I agree with Willbe's comments, but a little more intuitive and simplistic approach is available.
The easiest way to evaluate panel power is, of course, to measure it directly with a wattmeter as you vary the load with a potentiometer or resistor load bank having sufficient power/current rating. Observe the load current or the load voltage or the load resistance at which the displayed power is maximized as you vary the power resistor.
Although I have been lucky enough to pick up a few wattmeters surplus from a local university, unfortunately very few of us have wattmeters, the passive, analog form of which are instruments with meters that resemble the familiar d'Arsonval (pointer) type analog meter except that the magnetic field is created not from a permanent magnet, but from a fixed coil that carries the load current. The moving coil is energized by the load voltage. The result is that the pointer displacement indicates the PRODUCT of the load voltage and the load current - i.e., the power. The scale will typically be very non-linear. There are also wattmeters that employ active analog or digital circuitry to compute and display the power.
Lacking a wattmeter, you can determine the peak power point by measuring the load voltage and the load current and plotting the product of the two vs. load resistance on a chart. At zero load resistance (short circuit) the current will be maximum, the voltage will be zero, and the product (power) will be zero. At infinite load resistance, the current will be zero, the voltage will be maximum, and the product (power) will again be zero. As you vary the load resistance and plot the product of voltage and current at several points, you will find a maximum in the power output curve somewhere near the mid point of voltage and current. Ideally you would let the panel reach steady state thermal conditions at each new point but getting data at nearly constant solar irradiance requires that you take data quickly.
One way to obtain a first-order correction factor for varying solar irradiance during your measurements would be to plot simultaneous readings of voltage or current from a properly loaded separate solar panel or even a tiny solar cell like you can (or used to be able to) buy at Radio Shack that is held at constant conditions and orientation.
Have fun.
awright