Grading dictations and sight-singing

George Lam · August 2021

Hello everyone,

I'm wondering if I can get your take on how you assess student dictations and sight-singing, and in particular how to translate this to a grade? For example:

Given a short four-bar, single-voice melody that's heard three times, how do you grade this melodic dictation? Do you use a letter grade, a rubric, or points (e.g. out of 100)? If so, how do you determine the grade? How do students earn partial credit?

Given a short sight-singing example where students have to conduct while singing in solfège, how do you grade this? Similarly, do you use a letter grade / rubric / points or something else? How do students earn partial credit?

I think I've tried a lot of different ways to translate student performance to some kind of measurable value (for the purposes of calculating a grade, and to provide a data point for students that reflects their performance), but am trying to see if there are better practices out there.

Thank you!

George

Eric Knechtges · August 2021

For melodic dictation:

Some number of points per pitch (usually double-weighted if it's a chromatic pitch) -- I *may* give partial credit if they get a diatonic interval off for an extended period of time, but often not, since that indicates that they were not thinking about tonal function to reorient themselves.

Some number of points per half-measure of rhythm -- only one deduction if there is a consistent rhythmic displacement of a beat for a few measures, for instance.

At lower levels of aural skills, these come out to about 50/50 pitch and rhythm -- at upper levels, usually between 60/40 and 75/25.

Harmonic dictation is similar -- half-point for soprano, half-point for bass, and a number of points per Roman Numeral. Extra weight for chromatic pitches, chromatic chords, modulations, etc.

Sight-singing is tougher, because there are so many variations on how things can go awry. I experimented with rubrics like I used for sight-singing, but these led to overall scores that just didn't "feel" right.

We ended up moving to a "jury" format for end-of-the-semester final exams, where a panel of two or three aural skills instructors listen to a number of exercises, prepared and at-sight, and assign a letter grade (or a percentage, if it's less than a D). These are translated to percentages and averaged. It's a bit time-consuming (I had a huge spreadsheet set up that I could type numbers into, so I could also get metadata about class averages, average score per exercise, etc.), but it gives a bit more validity to the final grade.

Gary Karpinski · August 2021

George—

I addressed the grading of melodic dictation in my book Aural Skills Acquisition (on pp. 103-110), surveying a few published approaches (e.g., the AP Exam in Music Theory) and making some recomendations for ways to develop a standardized grading rubric. Without copying and pasting all that here, I'd simply encourage you to consider the distinctions between assessment and evaluation, and to think about ways to first assess students' work and offer meaningful feedback, then — as a separate part of the process — to evaluate that work (as a percentage, number, letter grade, or other indication). How we translate assessment into evaluation is a value system in itself, and can be developed and tweaked with full transparency for instructors and students alike.

—Gary

George Lam · August 2021

Thanks Eric and Gary! These are great ideas. Eric, that's great to hear about your take on grading sight-singing with a jury; I think I'll implement something similar at our (small) department, and also see if I can co-create a rubric together with the students.

Gary, your book has been a huge help! Thanks for the tip on assessment vs. evaluation. I'm structuring the semester with a lot of formative assessment, but it's always been tricky to translate performance to a grade in such a way that would be useful (and encouraging) for students.

Timothy Chenette · August 2021

Hello, George!

In addition to Gary's discussion in ASA, Jeffrey Gillespie has an article called "Melodic Dictation Scoring Methods: An Exploratory Study" in the Journal of Music Theory Pedagogy Vol. 15 (2001), while Stan Kleppinger's 2017 article in Indiana Theory Review, "Practical and Philosophical Reflections Regarding Aural Skills Assessment," has some important critiques of these traditional approaches.

I hope you'll forgive my climbing up on my soapbox for a minute, but I often find approaches to assessment in aural skills classes to be insufficiently goal-directed. In my own teaching, I always make sure I have a primary goal for each exercise: for example, perhaps a sight-singing (or prepared singing) exercise is designed for students to apply and perfect an approach to (say) across-the-beat triplets, or perhaps a dictation is intended for students to demonstrate their ability to determine tonic. Whatever my goal is, I make sure I have actually been giving instruction in that area recently, and if a student meets that goal, then regardless of what other errors they make, they receive a passing grade. I usually give either 0/10 (didn't meet the goal regardless of how good the rest was, please try again), 8/10 (met the goal but still room for improvement beyond that), or 10/10 (excellent).

I think the point-per-pitch/half-measure approach Eric describes is common. I have in the past used something like it, and have decided there are three aspects that don't seem right to me. First, it doesn't account for all the global skills involved (finding tonic, determining meter, etc.), and these are more important to me than individual errors. Second, it means students are graded on a standard of perfection right from the beginning. I imagine this as teaching a class on how to run a 10k, and all you do is ask the students to run 10ks and grade them each time on how close they came to completing it--some students will do great right from the beginning and others will do poorly, but those grades will have more to do with their preparation for the class than their learning/effort. Third, it focuses on little details (individual pitches, rhythmic cells) which can make students so terrified about these that they lose all sense of grouping and musicality (and, ironically, they often get worse at the details too). I think more holistic grading will help students focus on the specific skills we want them to acquire, while also allowing them to consider overall musicality and context, as one ideally should when sight reading.

I do find the "jury" format intriguing, because that could easily focus more on the overall musicality of sight singing. I like the idea of using that as a culminating exercise.

If you're interested in more soapbox thoughts, I recently wrote a blog post that touches on assessment in aural skills. It focuses on philosophy/approach and doesn't give practical tips that can be put into practice right away, but may still be helpful.

All best,

Tim

George Lam · August 2021

Thanks for all the thoughtful insights, Tim! And also for the resources. I think our need (mandate?) to give grades is getting increasingly difficult to square with student-centered pedagogy, but as you pointed out, there are some good middle-grounds that can help students achieve more, and to create a more constructive learning environment.

I'm trying out a flipped learning model for our musicianship classes this year, mostly because our schedule only allows one weekly 50-minute in-person session, so most of the time will be spent on working together in small sections on dictations and sight-singing. I like the ideas that you proposed, and am thinking about how we can implement different types of grading - especially to involve students in peer-grading contexts that can help everyone.

Thank you again!

George

Hello!

Categories

In this Discussion

Most Popular This Week

Grading dictations and sight-singing

Comments