Quality v. Quantity

It’s not how much; it’s how. ~Robert A. Duke, Amy L. Simmons, Carla D. Cash; Texas music education researchers

You’ll never make a mistake if you never make a mistake. ~Julius Baker

—–

Want to learn more about the best ways to practice? Get an e-mail with a discount code when The Practice of Practice is published (June, 2014). To learn more about the book, check out a sample from The Practice of Practice.

When I was a kid in middle school, a practice sheet was taped to the fridge. My music teacher required a weekly practice report and to get a good grade, you had to practice at least a couple hours a week, but it didn’t matter what you did, you just had to log the time (this resulted in some creative time-wasting on my part, some of which was actually helpful). Some research shows that the amount of time doesn’t really matter, although it does matter a little since if you spend zero hours doing something, you’re not going to get better at all. But it turns out that the number of hours practiced doesn’t really matter, it’s all about the quality of your practice. What you do is important, but not how much you do. Duh, right?

This seems like a no-brainer issue, but researchers are notoriously skeptical about common-sense issues. They want to know for sure whether things are true. That’s one of the reasons behind a study by Duke, Simmons, & Cash (2009), titled It’s not how much; it’s how: Characteristics of practice behavior and retention of performance skills. These researchers had 17 graduate and advanced undergraduate piano players practice a 3-measure excerpt of Shostakovich’s Concerto No. 1 for Piano, Trumpet and String Orchestra (here’s a clip of Shostakovich himself playing part of it). Here’s the excerpt:

Shastakovich Piano Concerto #1 excerpt — Shostakovich Piano Concerto #1 excerpt

The piece was chosen because it’s difficult to sight-read at tempo and is short enough (and challenging enough) to be learned in one practice session. Here’s what was required of the musicians:

players had 2 minutes to warm up, then were given the excerpt, a metronome, and a pencil
they had as much time as they needed to practice until they were confident they could play it accurately at 120 beats per minute during a test the following day.
The next day, they had 2 minutes to warm up (without playing any part of the excerpt), then played the excerpt.

And here’s what the researchers analyzed:

Everything was recorded to digital video; and the piano practice was recorded from the MIDI keyboards.
during the practice the researchers measured:

1. total practice time
2. number of performance trials
3. number of complete performance trials
4. number of correct performance trials (no errors or hesitation)
5. number of near-correct performance trials
6. sum of correct + near-correct trials
7. number of incorrect trials
8. percentage of complete trials correct
9. percentage of complete trials that were correct and near-correct
10. percentage of all trials that were correct.

The test the following day was 15 attempts. They measured

1. Number of correct trials
2. number of near-correct trials (correctness measured only in terms of pitch and rhythm accuracy, not emotional content, etc.)
3. sum of #1 and #2
4. Judges agreed 96% of the time on the correctness of the trials (this is a great consensus, or inter-rater reliability)

Here’s what they found:

A few things didn’t seem to affect how well the players did (no statistical significance): practice time; total number of practice trials, and number of complete practice trials.
What did separate the better pianists (statistically significant), was: the percentage of all performance trials that were correct; the percentage of complete performance trials that were correct; and the number of trials performed incorrectly during practice (this was a negative relationship, or the less mistakes the better the player ranked).

So what does this mean? In a nutshell, you have to practice slow enough to get things right as soon as possible. Playing anything incorrectly ever is teaching your motor neurons to play incorrectly. They don’t know any better and this is a great example of the GIGO principle: Garbage In, Garbage Out. What separated the top-ranked players from the others was how they treated errors when they occurred (p. 318):

The precise location and source of each error was identified accurately, rehearsed, and corrected.
Tempo of individual performance trials was varied systematically; logically understandable changes in tempo occurred between trials (slowed down enough; didn’t speed up too much).
Target passages were repeated until the error was corrected and the passage was stabilized, as evidenced by the error’s absence in subsequent trials.

Everybody makes mistakes. It’s inevitable! But how you handle those mistakes makes all the difference, as this and other research has shown. This raises a couple issues, the first being that you have to be able to perceive that you’ve made an error, something that is more difficult for beginners especially. Another issue is that you must know that errors have to be corrected immediately and not saved for later, or worse yet, made again the next time you go through a piece. Now that quote at the top of this post by flute pedagogue Julius Baker should make more sense.

But finally, the issue of who can use this information is one I always bring up. If you’re learning an improvised solo by ear, it’s nearly impossible to learn it without LOTS of trial and error. I’m always going back over a particular passage I’m trying to learn by ear, listening hard and comparing what I’m doing with the recording. Eventually, I get it right, but not immediately, and going slowly is only an option if I run the tune through Audacity or some similar program. So what we need is a good model on how this is best done. I’m sure interested in that, so I guess I’ll add it to my list of research projects.

Anyway, have fun, and good luck with your practice!

Here’s a YouTube copy of Shostakovich’s Piano Concerto no.1 in C minor, Op.35. Shostakovich’s heirs play it: his son Maxim conducts, and his grandson Dimitri Jr. plays piano. James Thompson plays the trumpet part. (if you want to own it: CD, or mp3)

I.Allegro moderato