The role of diet in promoting health and preventing disease is difficult to elucidate due to its complex network of foods and nutrients. Besides total energy intake, dietary composition is probably the most important discriminator within and between populations. Dietary composition is reflected in dietary patterns, which have recently gained popularity. The present paper reviews the most commonly applied methods to identify dietary patterns, data-driven methods such as factor and cluster analysis, investigator-driven methods such as indices and score, and methods combining the two, namely reduced rank regression. We describe the techniques and their application, discuss strengths and limitations, and discuss the usefulness of dietary pattern analyses.