Hypothesis help for "MP3 Sqeeze" project.

[Bigal369]

I am having trouble coming up with a research based hypothesis for the "MP3 Squeeze: How Much Compression is Too Much?" project. I am not sure how I should put it in "If...then" form. Also, I don't know exactly how to make a hypothesis for this project. I am in need of assistance. My project is due on December 14, 2009. Please help me.

[deleted-71709]

Your hypothesis might be stated in this way, "High compression of digital audio files will generate audio that is less pleasing to the human ear". Or, using your "If...then" approach, "If CD quality audio files are highly compressed, then the quality of the audio played back from those files is less pleasing to the human ear."

These should work.

In reviewing the Science Buddies project description, there is one thing I would suggest. Because the output of your experiment is an opinion of the quality of the sound, I would suggest you collect the opinion of more than one person. Usually friends at school are quite willing to help with an experiment like this.

Good luck, and have fun.

Ed Neu
Buffalo, MN

[deleted-71588]

I am having trouble coming up with a research based hypothesis

I'm not surprised. It will be hard to come up with something that doesn't have subjective aspects. Subjective aspects usually means you will have to have multiple people (subjects) evaluate several MP3 compression ratios of the same material and use statistics to determine your results.

MP3 by design is a lossy compression scheme unless the source is already compressed but this is not the only factor that is involved. The "playback" system can also be lossy. For example, if you send an audio signal through a phone line, frequencies above 3 KHz are removed. Statistically significant research was done that identified the ability to recognize speach required frequencies between 200 Hz and 2700 Hz. Other statistically significant research on the response of a young adult human ear indicates frequencies between 20 and 15,000 Hz are typical. Top end audiophile type equipment tries to reproduce 16 to 25,000 Hz. Note, these kinds of phone and music results come from large sample sizes and are used by engineers when designing equipment. In most cases, the speaker/headphone part of the system will limit the frequency response the most. Physically small speakers typically can't reproduce low frequencies well and large speakers typically can't reproduce high frequencies.

Go back and re-read the project idea considering the subjective aspects I've mentioned and see if you can come up with a testable hypothesis.