I recently experimented with Stereophile's Test CD 3, which had three tracks, each one of which provide a series of rising (or falling, in the case of the bass) test tones that sound for about 5 seconds each, before moving to the next track. I was also using a friends old Realistic/Radio Shack analog Sound Level Meter.
The advice provided above - about developing a calibrated table of responses for the meter - is VERY important (Rives, Room EQ).
In my experience the meter significantly under-measures tones below 100 Hz and over 10,000 Hz. If you follow the procedure for setting a test tone level around 1000 Hz, and look at the graph that results form the tests on your speakers, WITHOUT SOME ADJUSTMENT/COMPENSATION FOR THE INEXPENSIVE SOUND LEVEL METER readings, you will be surprised to see a very curved (as opposed to flat) response. Do not throw your speakers away at this point.

The inexpensive sound level meters need to have their measurements adjusted for the inadequacies of the meter. The Rives and other sources can guide you in this.
ALSO - just as important - the decibel level (dB) you read on the meter, depending on where you do your testing and how far you are from the speaker, will include the effects of the room. Some frequencies will be enhanced, some attenuated. Do not worry too much, but recognize that you will can see up to 3 dB variation (or more) from the true reading just related to room acoustics. The goal with this meter is to get a sense of the "response curve" for your speakers and system, and then you can compare the effects of different speakers, shifting speaker position, and adding room acoustic treatments on the frequency response curve. You can also get a "sense" of the response of your speakers, but recognize that an inexpensive meter, in an acoustically reflective room, will not be able to reproduce the results of "scientific" level test equipment.
I recently went through the process of testing the frequency response curves on some new speakers, and I learned more about the meter and testing process than I did about the speakers, initially. The meter IS useful for comparative testing (speakers, acoustic treatment, speaker placement), but is not really all that useful for "absolute" testing of a speaker, IMHO. At least not without a lot of experience and adjustment.
Bill C