Hi,
I think you have two options:
1)if it's just for learning purposes and you don't really care about the accuracy(not intended for silicon manufacturing) you can take a look at MOSIS, they provide SPICE models from their processes(I suppose it's what it's used in the webcasts), there are several processes to choose:
http://www.mosis.com/Technical/Testdata/menu-testdata_mep.html.
2)A rough approximation with the model that you have: formulate the standard square-law equation and isolate the variables that you want to estimate-with some assumptions because you will find Vth and lambda all together-, perform DC simulations and calculate the values. After hand calculation you can see how much divergence exist between hand calculation and the square law model. As a reference I remember:
http://personalpages.to.infn.it/%7Ecobanogl/lowlevelstuff/tutparext and
http://personalpages.to.infn.it/%7Ecobanogl/lowlevelstuff/tutparext2.
As Geoffrey said modern processes doesn't fit the standard equation, so this results(1 or 2) are only approximations for quick math, final circuits must be tunned using simulations.
Hope it helps,
edit:MOSIS don't provide LAMBDA, my memory it's starting to fail...