Here is an excerpt from an example discussion for Experiment III, Formula Weight of an Ionic Compound:

The MSO4 sample, weighing 0.213g, may have been too small. In order to isolate BaSO4, the procedure called for many preparatory steps involving filtration and precipitation. Each step in the procedure allowed for a certain margin of error. All of the BaSO4 may not have come out of solution during the crystallization, and when filtering the precipitate that did come out, some may have redissolved in the distilled water used to filter it, even though it was cold. Furthermore, the BaSO4 that was recovered may have contained impurities because sometimes during the crystallization process, large crystals trap surrounding residues. The filtration process required us to constantly stir our precipitate in solution which may have disrupted the equilibrium between the crystals and the solution and redissolved some of them. If this happened, we may have lost weight through the filter. We could not rinse all of the precipitate out of the crystallization beaker also, which means we lost weight at that point as well. All of these errors constitute a loss in BaSO4 mass which means a loss in the %SO4 calculation. This loss in turn would yield a higher value for the atomic weight of the metal because the atomic weight of the metal is inversely proportional to the moles of SO4 in a 100g sample of our unknown salt. These errors may explain why the possible metals that I came up with are of such high atomic weights.

I expected the actual metal to be one that I have heard of, one that is commonly found in college experiments, and one that is safe and light in weight. However, according to my results, the best possible molecular formula for the unknown salt turns out to be CuSO4.7H2O which is not a plausible compound. CuSO4 is a white powdery salt when it is anhydrous...but CuSO4.5H2O is a blue compound. Furthermore, the hydrous form of CuSO4 absorbs two less water molecules than the compound I came up with.