A general typology of research designs featuring mixed methods.

Abstract

This article presents a general typology of research designs that features those utilizing mixed methods. The Methods-Strands Matrix includes both monomethod and mixed methods designs, but the emphasis is on the more complex and adaptable mixed ones. The article starts with a brief discussion of why typologies of mixed methods designs are valuable at this time. The Methods-Strands Matrix is produced by crossing number of methods employed (monomethod, mixed methods) by number of research strands (single, multiple). The multistrand, mixed methods cell in the matrix includes four families of MM designs: sequential, concurrent, conversion, and fully integrated. Examples of each of these strands are presented. Quasi-mixed designs are also described in which qualitative and quantitative data are collected, but there is no true integration of the findings or inferences for the overall study. We conclude that it is impossible to create a complete taxonomy of mixed methods designs, because they have an evolving nature that can spin off numerous permutations. The article concludes with a seven-step process for selecting the most appropriate mixed methods design for a research study. This article presents the Methods-Strands Matrix, which is a general typology of designs used in the social and behavioral sciences. This typology includes monomethod designs (qualitative or quantitative), but it