Nationally, there is continuing debate about appropriate methods for conducting educational evaluations. The U.S. Department of Education has placed a priority on "scientifically" based evaluation methods and has advocated a "gold standard" of randomized controlled experimentation. The priority suggests that randomized control methods are best, followed by quasi-experimental designs with carefully matched comparison conditions, and then longitudinal designs and regression discontinuity designs. Using only one research method ultimately allows only one system of values to operate within an evaluation. In the case of randomized controlled experiments, the value system is nomothetic as opposed to idiographic and utilitarian as opposed to intuitionist. The nomothetic approach is characterized by procedures and methods designed to discover general laws relying on reproducible facts and outcomes. By contrast, an idiographic view stresses the importance of understanding how the individual creates, modifies, and interprets the world. A utilitarian approach emphasizes the greatest good for the most people as measured by a predetermined outcome, while an intuitionist approach emphasizes value to an individual. This article illustrates the methodological pluralism inherent in STEM educational evaluations. This article first describes the connections between diverse methodological approaches and STEM evaluation. Then a web model for developing scientific understanding of STEM projects is presented. Examples illuminate the strands portrayed by the web. It concludes by summarizing the case for methodological pluralism in evaluation. (Contains 1 figure.)