Thank you, Doris! Some topics are definitely harder to write songs about than others and before our project there were several topics that had probably never before been the focus of a song. As for effectiveness with students, we're still analyzing data, but results may end up showing an effect by things in addition to the topic of the song, such as the role of the song (i.e., whether the song is being used to recall facts/procedures, introduce a concept, synthesize and apply knowledge to a context, etc.), the genre of the song, the length of the song, etc.

Thank you, Ron!  We haven't used sonification (e.g., this example) in this project.  That said, our interactive song "Central Limit Theorem" has students reflect on different distribution shapes and near the end of our song "Chi-squared Dance" is a lyric line ("The chi-squared distribution has a long right tail") whose melody follows the contour of that distribution's right tail.  Interestingly, one of our PIs (Dennis Pearl) was also a PI on a major project on probability distributions.

Thanks for the kind words.  One goal of the project is to lessen students' anxiety about statistics and quantitative evidence generally.  We are hoping what you describe as "delightful" has that result.

Thanks for your interest, Jessica! SMILES songs have been used in courses with diverse delivery modes.  A teacher could do it during a live lecture in any size classroom by having students use clickers to vote on responses to fixed-choice questions and having students text or call out suggestions for the open-ended questions.  Or students could each be in front of a computer with earbuds or headphones.  Or a teacher could assign particular songs for the students to access at home as homework.  We're actually giving a 2-hour virtual workshop on Project SMILES pedagogy on Thursday May 24 at 11am Eastern, as part of a week-long virtual conference on teaching statistics that you can register for here. Also, we're working to create more supplementary resources for instructors on the SMILES website to address not only different course formats but also provide information about how their students performed on the prompts, offer sample assessment items, offer tips on creating interactive songs, etc.). We also plan to create a faculty community of instructors using SMILES so we can get good feedback and learn from each other and we launched the VOICES initiative to gain insight into alternative ways that song can have a positive effect on learning across disciplines.

Thanks, Lou! Actually, three of our SMILES songs were written by the 2017 VOICES chair, biology professor Greg Crowther. Because of how common statistics enrollment (one million students each year) and statistics anxiety is, we wanted to start with a collection of songs that did a good job spanning the topics of introductory statistics.  Because that content domain overlaps with many disciplines in the natural and social sciences, we expect a variety of instructors to start using our collection and one way we hope to measure impact is by following the number of instructors seeking login credentials to the SMILES site and watch the analytics on the site data to see how many sessions are launched, how many songs are used, and how those change over time. We feel SMILES are contagious and our approach certainly can be used to write interactive songs in those other content areas (with the statistics parts of those areas being a natural bridge to introduce the SMILES approach), and we're happy to share our process (via workshops, articles, etc.) with interested folks to support that. One thing we'll discuss at our VOICES meeting next month in Denver is what it would take to compile and consolidate such song-in-STEM resources, as well as STEM song databases (e.g., CAUSE and Crowther's databases).

Thanks Lou.  To add a couple points to Larry's response...

Anytime you get students being more active and the instructors doing less lecturing, you are bound to have more successful learning.  So I think the idea of making songs interactive is generally very favorably received in the VOICES community.  Of course, there are a variety of opinions about the best way to do that so we will probably see more adaptations than adoptions of our work.

Also, we did write a paper (in review) on the process involved in creating the  songs so that others can hit the ground running if they want to try out the general idea in their discipline.  

Dennis and Larry, Thanks for the updates on the project and I look forward to seeing the paper on the process of song creation - a topic about which a lot of people are interested!

Thanks Alejandra.  Let us know if you want instructor log-in credentials to the SMILES materials so you can see the assessment items we use in our randomized experiments to gauge their effectiveness. (the link for that is on the project homepage).

Interesting point Courtney.  We have a large number of English as a Second Language students at Georgia State University where we've done a good deal of our field testing.  Because of that, we made a special effort to use appropriate vocabulary in our student prompts to avoid ambiguities for that group.  

We have also wondered about cultural effects in our music choices and, based on feedback from our advisory board, have been trying recently to broaden the playlist to provide more music genres.

To add to Dennis' thoughts: While we haven't had a chance to get results on this for Project SMILES, my overall work on ELL/language issues in math/statistics education gives me several reasons to believe our interactive songs could help ELLs: First, lyrics have more direct, conversational sentence structure (e.g., minimal use of clauses or inverted order) and have less jargon than textbook prose. Students can pause, rewind, and replay songs (which are already slower than speech in conveying content). The interactive interface often provides scaffolding for those new to English and/or to the US (e.g., "Hypothesis on Trial"). And our song "A Fitting Conclusion" directly adapts the ELL teaching tool of sentence frames (e.g., p. 6 of this 2009 article in Statistics Education Research Journal).  I'm looking forward to speaking about this at the 2018 national conference of TODOS - Mathematics for ALL next month!

hi Dave -- great meeting you at last month's inspiring NCTM conference (I'm the one who inquired about restoring the archive of ON-Math issues to NCTM's website). When I presented our project at the conference, it was interesting how attendees spanned all grade bands.  I agree that the educational uses of song that most people are familiar with involve recall of K-8 material (e.g., multiplication facts). I'd written about more advanced content and roles beyond recall (e.g., in JMA and JME), but it's been a whole new level how we've put together this library of songs that build conceptual reasoning across the introductory statistics course. We're beginning to analyze data from randomized experiments, and look forward to reporting results soon. Given how many students struggle to form a positive identity as doers of mathematics/statistics, and given how virtually everyone loves engaging with music, your last question is quite intriguing and promising!  

Thanks Katie. The previous work that motivated this project showed that songs seemed to help student learning more than cartoons and we speculated that it was the interactivity of songs that powered their effect. You can find our main report on that project at https://tandfonline.com/doi/pdf/10.1080/10691898.2016.1190190?needAccess=true. We also have a web page with references to other work by our group at http://www.math.utep.edu/Faculty/lesser/Fun.html. Thus, we wanted to try and make songs even more interactive by getting students involved in creating the examples driving the lyrics.

We completed a pilot test at a two-year college and at a four-year research institution. The results of survey questions, described in our VOICES poster (available at https://www.causeweb.org/voices/2017/poster/8 at the 2:45 mark), showed that students considered the songs a good way to engage students in learning statistics, relevant to learning statistical topics, and a good tool in helping relieve student anxiety.

We did an experiment in the Fall 2017 semester to compare students randomly assigned to the SMILES intervention with students randomly assigned to readings on the topics (the concepts and examples in the readings were essentially the same). We then put assessment items on those topics in the final exams for those courses. We also measured student anxiety toward statistics using a nationally normed instrument. Thus, we should be able to see if the SMILES approach improves learning and/or reduces anxiety.

Thank you, Katie!  To add to John's post, other initial inspirations/precedents for writing "interactive songs" include: an icebreaker "script" song in a Mathematics Teacher article, fill-in-the-blank songs by Dane Camp, and the Mad Libs template game (which has been used in teaching statistics). And in the music world beyond education, I'm also inspired by a "choose your own adventure" song video by Roy Zimmerman, and by seeing singer-songwriters (e.g., Peter Himmelman or David Wilcox) create a song on the spot that incorporates information about a stranger from the audience. And, along the way, our process has been informed by great feedback from workshops, focus groups of instructors, and our Advisory Board.

Kristina, thank you for watching our video and posting your comment.

We hope our resources help you develop ways for your museum visitors explore statistical ideas.