Home » Departments, Education, Featured, undergraduate statistics education

FUNdamental Engagement in Statistics Education

1 September 2017 138 views No Comment
Lawrence M. Lesser, The University of Texas at El Paso

    Larry Lesser is a statistics/mathematics education professor at The University of Texas at El Paso. He has written books and 100 peer-reviewed papers, done editorial service for JSE and SERJ, and won statewide/regional teaching awards. His work on mathematical sciences “fun” has yielded 75 published songs, a score of national contest awards, myriad conference performances and radio/TV appearances, several issues of The Mini Page, and five journal articles.


    As John Tukey says, statistics folks enjoy getting “to play in everyone’s backyard,” so it’s no surprise we enjoy various forms of play. Sometimes playful content-related entertainment (or, “edutainment”) is showcased when the ASA celebrates a special anniversary (e.g., the recent 175th), when CAUSE A-mu-sing contest songs or videos are played at the United States Conference on Teaching Statistics (USCOTS), or when videos are displayed from a contest such as the ASA’s 2011 “Promoting the Practice and Profession of Statistics” competition. The use of fun in statistics education also has implications for teaching, educating teachers, and educating the public.


    Given how important and interesting we know our discipline is, it may be hard to believe many of our students (especially our non-major students) have high levels of statistical anxiety that keep them from fully engaging in and embracing our course. In fact, statistics anxiety is fairly common in students and society. Education researchers have developed and validated two discipline-specific instruments to assess this—the Statistical Anxiety Rating Scale (STARS) and the Statistics Anxiety Measure (SAM). Fortunately, anxiety is more amenable than attitudes (which are considered more stable over time) to being changed by appropriate interventions.

    One strategy several researchers have suggested for lowering anxiety (so students can devote more mental and emotional energy to learning) is selective introduction of material or moments incorporating fun. In a recent book chapter about statistics anxiety, I cite many conceptually oriented statistics books over the last few decades that employ humor, irreverence, or a mode such as cartoons to lower students’ statistics anxiety. That said, there are many other methods of reducing anxiety, and reducing anxiety is just one motivation many instructors consider when deciding if or how to use fun.

    The five top (by far) motivations my colleagues and I identified (in descending order of how often they were chosen on a survey of n = 249 USCOTS 2011 attendees, a 66% response rate) in our March 2013 Journal of Statistics Education (JSE) paper are: increases engagement; increases learning; makes teaching more enjoyable; reduces anxiety; and builds classroom community.

    There’s no doubt the inimitable Hans Rosling (who passed away in February) was enjoying and sharing a healthy dose of fun whenever he presented playful graphics and went into “sportscaster play-by-play mode” to narrate a data animation.


    Fun comes in many forms, and my colleagues and I have compiled more than a score in our JSE papers—cartoons, celebration days, clothing, comic strips, food, games (commercial, cultural, or other), game shows, humor/jokes, kinesthetic activity, literature (e.g., short stories), magic, media bloopers, movies, music/songs/raps, poems, quotations, statistics fun books, strange news, striking examples, videos, and wordplay (e.g., anagram puzzles). Different forms of fun may have different interactions with students’ learning styles and/or cultural norms, and our 2013 JSE paper indicated some interaction between form of fun and instructor gender: Male instructors were more likely than females to use humor, while females were more likely than males to use games.

    Fun can also be classified by how much risk may be involved—ranging from low (e.g., prepared “safe” jokes, humor, and cartoons; pressing PLAY on professionally recorded song/video clips) to high (e.g., singing an original song or improvising humor). Whatever the modality and level of risk, the common denominator is that the fun uses characteristics of surprise, humor, or play (SHoP) in the service of education. By “in service to education,” we mean the fun is connected to content in the course and not used as a generic distraction.


    The discipline of statistics happens to have one of the most searchable, curated repositories of discipline-specific fun: the CAUSE fun collection. Curated for more than a decade by Dennis Pearl of Penn State and me (with occasional input from many others), this collection is housed within the digital library of the Consortium for the Advancement of Undergraduate Statistics Education (CAUSE) and is one of CAUSEweb’s most-visited resources. The collection contains 718 (and counting) items spanning cartoon, art gallery, game, joke, magic, poem, puzzle, quote, song, story, and video.

    An example of a song and a cartoon from the collection appear in Figures 1a and 1b. Also, the collection includes an introduction (organized by modality) to the literature on using fun in teaching statistics, together with lesson guidance on using fun items:


    Figure 1a. A song from the what-p-value-means CAUSEweb collection


    Figure 1b. A cartoon from the CAUSEweb collection


    While the CAUSEweb collection is the largest and most searchable collection of fun items in statistics, it is certainly not the only source. For example, there is a collection of 200 statistics jokes, indexed by topic, compiled by the late Illinois State University Emeritus Professor of Psychology Gary Ramseyer and now maintained by his daughter, Vicki S. (Ramseyer) Morrow.

    Much activity on fun in statistics can be traced to the biannual United States Conference on Teaching Statistics (USCOTS). For example, the 2008 JSE review paper on fun came from an invited breakout session at USCOTS 2007, and the 2013 JSE paper started with questionnaires collected at USCOTS 2011 by a team of 11 authors, most of whom were in the CAUSE-supported Study of Fun cluster launched at USCOTS 2009. Beyond the cluster’s joint work, many individuals have done their own work on fun:

    • Shonda Kuiper’s publications (e.g., November 2015 The American Statistician) and NSF-funded work on the use of learning through games.
    • Mark Glickman’s and my 2009 review paper (in Model Assisted Statistics and Applications) about using magic to teach probability and statistics and his USCOTS 2011 action footage
    • Michael Posner’s performance of statistical magic and song at USCOTS
    • Pearl, John Weber, and I have received two NSF grants for preparing and assessing fun items (e.g., cartoons and songs) in the teaching of statistics.


    Because some believe that fun has no “serious” role in the classroom, it is important to acknowledge and address common hesitations instructors may have, as my coauthors and I identified in the 2013 JSE paper, in descending order of how often the hesitation was chosen by the instructors surveyed:

    • Can’t quickly find good examples
    • Uses too much class time
    • No skills/talent
    • Need to be perceived as serious by students
    • Weak evidence of helping student learning
    • Incompatibility with students’ cultures
    • Size of class
    • Need to be perceived as serious by colleagues/supervisor

    Many of these hesitations are readily addressed. Once instructors find out about the ever-growing user-friendly CAUSEweb collection, they can much more readily find good examples. If class time is a concern, an instructor can, for example, have a statistics song playing or a cartoon displayed as students enter or as papers are handed back, or it can be assigned to experience on a course webpage outside of class time. (The song in Figure 1a literally takes 10 seconds.)


    Well-chosen, well-placed statistical fun can be used effectively to build community in your classroom, in your department, or at your conferences. Nationally, there is also a community of kindred fun statistical spirits that always has room for newcomers. After all, it is important to make sure the study or practice of fun is not viewed as the domain of only those with rare talent or creativity.

    Opportunities to get involved include the following:

    Signing up to be a rater of existing CAUSEweb items and be rewarded for it in the CAUSE rewards program

    Submitting a new fun item to Dennis Pearl at dkp13@psu.edu for possible addition to the CAUSEweb collection; it can even just be an idea for an item—for example, CAUSE has access to a professional cartoonist who can take an idea for a cartoon and manifest it with professional quality—or it can be a suggestion of a specific learning objective not covered by any item in the collection

    Joining periodic “fun folks” virtual or in-person gatherings, such as the Birds of a Feather lunch table at USCOTS this past May

    The concern about fit with student culture (which applies not just to fun items but also, for example, to the context of data sets discussed in class) can be addressed with an initial survey (of student backgrounds and favorite bands, movies, etc.) and by making sure to avoid, for example, items whose humor comes at the expense of individual students or marginalized groups, or that unduly invoke or stereotype partisan politics, religion, sexuality, etc. In general, each instructor must reflect upon what dosage and modalities fit their personality, their course objectives, and the culture of their institution and students.

    Most modalities work equally well in large classes as in small classes (in fact, a larger roomful of people can more readily yield a critical mass of people singing along to a statistics song). Special skill or talent is not needed when an instructor can, for example, simply press play on a video or song from the CAUSEweb collection.

    By keeping the fun items connected to course content and not overdoing their frequency (e.g., education/humor researcher Avner Ziv found three or four jokes per class meeting to be an optimal dosage), the instructor should sufficiently maintain overall seriousness in students’ eyes while increasing student engagement and attention. And citing some of the emerging critical mass of research on the use of fun can address the need for evidence and being taken seriously by colleagues.

    Getting students involved in the creating or presenting of fun items can be a wonderful learning opportunity and community-building experience, whether done as a required or optional part of the course, and whether done as individuals or in teams. There are well-known examples of statistics educators who have had success with student-created items such as jokes (see John Wierman’s piece in the August 2016 The American Statistician), videos, or cartoons (see Page 69 of the 2016 Guidelines for Assessment and Instruction in Statistics Education College Report. Often from the encouragement of their instructor, many students have submitted entries (song, cartoon, video, poem) to the biennial A-mu-sing statistics fun contest or to the monthly blindly judged cartoon caption contest sponsored by CAUSE, and many of those student entries have won recognition or prizes.


    While it may sound surprising that there has been serious research on something like humor or fun, there are international refereed journals on humor research, and research on humor in educational settings actually goes back at least four decades. Our focus here, of course, is specifically on the context of teaching and learning statistics.

    Because humor and fun are context-dependent, it is not surprising that some research in this area has had a focus on ecological validity and sustained duration. For example, Ziv reported in 1988 in the Journal of Experimental Education on a naturalistic experiment in which an instructor (trained in the use of educational humor) taught two sections of a semester-long introductory statistics course, in which the humor condition section (which had three or four jokes or cartoons inserted into each lecture) had average grades nearly 10 percentage points higher than the section with the nonhumorous condition.

    Other studies are not comparative in nature, but focus on a better understanding of the role or dynamic of humor or fun in the course. For example, in the June 2015 Transformative Dialogues, Reynaldo Reyes and I reported on a case study of multiple sources of data from an instructor’s introductory statistical literacy course. Findings revealed that students in this course valued, engaged in more, and felt they better understood the material when the instructor enacted a pedagogy that built on the emotional aspects of teaching and learning. Students’ responses indicated it was more about the instructor’s immediacy or delivery than the specific fun item/activity, and this is consistent with the findings of other researchers.

    The reaction or direction of ensuing conversation after the insertion of fun into a lesson might differ in ways that set up additional content connections, and that trajectory can have ripple effects on the tenor of the rest of that day’s lesson. Because of such challenges of having an instructor teach one section with fun and an otherwise identical, replicable control section without fun, some studies seek to remove the role of the instructor. Removing instructor variability has the secondary benefit of addressing the second most common hesitation on using fun mentioned by college instructors in our survey—namely, lacking skill/talent. In particular, researchers may conduct a student-randomized experiment in which the treatment group has fun artifacts inserted into an otherwise identical copy (e.g., of a reading, test, video, customizable open-source textbook, etc.).

    In a 2006 issue of Humor, Ronald Berk and Joy Nanda report their randomized experiment with a graduate biostatistics of n = 98 students to assess the effect of adding humor to exam items, exam directions, or both on test anxiety and test performance. In a 2006 issue of College Teaching, R. L. Garner reports that 42 of 94 college students whose assigned three 40-minute statistics lecture videos had (three each) humor inserts gave significantly higher ratings of the lessons, how well the lessons communicated information, and quality of the instructor. They also recalled and retained more information on the topic (each p < .001) than did the 52 students who were randomly assigned to receive those videos without the inserts.

    Asil Özdoğru and Robert McMorris in a 2013 issue of Humor reported that n = 156 undergraduates studied six one-page concept readings where three had content-related cartoons inserted and three did not. The inserts appeared to yield favorable student attitudes, but did not affect learning. David Neumann, Michelle Hood, and Michelle Neumann review other studies in their 2009 article in Journal of Statistics Education.

    In 2016, Pearl, Weber, and I published a randomized experiment in Journal of Statistics Education in which students (within several classes at a two- and a four-year college) were randomly assigned to either always receive or never receive a fun insert (e.g., cartoon, song, etc.) into each of a dozen otherwise identical readings within the course learning management system. Students exposed to song inserts performed an average of 7.7 percentage points better on the six related multiple-choice items embedded on exams (p ≈ 0.04).

    Our experiment’s promising results with songs (relative to cartoons) and a conjecture that songs may offer more sustained engagement (e.g., due to having both visual and auditory channels of input, more mnemonic power, or inviting students to sing along) led us to explore using interactive statistics songs in which students supply concepts or context/examples (i.e., going beyond low-level recall) to be inserted (somewhat like Mad Libs) into a song that could then be played back to the student. This led us to our subsequent NSF grant (Project SMILES: Student-Made Interactive Learning with Educational Songs for introductory statistics) to create and field-test (with another randomized experiment) the effectiveness in enhancing learning and reducing anxiety of interactive songs spanning the introductory course.

    A collaborative with professional expertise in both statistics and songwriting produced more than 20 songs high in aesthetic and pedagogical values, and the web-based platform allows auto-grading of student inputs. It is also usable in lecture, flipped, or online courses. When college students tried the first version of the SMILES-style songs in a lab, they thought the activity reduced anxiety and had benefits for learning the material. They also supplied feedback that informed the refinement of the intervention for a larger follow-up trial now underway. Upon completion of the research study, the SMILES collection will be available on CAUSEweb for all instructors.

    A nice feature of this innovation is that it should be readily transferable to other disciplines. Indeed, we are launching VOICES: Virtual Ongoing Interdisciplinary Conferences on Educating with Song. Primarily targeting STEM and health sciences, this forum debuts this month (September 27–28).


    College statistics courses are often taught by faculty without a PhD in statistics who are housed in a variety of disciplines and departments. Thus, ways we develop to make statistics more understandable, relatable, and enjoyable can have that much more of an impact across a campus.

    Fun can be used to make an impact even beyond a campus by helping popularize and humanize our discipline to broader audiences. There have been many recent outreach efforts in statistics, such as the ASA’s This Is Statistics career awareness website and campaign launched in 2014, the quinquennial World Statistics Day (inaugurated October 20, 2010, by the United Nations Statistics Division), or the World of Statistics website (the successor to the International Year of Statistics campaign of 2013). Such efforts often can or do employ bits of statistics fun to help attract and engage new audiences.

    Statistical outreach can also happen within venues of mathematics organizations. The National Museum of Mathematics (MoMath), North America’s only major mathematics museum, has a couple of hands-on exhibits with probability or statistics content (e.g., in Edge FX, a quincunx interactively explores binomial distributions and when they are approximated well by normal distributions) and has had statistics songs played at some of its events. Statistics played a role in most entries in the “QL in the Media” contest held by the MAA’s SIGMAA on Quantitative Literacy and in many Media Clips columns in the NCTM practitioner journal Mathematics Teacher.

    Statistical outreach can apply also to the mainstream media, such as the “Stats and Stories” podcast series (with 38+ episodes archived at statsandstories.net) sponsored by the ASA and Miami University. Communicating to an even larger, younger audience is an enjoyable departure from almost all of our usual work, forcing us to think about issues in the most accessible, big picture manner possible. I included some playfulness in an issue of The Mini Page, a weekly syndicated children’s feature in 500+ newspapers with roughly 20 million readers. I wrote about statistics polls three months before the most recent presidential election. Also on that topic, I was the featured presenter (and lyricist) on a 2012 episode of the children’s educational television show “Blast Beyond,” which aired on a regional PBS station.

    The thoughtful use of fun is a great way to make our subject matter accessible, engaging, and memorable to our students, as well as to promote and celebrate our discipline.

    1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)

    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.