In conventional content analysis, codes emerge from and are defined during the analysis of the data, making it an inductive approach. In directed content analysis, researchers start with a theory or prior research findings, define codes before data analysis, and then expand those codes as analysis proceeds, combining deductive and inductive coding.

According to Hsieh and Shannon (2005), the two approaches differ mainly in when and how codes are developed. Conventional content analysis, often called inductive content analysis, begins with the data itself; the codes are not predetermined but emerge and are defined during the process of data analysis. This approach is used when the goal is to let patterns and themes arise naturally from the narratives. Directed content analysis, by contrast, starts with an existing theory or earlier relevant findings. Researchers define codes before beginning the data analysis and then expand those initial codes while analyzing the data. This combination of pre-existing, theoretically based codes and newly emerging codes means directed content analysis typically uses both deductive and inductive reasoning, and it is commonly used to validate or extend a theory or conceptual model.

Key points

Conventional content analysis is inductive: codes are developed from the data during analysis.
Directed content analysis is deductive at the outset: codes are defined before data analysis, based on theory or prior findings.
In directed analysis, codes are expanded as analysis proceeds, adding an inductive component.
Directed content analysis is often used to validate or extend an existing theory or conceptual model.
Conventional analysis is used when no prior framework drives the coding process.

How does directed content analysis differ from conventional content analysis?