To date there have been no systematic, quantitative investigations of the effect of sample preparation on the matrix-assisted laser desorption/ionization time-of-flight (MALDI) mass spectrometry response for polydisperse systems. To this end, the interrelationships between sample preparation, analyte molecular weight distribution (MWD) and solubility, and signal response were investigated for mixtures of alkylated polycyclic aromatic hydrocarbon (PAH) oligomers, the constituents of petroleum pitch that serve as precursors for advanced carbon materials. These PAH oligomers served as a useful analyte system for study, as their solvent solubilities decrease significantly with each increasing oligomeric unit. Molecular weight standards consisting of relatively pure dimer and trimer cuts of the starting M-50 petroleum pitch were produced using a dense-gas/supercritical extraction (DGE/SCE) technique and were then used to produce oligomeric mixtures of well-defined composition for study. Both traditional, solvent-based and newer, solvent-free sample preparation methods were evaluated, and their effects on both homogeneity and signal response were determined. While solvent-free sample preparation methods produced homogeneous samples and reproducible results regardless of the MWD of the analyte, solvent-based samples that contained more than one oligomeric cut produced non-homogeneous samples and poor reproducibilities. The differing solubilities of dimer, trimer, and tetramer oligomers in a given solvent (e.g., CS(2) or toluene) were found to be the cause of the inhomogeneities observed in solvent-based sample preparation. A quantitative analysis study performed with dimer/trimer mixtures over a wide range of compositions via solvent-free sample preparation indicates that linear, reproducible calibration curves can be generated and used to calculate the molecular composition of unknown dimer/trimer mixtures with confidence.