Also, since we're using what are really technical terms here, let's use the right ones:
"normalizing" sets the PEAK level to a standard amplitude, often (but not necessarily) the maximum possible level for the signal carrier or transmission channel.
What you probably want to do is set the apparent loudness of all your music to a common level, which requires adjusting the AVERAGE level, not the peak level, and is a much harder task.
There is software that exists to analyze your files and supply a suggested volume change, called ReplayGain, but unless you want to alter the actual sample values and rewrite the files, you need player software that honors the ReplayGain tags to make the adjustment at playback time.
Also keep in mind that if you set each TRACK to the same apparent loudness, you will destroy any intended changes in loudness that exist within an album. That's an artistic/musical decision, and if you ever listen to complete albums you will notice that some tracks are supposed to be quieter and some are supposed to be louder. It's possible to use ReplayGain in an "album" mode, which preserves those differences among the tracks of an album, but in random or shuffled mode you will still have (possibly unwanted) volume changes.