/-/https/www.sciencebuddies.org/cdn/Files/21464/5/sb-logo.png){kind=logo}
/-/https/www.sciencebuddies.org/cdn/Files/21464/5/sb-logo.png){kind=logo}
I would like to conduct an experiment that answers the question, "what makes some chords sound better than others?" I plan on collecting data by recording different chords on my guitar and analyzing the waveforms in Audacity to find any patterns and explanations to help answer my question. Is this a good procedure? What suggestions can you make for me?
I also have a question about my independent and dependent variables as well. (1) Could my independent variable be the different shapes/forms of a particular chord like, for example, A-flat 13 (Ab13)? There are different forms of the Ab13 chord--some forms use certain notes that other forms don't, but it's still an Ab13 chord. Does this qualify as a "fair test?" (2) What would dependent variable be? The patterns I find in the waveforms of the chords, right?
If you need me to explain anything above, I will elaborate upon request. Thank you!
Help with Planning a Music-related Project
Moderators: kgudger, bfinio, MadelineB, Moderators
-
deleted-200508
- Posts: 1
- Joined: Fri Feb 28, 2014 5:35 pm
- Occupation: Student: 11th grade
- Project Question: http://www.sciencebuddies.org/science-f ... ml#summary
- Project Due Date: 1 April 2014
- Project Status: I am just starting
-
deleted-137857
- Posts: 2
- Joined: Thu Mar 13, 2014 10:50 pm
- Occupation: Student: 10th grade
- Project Question: Trying to learn and help others along the way.
- Project Due Date: n/a
- Project Status: Not applicable
Re: Help with Planning a Music-related Project
Hi seldomseenkid,
From what I currently understand of your project, you want to do something that tries to quantify what makes a certain chord sound "better" than other chords. Now, when presented with your question ("What makes some chords sound better than others?"), I came up with a fairly simple addition to what you already have in mind: survey people. Basically, if you're interested in analyzing, say, ten chords, you can survey a group of people (the larger, the better) by playing all ten chords (although, I'd suggest finding a software that can deliver the chords consistently to reduce instances of human error) and asking them to rank the chords by preference. You assign each rank a numerical value (1-10 in this case) and, after you've surveyed a good number of people, rank the chords in terms of score (the sum of the numerical ranks assigned by each person). Take the top half (in this case, the five most popular chords) and analyze the waveforms in Audacity like you said, and then do the same with the bottom half. If you can establish a more general reason for why the top five chords were the top five chords and why the bottom five chords were the least liked, you can go further by analyzing the subtler differences between the first five chords to see what subtle differences made people like chord number 1 vs chord number 2. To clarify, let's consider you're only interested in analyzing four chords (for our sake, we'll call them A, B, C, and D). You have a five people listen to the four chords and you tell them to give their favorite chord a 3, the next favorite a 2, the next favorite a 1, and the least favorite a 0. Say our five people say something like this:
Person 1-
Chord A=3
Chord D=2
Chord C=1
Chord B=0
Person 2-
Chord B=3
Chord A=2
Chord D=1
Chord C=0
Person 3-
Chord D=3
Chord A=2
Chord B=1
Chord C=0
Person 4-
Chord A=3
Chord C=2
Chord D=1
Chord B=0
Person 5-
Chord C=3
Chord A=2
Chord B=1
Chord D=0
Total-
Chord A: 12
Chord B: 5
Chord C: 6
Chord D: 7
As you can see, Chord A is the most popular, followed by D, C, and finally, B. So, if you study the waveforms of Chord A and D and compare them to the waveforms of C and B, you may be able to find some general differences between them; as in, chords A and D may have a characteristic in their waveform that C and B do not have. Once you've established that, you can compare the waveforms of A and D to see if there are any differences that make A more popular than D. Repeat this procedure with the other chords to see the more subtle differences. If you need me to clarify further, I'd be more than happy to do so.
From what I currently understand of your project, you want to do something that tries to quantify what makes a certain chord sound "better" than other chords. Now, when presented with your question ("What makes some chords sound better than others?"), I came up with a fairly simple addition to what you already have in mind: survey people. Basically, if you're interested in analyzing, say, ten chords, you can survey a group of people (the larger, the better) by playing all ten chords (although, I'd suggest finding a software that can deliver the chords consistently to reduce instances of human error) and asking them to rank the chords by preference. You assign each rank a numerical value (1-10 in this case) and, after you've surveyed a good number of people, rank the chords in terms of score (the sum of the numerical ranks assigned by each person). Take the top half (in this case, the five most popular chords) and analyze the waveforms in Audacity like you said, and then do the same with the bottom half. If you can establish a more general reason for why the top five chords were the top five chords and why the bottom five chords were the least liked, you can go further by analyzing the subtler differences between the first five chords to see what subtle differences made people like chord number 1 vs chord number 2. To clarify, let's consider you're only interested in analyzing four chords (for our sake, we'll call them A, B, C, and D). You have a five people listen to the four chords and you tell them to give their favorite chord a 3, the next favorite a 2, the next favorite a 1, and the least favorite a 0. Say our five people say something like this:
Person 1-
Chord A=3
Chord D=2
Chord C=1
Chord B=0
Person 2-
Chord B=3
Chord A=2
Chord D=1
Chord C=0
Person 3-
Chord D=3
Chord A=2
Chord B=1
Chord C=0
Person 4-
Chord A=3
Chord C=2
Chord D=1
Chord B=0
Person 5-
Chord C=3
Chord A=2
Chord B=1
Chord D=0
Total-
Chord A: 12
Chord B: 5
Chord C: 6
Chord D: 7
As you can see, Chord A is the most popular, followed by D, C, and finally, B. So, if you study the waveforms of Chord A and D and compare them to the waveforms of C and B, you may be able to find some general differences between them; as in, chords A and D may have a characteristic in their waveform that C and B do not have. Once you've established that, you can compare the waveforms of A and D to see if there are any differences that make A more popular than D. Repeat this procedure with the other chords to see the more subtle differences. If you need me to clarify further, I'd be more than happy to do so.