Science

Visualizing oneway ANOVA

I spent a little while over this winter break expanding upon how I might use Shiny to help demonstrate and familiarize students with statistical concepts in music education research methods courses. In general, it seems to me that clear explanations are needed, practice examples are good, but “live” manipulable demos are really, really great.

I built an app that allows students to visualize and manipulate a hypothetical experimental scenario so that they can see how the oneway ANOVA procedure partitions variation into “group (i.e., model)” and “residual” sources or more colloquially, “between” and “within” group sources. The plot at the top shows stripcharts of three groups of hypothetical experimental participants in blue, and a single stripchart of all of the hypothetical participants at once in red. Below the strip chart are controls for manipulating the data.

Students can play with the means of three groups and watch the between sums of squares go up and down. When students manipulate the standard deviations of the groups, they’ll see the within sums of squares go up and down. The numbers corresponding to the between, within, and total sums of squares appear reactively in the sub-title of the plot on top.

Below the controls are (a) a corresponding ANOVA table that updates reactively with each change the student makes and (b) a density plot to depict the overlap among the groups a bit more clearly.

This app is similar in design to “this ANOVA handout” I use in my courses.

Click this link or the image below to access the app:

https://petemiksza.shinyapps.io/Visualizing_oneway_ANOVA/

http://js-156-26.jetstream-cloud.org:3838/oneway_ANOVA/

Let me know what you think if you have a moment or have any suggestions if you’d like to use this or something like it and I can be of help.

onewayANOVAapp

PS…

I’ve updated the previous “standard error of the mean” app I built to be cleaner and easier to use (see the previous post)

Advertisements

My First Data Visualization Toy

I made a toy!

I made a fun little web app for visualizing some interesting things about sampling distributions and the standard error of a mean (pictured below). The app allows you to change parameters with the sliders on the left to see how sample size and standard deviation affect the standard error of a mean — and how sampling distributions of the mean approach normality as the number of samples drawn from a population increases.

You can play with it at this link (or click the picture): 

http://js-156-26.jetstream-cloud.org:3838/standard_error/

semAPP

I built this app in R

I’ve recently been enjoying learning about the possibilities that the free, open-source statistical computing and graphics language “R” offers. It’s an amazingly flexible platform for organizing, visualizing, and analyzing data. Working in R requires acquiring some fluency with basic programming language. However, there is an astounding amount of free learning resources available on the web in the form of websites, books, blogs, tutorials, etc. There are also great web-based courses (https://www.datacamp.com/) that often have free trials.

The program can be downloaded here:

https://www.r-project.org/

Many recommend also using R Studio, a platform to help organize workflow when using R. The free version of R-Studio is what I use, and I’m fairly sure it has all that most of the folks in music education would likely every need.

https://www.rstudio.com/

More to the point… …R Studio has an integrated web app builder called “shiny“. Shiny makes building data visualization tools and web-based dashboards for exploring data fairly straight-forward. The apps that are produced with shiny are built in roughly the same type of code that is used to run analyses and make plots in R in general.

https://www.rstudio.com/products/shiny/

Why R?

Music ed research folks may wonder why use R instead of say, SPSS, or another commercial statistical package? Well, that could probably be a rather lengthy discussion of pros and cons. However, here’s a short list of some simple things that R is nice for:

  • It’s free and open-source with an immense community of users and developers which keeps it well-documented and up-to-date, as well as relevant to the needs of contemporary analysts.
  • The R community is very helpful. You could google almost any kind of problem you may be having and would be likely to find a few forums where people offer solutions.
  • It’s modular such that there is a package or more than a few packages to do any kind of data wrangling or analyses or plotting you’d ever want to do. For example, there was a point when I would toggle back and forth between SPSS, Stata, and Lisrel depending on what sorts of analyses I was doing. Now I can do it all in R.
  • The graphing capabilities are very impressive. It’s fairly easy to get nice-looking, presentation and/or publication ready figures and you can customize any aspect of a graph.
  • Organizing the coding for analyses with code in scripts allows you to create a very clear reproducible record of all of the work you do to arrive at your results in any given project. This saves lots of time in the long run and is good scientific practice in general.

 

 

The IU JSOM Music and Mind Lab: A Year in Review

Last year included the first full academic year of Music and Mind Lab meetings and activities at the IU Jacobs School of Music. I thought I’d post a quick note about some of the fun and exciting things we were able to do together as we begin looking forward to another productive year.

IU - MusicAndMind-Banner.jpg

Some quick background (visit the MaML Website)

The lab was overseen last year by co-directors Pete Miksza, Frank Diaz (Dept. of Music Education) and Daphne Tan (Dept. of Music Theory). Dr. Tan will be shifting to a role of “collaborator” this year as she transitions to a new position at the University of Toronto – she will be sorely missed! The student lab members include undergraduates and graduate students from a variety of disciplines and academic specializations: music education, music theory, musicology, music performance, cognitive science, psychological and brain sciences, and telecommunications. Our goal is to produce original research that will contribute to a general understanding of the role of music in the human condition.

Recent events

Guest speakers and presentations featured heavily in our activities this past year along with sessions devoted to faculty and students’ research interests. We were fortunate to have some truly brilliant people join us and share their research. Last year’s special topics and guests included the following:

  • Musical improvisation as a way of knowing
    • Andrew Goldman, Presidential Scholar in Society and Neuroscience, Columbia University
  • Working as a lab in the cognitive humanities
    • Fritz Breithaupt, Professor of Germanic Studies, Indiana University, Bloomington
  • Rhythm and movement
    • Justin London, Andrew W. Mellon Professor of Music, Cognitive Science, and the Humanities, Carleton College
  • Music, empathy, and cultural understanding
    • Eric Clarke, Heather Professor of Music; Professorial Fellow, Wadham College, University of Oxford, UK
  • Music, trauma, and the Polyvagal Theory
    • Jacek Kolack, Postdoctoral Research Fellow, Kinsey Institute, Indiana University, Bloomington
  • The neuroscience of musical skill learning
    • Anna Kalinovsky, Assistant Research Scientist, Gill Center for Bimolecular Science, Indiana University, Bloomington
    • Grigory Kalinovsky, Professor of Music (Violin), Jacobs School of Music, Indiana University, Bloomington

Ongoing projects

Although just getting started, our lab group has been quite productive in generating projects with interdisciplinary connections throughout IU and completing research that has found its way into the world as conference presentations and journal articles. Our most recent project involves a collaboration between Drs. Tan, Diaz, and I on the topic of musical communication. We are investigating the nature of expressive vocal performance. Broadly speaking, we are studying the ways singers prepare and produce performances to be evocative of specific emotive intentions. We are also interested in how inducing a mindful state will impact their singing and, ultimately, how these performances will be received by listeners. We’ve collected a good deal of data and are excited about the potential for this project going forward.

How students are involved

Students can participate as investigators or contributors. Investigators typically come to the lab with some prior experience in empirical research and are expected to co-design, propose, and lead projects. Contributors primarily serve support roles. They are expected to participate in weekly meetings and discussions, and help to manage projects and collect data. Through their participation, contributors can gain the experience necessary to be investigators in future projects.

Coda

All in all, I’m happy to say the Music and Mind Lab provides an exciting intellectual space for those at Indiana University who are curious about intersections between their musical and scientific interests. I’m hopeful that this group will continue to grow and look forward to a productive ’17-’18 school year!

Take some time to learn about the Jacobs School of Music if you’re interested in joining us.

Music & Mind Emblem CROPPED

2016 IMEA Presentation

Hello!

Click below for slides from my presentation today at the Indiana Music Educator’s Professional Development Conference…

The science of music performance skill acquisition:
Planning, executing, and reflecting for achievement

Presentation slides

All the best,

Pete Miksza

A SMiLE-ing experiment

I wanted to write a quick post to share my excitement about a new research group starting up at IU called: SMiLE: Science of Music in Learning and Experience.

The purpose of the group is to cultivate conversations about music research among students and faculty across departments in the school of music and across units on campus. So far, interest in the group has been expressed by grad students and faculty from… …Music (education, theory, conducting), Cognitive Science, Psychological and Brain Sciences, Telecommunications, Speech and Hearing, Informatics, Computer Science, Linguistics, and Second Language Studies. This image is a wordle of the groups’ collective interests from our first meeting.

Topics of interests

Our first two meetings were interdisciplinary, collegial, energetic, and overall – lots of fun. Many thanks to the colleagues and friends I spoke to who encouraged doing something like this and gave such great advice about how to do it. Many thanks too, to the wonderful colleagues and students taking time out of their busy days to join in the fun!

It’s sure to be a fun adventure and I’m excited for what the Spring may bring!

Research reads to keep you warm on cold winter nights

Because, really, what’s more comforting in the middle of winter than curling up with a great book on quantitative research design and/or statistical analysis!

🙂

This semester I have the pleasure of teaching a course on quantitative research methods for music education. In preparation for this class, I’ve been looking at all kinds of resources that could be helpful for the students (and me) to dig into compelling issues of research design and analysis. A couple of standouts from the pile of books I was wading through are listed below. In addition to being clearly written and approachable in style, each book does a great job elaborating on issues related to quantitative research that are often difficult to digest. These books aren’t designed to be suitable as a text for a music education research methods course, but, they’re certainly excellent supplements.

photo

(snowflake courtesy of Lucy Miksza, 5 yrs old)

Stanovich, K. E. (2001). How to think straight about psychology (6th ed.). Needham Heights, MA: Allyn & Bacon.

  • This book is a terrific, down-to-Earth read about some of the most basic characteristics of scientific inquiry. I particularly enjoy the discussions of scientific inquiry as a converging process, the importance of falsification, and the challenges inherent in probabilistic thinking. Being focused on the social science of psychology, it comes across as a good introduction to issues of scientific activity that comes across in a way that I think is relevant to many of the types of questions that music ed researchers may be interested in.

Abelson, R. (1995). Statistics as principled argument. Hillsdale, NJ: Erlbaum Associates.

  • This is a humorous and plain-spoken collection of wisdom for those who are writing about statistical findings. The first chapter, “making claims with statistics,” raises a host of simple, yet important considerations for stats folk. All of the chapters, though, will be helpful – especially when thinking about developing a writing style.

Jaeger, R. M. (1990). Statistics as a spectators sport. Newbury Park, CA: Sage Publications.

  • I’ve relied on this book for one reason or another many times since getting bitten by the music ed research bug. This book lays out basic and intermediate statistical topics in an easy-to-grasp, conceptual manner. Jaeger’s explanations could be a great help for those who find that math and formulas seem to get in the way of understanding how statistical analysis techniques could serve music ed researchers. Or, if you’re looking for a book that ties together some loose ends and fills conceptual gaps – this could really help.