to these comments via RSS.

Be nice. Keep it clean. Stay on topic. No spam.

You can use these tags:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

This is a Gravatar-enabled weblog. To get your own globally-recognized-avatar, please register at Gravatar.