btw, I think the size of the Lego brick will factor out of the analysis because compressive strength is in units along the lines of "pounds per square inch" so if you have more square inches to distribute the load, the load will be higher...while the "psi" compressive strength would remain the same...excuse my lack of metric units which totally suck anyway.
Perhaps consider testing both the 4X2 and 2X2, record both the load at failure and the area it was applied to; then see if the "load per area" remains about the same.
Since the venerable Tinius Olsen company is right down the street from where I work, I looked at their site knowing that they're a bunch of whackos over there, and wouldn't you know it ...see the caption to Figure 18 in this brochure:
http://www.parameters.com.au/library...ngMachines.pdf
Perhaps we should ask them if they can share their data as a check of telekinetic's work.