The increased popularity of SEM has been accompanied by two changes in the statistical analysis of research data. First is that the complexity of the models and methods being used has increased dramatically. In statistical modeling, problems that were previously regarded as impossibly complex have become regarded as tractable. This trend is partly driven by Moore's Law, which states that the complexity of computer circuits (i.e., computing power) doubles approximately every 18-24 months (Moore, 1965). Both methodological and substantive researchers have sought to exploit developments in computer architecture with statistical methods to improve the quality of their scientific output.
